This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikipedia_should_have...
I should probably say that I don't believe our present "lists of articles every Wikipedia should have" are really good. I believe the lists should reflect what people from different places actually reads, or try to read, but normalized to a global perspective. That is also a necessity if the purpose is to create local communities in the different languages. Think globally, act locally!
On Sat, Feb 24, 2018 at 1:51 PM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_ every_Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_ every_Wikipedia_should_have/Expanded
Excellent idea.
I have the idea that the WMF invests $10,000 in the developing world to recruit $1000 of volunteer labor.
We need to be realistic about the relative costs of doing Western-style, rich country outreach in all economies. In the past, the strategy has been to fund the recruitment of volunteers and avoid hiring content producing staff no matter the outcomes, cost, or impact.
In the spectrum of the average income of individuals in some places, it is obviously easier and more impactful to hire someone with a masters degree to outright produce content than to pay for a program which will recruit volunteers.
Obviously, the WMF cannot and will never pay for content. However, I think that we need to make it easier for Wikimedia chapters, community groups, and partner organizations to hire paid contributors. Translation is the most obvious place to start because having base content in an encyclopedia is the foundation for demonstrating the legitimacy and value of Wikimedia projects. Funding should go from WMF to chapters to paid staff for content.
To make this project a go we would need to have a conversation about what sort of content is a priority for translation. I have a draft of an idea for prioritizing content for translation. https://meta.wikimedia.org/wiki/Wiki99 The idea is that for any given field of study, subject matter experts identify 99 articles in that field which they deem and come to consensus as priorities for having a global conversation in that field. So for example, if a group funds translation of LGBT+ content, then we would need to develop a canon of LGBT topics to which everyone in the world would have access. I have no idea how to choose topics, but fewer than 100 is probably not enough and more than 100 is probably too much for an all-languages translation project. I could use some help drafting guidelines for how to make priorities for what to translate given limited resources.
On Sat, Feb 24, 2018 at 7:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Articles about LGBT topics would be great! Similarly topics about important women in the third world. I would also like a focus on articles about primary health. Perhaps also agriculture.
I'm not sure if it is wise to move this out into chapters, keep it simple, but perhaps community groups should be able to make direct feedback.
On Sat, Feb 24, 2018 at 2:04 PM, Lane Rasberry lane@bluerasberry.com wrote:
Excellent idea.
I have the idea that the WMF invests $10,000 in the developing world to recruit $1000 of volunteer labor.
We need to be realistic about the relative costs of doing Western-style, rich country outreach in all economies. In the past, the strategy has been to fund the recruitment of volunteers and avoid hiring content producing staff no matter the outcomes, cost, or impact.
In the spectrum of the average income of individuals in some places, it is obviously easier and more impactful to hire someone with a masters degree to outright produce content than to pay for a program which will recruit volunteers.
Obviously, the WMF cannot and will never pay for content. However, I think that we need to make it easier for Wikimedia chapters, community groups, and partner organizations to hire paid contributors. Translation is the most obvious place to start because having base content in an encyclopedia is the foundation for demonstrating the legitimacy and value of Wikimedia projects. Funding should go from WMF to chapters to paid staff for content.
To make this project a go we would need to have a conversation about what sort of content is a priority for translation. I have a draft of an idea for prioritizing content for translation. https://meta.wikimedia.org/wiki/Wiki99 The idea is that for any given field of study, subject matter experts identify 99 articles in that field which they deem and come to consensus as priorities for having a global conversation in that field. So for example, if a group funds translation of LGBT+ content, then we would need to develop a canon of LGBT topics to which everyone in the world would have access. I have no idea how to choose topics, but fewer than 100 is probably not enough and more than 100 is probably too much for an all-languages translation project. I could use some help drafting guidelines for how to make priorities for what to translate given limited resources.
On Sat, Feb 24, 2018 at 7:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community
at
all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided.
Perhaps
we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a
lot
of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Lane Rasberry user:bluerasberry on Wikipedia 206.801.0814 lane@bluerasberry.com _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
We learned a few things during the medical translation project which started back in 2011:
1) You must start with high quality content and thus all articles are extensively improved before being proposed for translation.
2) A lot of languages want "less" content than is present on EN WP. Thus we moved to just improving and suggesting for translation the leads of the English articles.
3) The "Content Translation" tool developed by the WMF made efforts more efficient than handing around word documents. Would love to see that tool improved further such as having it support specific lists of articles that are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
4) We used volunteer translators mostly associated with our partner Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question. The issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
5) With respect to "paying per word" the problem is this would require significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so languages in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
6) I hired a coordinator for the translation project for a couple of years. The translators at TWB did not want to become Wikipedians or learn how to use our systems. The coordinator created account like TransSW001 (one for each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
7) Were are we at now? There are currently just over 1,000 leads of articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job of teaching high school students.
8) The project continues but at a slower pace than before. The Wikipedian and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken by 40 million people in Eastern India. The amazing thing is that for many of these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into Chinese. There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad to make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Meant to write "more than 5 million words translated". Apologies.
James
On Sat, Feb 24, 2018 at 6:26 AM, James Heilman jmh649@gmail.com wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP. Thus
we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that tool improved further such as having it support specific lists of articles that are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question. The issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so languages in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple of
years. The translators at TWB did not want to become Wikipedians or learn how to use our systems. The coordinator created account like TransSW001 (one for each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job of teaching high school students.
- The project continues but at a slower pace than before. The Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken by 40 million people in Eastern India. The amazing thing is that for many of these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into Chinese. There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad to make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian
One further case, some of the translations we did into Swahili had funding associated with them. Few people in the country have easy access to a computer and cellphones are not as suitable for translation work. Basically TWB has a brick and mortar translation center in Nairobi with computers. They have staff that keep an eye on the center. People were recruited, provided instruction, provided access to the computers, and provided cell phone credits for their involvement. What they worked on helped them develop a CV.
James
On Sat, Feb 24, 2018 at 6:30 AM, James Heilman jmh649@gmail.com wrote:
Meant to write "more than 5 million words translated". Apologies.
James
On Sat, Feb 24, 2018 at 6:26 AM, James Heilman jmh649@gmail.com wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP. Thus
we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that tool improved further such as having it support specific lists of articles that are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question. The issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so languages in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple of
years. The translators at TWB did not want to become Wikipedians or learn how to use our systems. The coordinator created account like TransSW001 (one for each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job of teaching high school students.
- The project continues but at a slower pace than before. The Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken by 40 million people in Eastern India. The amazing thing is that for many of these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into Chinese. There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad to make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian
-- James Heilman MD, CCFP-EM, Wikipedian
... make sure people are taking the work seriously and not simply using Google translate....
People are likely to start with Google Translate whether they are taking the translation seriously or not, so it would still help if we could get Google to provide numeric per-word translation confidence scores.
So please star the request for those at: https://issuetracker.google.com/issues/73830349
Thank you!
Best regards, Jim
On Sat, Feb 24, 2018 at 7:18 AM, James Heilman jmh649@gmail.com wrote:
One further case, some of the translations we did into Swahili had funding associated with them. Few people in the country have easy access to a computer and cellphones are not as suitable for translation work. Basically TWB has a brick and mortar translation center in Nairobi with computers. They have staff that keep an eye on the center. People were recruited, provided instruction, provided access to the computers, and provided cell phone credits for their involvement. What they worked on helped them develop a CV.
James
On Sat, Feb 24, 2018 at 6:30 AM, James Heilman jmh649@gmail.com wrote:
Meant to write "more than 5 million words translated". Apologies.
James
On Sat, Feb 24, 2018 at 6:26 AM, James Heilman jmh649@gmail.com wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP. Thus
we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that tool improved further such as having it support specific lists of articles that are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question. The issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so languages in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple of
years. The translators at TWB did not want to become Wikipedians or learn how to use our systems. The coordinator created account like TransSW001 (one for each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job of teaching high school students.
- The project continues but at a slower pace than before. The Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken by 40 million people in Eastern India. The amazing thing is that for many of these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into Chinese. There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad to make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian
-- James Heilman MD, CCFP-EM, Wikipedian
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the project.
3) The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that tool improved further such as having it support specific lists of articles that are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
4) We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question. The issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
5) With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so languages in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP. Thus we
moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that tool improved further such as having it support specific lists of articles that are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question. The issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so languages in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple of years.
The translators at TWB did not want to become Wikipedians or learn how to use our systems. The coordinator created account like TransSW001 (one for each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job of teaching high school students.
- The project continues but at a slower pace than before. The Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken by 40 million people in Eastern India. The amazing thing is that for many of these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into Chinese. There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad to make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community
at
all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided.
Perhaps
we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a
lot
of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Should have added that the remaining points are somewhat less interesting in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should also be pretty broad, not very narrow technical or medical, ie vertical articles, as the number of editors that can handle those will be pretty small.
In particular: Do not believe you can turn a teanslator into a new editor! You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the project.
- The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that tool improved further such as having it support specific lists of articles that are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question. The issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so languages in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP. Thus
we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that tool improved further such as having it support specific lists of articles that are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question. The issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so languages in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple of
years. The translators at TWB did not want to become Wikipedians or learn how to use our systems. The coordinator created account like TransSW001 (one for each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job of teaching high school students.
- The project continues but at a slower pace than before. The Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken by 40 million people in Eastern India. The amazing thing is that for many of these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into Chinese. There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad to make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a
community at
all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be avoided.
Perhaps
we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they need a
lot
of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my simple opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com wrote:
Should have added that the remaining points are somewhat less interesting in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should also be pretty broad, not very narrow technical or medical, ie vertical articles, as the number of editors that can handle those will be pretty small.
In particular: Do not believe you can turn a teanslator into a new editor! You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the project.
- The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP. Thus
we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple of
years. The translators at TWB did not want to become Wikipedians or learn how
to
use our systems. The coordinator created account like TransSW001 (one
for
each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job
of
teaching high school students.
- The project continues but at a slower pace than before. The
Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken
by
40 million people in Eastern India. The amazing thing is that for many
of
these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into
Chinese.
There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad
to
make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles
from
"List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word
(about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that
lacks
good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a
community at
all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be avoided.
Perhaps
we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does
not
have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they need
a
lot
of adjustments. Perhaps such a project would identify our inherit
bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
colonialism *
On Sat, Feb 24, 2018 at 9:55 AM Jean-Philippe Béland jpbeland@wikimedia.ca wrote:
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my simple opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com wrote:
Should have added that the remaining points are somewhat less interesting in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should also be pretty broad, not very narrow technical or medical, ie vertical articles, as the number of editors that can handle those will be pretty small.
In particular: Do not believe you can turn a teanslator into a new editor! You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the project.
- The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian
there
is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com
wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP.
Thus
we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian
there
is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple of
years. The translators at TWB did not want to become Wikipedians or learn how
to
use our systems. The coordinator created account like TransSW001 (one
for
each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job
of
teaching high school students.
- The project continues but at a slower pace than before. The
Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken
by
40 million people in Eastern India. The amazing thing is that for many
of
these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our
partnerships
with WMTW and medical school in Taipai continue to translate into
Chinese.
There the students translate and than their translations are reviewed
by
their profs before being posted. They translate in groups using
hackpad to
make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators?
There
are
several lists of such base sets. We have both the thousand articles
from
"List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word
(about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that
lacks
good translation tools.
I believe this would be an _enabling_ activity for the communities,
as
without a base set of articles it won't be possible to build a
community at
all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be avoided.
Perhaps
we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does
not
have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they
need a
lot
of adjustments. Perhaps such a project would identify our inherit
bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wik
i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
You can turn it around; give added credits for translations from small language projects and into the larger ones, that is a lot more interesting than strictly translating from the larger language projects.
On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland <jpbeland@wikimedia.ca
wrote:
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my simple opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com wrote:
Should have added that the remaining points are somewhat less interesting in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should also
be
pretty broad, not very narrow technical or medical, ie vertical articles, as the number of editors that can handle those will be pretty small.
In particular: Do not believe you can turn a teanslator into a new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the project.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love
the
tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian
there
is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And
for
languages in which we have little content there are often few
avaliable
volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the
work
seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com
wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP.
Thus
we moved to just improving and suggesting for translation the leads of
the
English articles.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love
the
tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian
there
is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And
for
languages in which we have little content there are often few
avaliable
volunteers.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the
work
seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple of
years. The translators at TWB did not want to become Wikipedians or learn how
to
use our systems. The coordinator created account like TransSW001 (one
for
each volunteer) and preloaded the article to be translated into
Content
Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job
of
teaching high school students.
- The project continues but at a slower pace than before. The
Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken
by
40 million people in Eastern India. The amazing thing is that for many
of
these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our
partnerships
with WMTW and medical school in Taipai continue to translate into
Chinese.
There the students translate and than their translations are reviewed
by
their profs before being posted. They translate in groups using
hackpad
to
make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators?
There
are
several lists of such base sets. We have both the thousand articles
from
"List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word
(about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that
lacks
good translation tools.
I believe this would be an _enabling_ activity for the communities,
as
without a base set of articles it won't be possible to build a
community at
all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be avoided.
Perhaps
we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does
not
have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they
need
a
lot
of adjustments. Perhaps such a project would identify our inherit
bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wik
i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I agree with John that it is very difficult to turn a translator into a new editor. I also agree with Jean-Philippe that it is key to have involvement of the local projects and preferable if they lead the efforts. Of the languages we worked in only one explicitly requested not to be involved / have translations from TWB.
James
On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad jeblad@gmail.com wrote:
You can turn it around; give added credits for translations from small language projects and into the larger ones, that is a lot more interesting than strictly translating from the larger language projects.
On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < jpbeland@wikimedia.ca
wrote:
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my simple opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com
wrote:
Should have added that the remaining points are somewhat less
interesting
in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should
also
be
pretty broad, not very narrow technical or medical, ie vertical
articles,
as the number of editors that can handle those will be pretty small.
In particular: Do not believe you can turn a teanslator into a new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the project.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of
articles
that
are deemed ready for translation by certain groups. Would also love
the
tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages
in
which their are lots of translators such as French, Spanish, and Italian
there
is often already at least some content on many of the topics in
question.
The
issue than becomes integration which needs an expert Wikipedia. And
for
languages in which we have little content there are often few
avaliable
volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would
require
significant checks and balances to make sure people are taking the
work
seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a
second
review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com
wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles
are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP.
Thus
we moved to just improving and suggesting for translation the leads of
the
English articles.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of
articles
that
are deemed ready for translation by certain groups. Would also love
the
tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages
in
which their are lots of translators such as French, Spanish, and Italian
there
is often already at least some content on many of the topics in
question.
The
issue than becomes integration which needs an expert Wikipedia. And
for
languages in which we have little content there are often few
avaliable
volunteers.
- With respect to "paying per word" the problem is this would
require
significant checks and balances to make sure people are taking the
work
seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a
second
review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple of
years. The translators at TWB did not want to become Wikipedians or learn
how
to
use our systems. The coordinator created account like TransSW001
(one
for
each volunteer) and preloaded the article to be translated into
Content
Translation. They than gave the volunteer translator the user name
and
password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO
Essential
List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into
different
Wikipedias. The coordinator has unfortunately moved on to his real
job
of
teaching high school students.
- The project continues but at a slower pace than before. The
Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically
single
handedly translated nearly all 1,000 leads into Odia a language
spoken
by
40 million people in Eastern India. The amazing thing is that for
many
of
these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our
partnerships
with WMTW and medical school in Taipai continue to translate into
Chinese.
There the students translate and than their translations are
reviewed
by
their profs before being posted. They translate in groups using
hackpad
to
make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad <jeblad@gmail.com
wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators?
There
are
several lists of such base sets. We have both the thousand
articles
from
"List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word
(about
$1
for a 1k-article) for translating one of those articles into
another
language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that
lacks
good translation tools.
I believe this would be an _enabling_ activity for the
communities,
as
without a base set of articles it won't be possible to build a
community at
all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be
avoided.
Perhaps
we should also identify good source articles, that would be a
help.
Translated articles should be above some minimum size, but they
does
not
have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they
need
a
lot
of adjustments. Perhaps such a project would identify our inherit
bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wik
i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
How about training language experts in academic institutions on how to translate contents from one language Wikipedia (Eg. English wikipedia) to another? I believe this would be more productive than paying people directly to contribute or translate contents.
Sometimes in 2016, I discussed with a professor of Yoruba language and Head of Department of Yoruba language on possible collaboration between the department and the Yoruba Wikipedia community. We agreed that students could be assigned to translating high quality articles from the English Wikipedia to Yoruba Wikipedia and they could be doing these translations as part of their course work in Yoruba language.
In Nigerian universities for example, Yoruba students take "Àyan Ògbùfò (the principle of translation) " as part of a course(s) they must pass to be awarded a degree in Yoruba language.
We could take advantage of this and approach them on possible collaboration.
Today, I had about 30 minutes discussion with one of the contributors to the Yoruba language version https://www.jw.org/yo/awon-itejade/%C3%A0w%E1%BB%8Dn-%C3%ACw%C3%A9-%C3%ACr%C3%B2y%C3%ACn/ of The watchtower and awake! magazine. https://www.jw.org/en/publications/magazines/ on possible collaboration. He was excited and agreed to be fully involved.
There are institutions and individuals that would be interested in translating high quality contents, we just need to reach out to them and devise a means to get them fully involved.
Regards,
Isaac
On Sat, Feb 24, 2018 at 4:04 PM, James Heilman jmh649@gmail.com wrote:
I agree with John that it is very difficult to turn a translator into a new editor. I also agree with Jean-Philippe that it is key to have involvement of the local projects and preferable if they lead the efforts. Of the languages we worked in only one explicitly requested not to be involved / have translations from TWB.
James
On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad jeblad@gmail.com wrote:
You can turn it around; give added credits for translations from small language projects and into the larger ones, that is a lot more
interesting
than strictly translating from the larger language projects.
On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < jpbeland@wikimedia.ca
wrote:
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my simple opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com
wrote:
Should have added that the remaining points are somewhat less
interesting
in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should
also
be
pretty broad, not very narrow technical or medical, ie vertical
articles,
as the number of editors that can handle those will be pretty small.
In particular: Do not believe you can turn a teanslator into a new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles
are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the
project.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see
that
tool
improved further such as having it support specific lists of
articles
that
are deemed ready for translation by certain groups. Would also
love
the
tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages
in
which their are lots of translators such as French, Spanish, and Italian
there
is often already at least some content on many of the topics in
question.
The
issue than becomes integration which needs an expert Wikipedia.
And
for
languages in which we have little content there are often few
avaliable
volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would
require
significant checks and balances to make sure people are taking the
work
seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a
second
review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the
project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com
wrote:
We learned a few things during the medical translation project
which
started back in 2011:
- You must start with high quality content and thus all articles
are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN
WP.
Thus
we moved to just improving and suggesting for translation the leads
of
the
English articles.
- The "Content Translation" tool developed by the WMF made
efforts
more
efficient than handing around word documents. Would love to see
that
tool
improved further such as having it support specific lists of
articles
that
are deemed ready for translation by certain groups. Would also
love
the
tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our
partner
Translators Without Borders. One issue we found was that languages
in
which their are lots of translators such as French, Spanish, and Italian
there
is often already at least some content on many of the topics in
question.
The
issue than becomes integration which needs an expert Wikipedia.
And
for
languages in which we have little content there are often few
avaliable
volunteers.
- With respect to "paying per word" the problem is this would
require
significant checks and balances to make sure people are taking the
work
seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a
second
review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple
of
years. The translators at TWB did not want to become Wikipedians or learn
how
to
use our systems. The coordinator created account like TransSW001
(one
for
each volunteer) and preloaded the article to be translated into
Content
Translation. They than gave the volunteer translator the user name
and
password to the account.
- Were are we at now? There are currently just over 1,000 leads
of
articles that have been improved and are ready for translation.
This
includes articles on the 440 medications that are on the WHO
Essential
List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into
different
Wikipedias. The coordinator has unfortunately moved on to his real
job
of
teaching high school students.
- The project continues but at a slower pace than before. The
Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically
single
handedly translated nearly all 1,000 leads into Odia a language
spoken
by
40 million people in Eastern India. The amazing thing is that for
many
of
these topics this is the first and only information online about
it.
Google translate does not even claim to work in this language. Our
partnerships
with WMTW and medical school in Taipai continue to translate into
Chinese.
There the students translate and than their translations are
reviewed
by
their profs before being posted. They translate in groups using
hackpad
to
make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad <
jeblad@gmail.com
wrote:
> This discussion is going to be fun! =D > > A little more than seventy Wikipedia-projects has more than 65k articles, > the remaining two hundred or so are pretty small. > > What if a base set of articles were opened for paid translators?
There
are > several lists of such base sets. We have both the thousand
articles
from
> "List of articles every Wikipedia should have"[1] and and the
ten
thousand > articles from the expanded list[2]. > > Lets say verified good translators was paid about $0.01 per word
(about
$1 > for a 1k-article) for translating one of those articles into
another
> language, with perhaps a higher pay for contributors in
high-cost
> countries. The pay would also have to be higher for languages
that
lacks
> good translation tools. > > I believe this would be an _enabling_ activity for the
communities,
as
> without a base set of articles it won't be possible to build a community at > all. By not paying for new articles, and only translating well-referenced > articles, some of the disputes in the communities could be
avoided.
Perhaps > we should also identify good source articles, that would be a
help.
> Translated articles should be above some minimum size, but they
does
not
> have to be full translations of the source article. > > A real problem is that our existing lists of good articles other projects > should have is pretty much biased towards Western World, so they
need
a
lot > of adjustments. Perhaps such a project would identify our
inherit
bias?
> > [1] > https://meta.wikimedia.org/wiki/List_of_articles_every_ > Wikipedia_should_have > [2] > https://meta.wikimedia.org/wiki/List_of_articles_every_ > Wikipedia_should_have/Expanded > _______________________________________________ > Wikimedia-l mailing list, guidelines at:
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ > wiki/Wikimedia-l > New messages to: Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
,
> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wik
i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
My reply can be read as a bit more harsh than intended, it was merely a statement about my present experience about translators in general.
The problem with lack of contributors (and translators) in a specialized area is that there is a small community, and within this community some kind of selection is made. Each time a selection is repeated the remaining group shrinks. Specialize the selection sufficiently many times and there will be no contributors (or translators) left. It is simply a game of probabilities. Thus, to make such a project work it must have a sufficiently broad scope for the articles. Articles about public health services will probably work even for a pretty small language group, but specialized medical articles might create a problem. But then you find a retired orthopedic surgeon like Subas Chandra Rout…
On Sat, Feb 24, 2018 at 4:04 PM, James Heilman jmh649@gmail.com wrote:
I agree with John that it is very difficult to turn a translator into a new editor. I also agree with Jean-Philippe that it is key to have involvement of the local projects and preferable if they lead the efforts. Of the languages we worked in only one explicitly requested not to be involved / have translations from TWB.
James
On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad jeblad@gmail.com wrote:
You can turn it around; give added credits for translations from small language projects and into the larger ones, that is a lot more
interesting
than strictly translating from the larger language projects.
On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < jpbeland@wikimedia.ca
wrote:
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my simple opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com
wrote:
Should have added that the remaining points are somewhat less
interesting
in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should
also
be
pretty broad, not very narrow technical or medical, ie vertical
articles,
as the number of editors that can handle those will be pretty small.
In particular: Do not believe you can turn a teanslator into a new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles
are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the
project.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see
that
tool
improved further such as having it support specific lists of
articles
that
are deemed ready for translation by certain groups. Would also
love
the
tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages
in
which their are lots of translators such as French, Spanish, and Italian
there
is often already at least some content on many of the topics in
question.
The
issue than becomes integration which needs an expert Wikipedia.
And
for
languages in which we have little content there are often few
avaliable
volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would
require
significant checks and balances to make sure people are taking the
work
seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a
second
review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the
project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com
wrote:
We learned a few things during the medical translation project
which
started back in 2011:
- You must start with high quality content and thus all articles
are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN
WP.
Thus
we moved to just improving and suggesting for translation the leads
of
the
English articles.
- The "Content Translation" tool developed by the WMF made
efforts
more
efficient than handing around word documents. Would love to see
that
tool
improved further such as having it support specific lists of
articles
that
are deemed ready for translation by certain groups. Would also
love
the
tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our
partner
Translators Without Borders. One issue we found was that languages
in
which their are lots of translators such as French, Spanish, and Italian
there
is often already at least some content on many of the topics in
question.
The
issue than becomes integration which needs an expert Wikipedia.
And
for
languages in which we have little content there are often few
avaliable
volunteers.
- With respect to "paying per word" the problem is this would
require
significant checks and balances to make sure people are taking the
work
seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a
second
review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple
of
years. The translators at TWB did not want to become Wikipedians or learn
how
to
use our systems. The coordinator created account like TransSW001
(one
for
each volunteer) and preloaded the article to be translated into
Content
Translation. They than gave the volunteer translator the user name
and
password to the account.
- Were are we at now? There are currently just over 1,000 leads
of
articles that have been improved and are ready for translation.
This
includes articles on the 440 medications that are on the WHO
Essential
List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into
different
Wikipedias. The coordinator has unfortunately moved on to his real
job
of
teaching high school students.
- The project continues but at a slower pace than before. The
Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically
single
handedly translated nearly all 1,000 leads into Odia a language
spoken
by
40 million people in Eastern India. The amazing thing is that for
many
of
these topics this is the first and only information online about
it.
Google translate does not even claim to work in this language. Our
partnerships
with WMTW and medical school in Taipai continue to translate into
Chinese.
There the students translate and than their translations are
reviewed
by
their profs before being posted. They translate in groups using
hackpad
to
make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad <
jeblad@gmail.com
wrote:
> This discussion is going to be fun! =D > > A little more than seventy Wikipedia-projects has more than 65k articles, > the remaining two hundred or so are pretty small. > > What if a base set of articles were opened for paid translators?
There
are > several lists of such base sets. We have both the thousand
articles
from
> "List of articles every Wikipedia should have"[1] and and the
ten
thousand > articles from the expanded list[2]. > > Lets say verified good translators was paid about $0.01 per word
(about
$1 > for a 1k-article) for translating one of those articles into
another
> language, with perhaps a higher pay for contributors in
high-cost
> countries. The pay would also have to be higher for languages
that
lacks
> good translation tools. > > I believe this would be an _enabling_ activity for the
communities,
as
> without a base set of articles it won't be possible to build a community at > all. By not paying for new articles, and only translating well-referenced > articles, some of the disputes in the communities could be
avoided.
Perhaps > we should also identify good source articles, that would be a
help.
> Translated articles should be above some minimum size, but they
does
not
> have to be full translations of the source article. > > A real problem is that our existing lists of good articles other projects > should have is pretty much biased towards Western World, so they
need
a
lot > of adjustments. Perhaps such a project would identify our
inherit
bias?
> > [1] > https://meta.wikimedia.org/wiki/List_of_articles_every_ > Wikipedia_should_have > [2] > https://meta.wikimedia.org/wiki/List_of_articles_every_ > Wikipedia_should_have/Expanded > _______________________________________________ > Wikimedia-l mailing list, guidelines at:
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ > wiki/Wikimedia-l > New messages to: Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
,
> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wik
i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I'll reply to the most recent email just for laziness.
I'm doubtful for a series of reasons, most of were already expressed in a better way by others: *a remuneration in terms of quantity will weaken the quality of translations unless there's a strong mechanism of quality verification requiring a quantity of resources comparable to translations themselves; *articles are the result of a long process which reflects cultural identity of different communities, I'm not confident with transferring them to a different "weaker" cultures. My usage of "weaker" adjective only focuses about the strength of a cultural presence on the Internet; *articles to be translated are at high risk of reflecting the cultural identity (and biases) of the Western culture; *finally I think paid translators would hardly turn into stable Wikipedians.
IMHO some paid editing may be better exploited in order to digitalise texts of unrepresented cultures (wikisource) or preserving their vocabularies (wiktionary).
Also those languages which are secondary for all their speakers should be dealt with in a different fashion. I, for one, am a native speaker of specific variant of Sicilian, Sicilian is a secondary language to any of its speakers. Honestly, I'd find pointless to read the biography of Leonardo da Vinci in Sicilian while I can find thousands of books about him in Italian. Also I find this kind of translation creates a fake "literary" language totally detached from reality: there's no "encaustic painting" in Sicilian, still a Sicilian article about Leonardo will invent one.
As a general principle we should always collect, rather than create, knowledge.
Vito
2018-02-24 16:30 GMT+01:00 John Erling Blad jeblad@gmail.com:
My reply can be read as a bit more harsh than intended, it was merely a statement about my present experience about translators in general.
The problem with lack of contributors (and translators) in a specialized area is that there is a small community, and within this community some kind of selection is made. Each time a selection is repeated the remaining group shrinks. Specialize the selection sufficiently many times and there will be no contributors (or translators) left. It is simply a game of probabilities. Thus, to make such a project work it must have a sufficiently broad scope for the articles. Articles about public health services will probably work even for a pretty small language group, but specialized medical articles might create a problem. But then you find a retired orthopedic surgeon like Subas Chandra Rout…
On Sat, Feb 24, 2018 at 4:04 PM, James Heilman jmh649@gmail.com wrote:
I agree with John that it is very difficult to turn a translator into a
new
editor. I also agree with Jean-Philippe that it is key to have
involvement
of the local projects and preferable if they lead the efforts. Of the languages we worked in only one explicitly requested not to be involved / have translations from TWB.
James
On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad jeblad@gmail.com wrote:
You can turn it around; give added credits for translations from small language projects and into the larger ones, that is a lot more
interesting
than strictly translating from the larger language projects.
On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < jpbeland@wikimedia.ca
wrote:
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my
simple
opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com
wrote:
Should have added that the remaining points are somewhat less
interesting
in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should
also
be
pretty broad, not very narrow technical or medical, ie vertical
articles,
as the number of editors that can handle those will be pretty
small.
In particular: Do not believe you can turn a teanslator into a new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
jeblad@gmail.com>
wrote:
- You must start with high quality content and thus all articles
are
> extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the
project.
- The "Content Translation" tool developed by the WMF made
efforts
more
> efficient than handing around word documents. Would love to see
that
tool
> improved further such as having it support specific lists of
articles
that
> are deemed ready for translation by certain groups. Would also
love
the
> tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty
obvious.
- We used volunteer translators mostly associated with our
partner
> Translators Without Borders. One issue we found was that
languages
in
> which > their are lots of translators such as French, Spanish, and
Italian
there
> is > often already at least some content on many of the topics in
question.
The
> issue than becomes integration which needs an expert Wikipedia.
And
for
> languages in which we have little content there are often few
avaliable
> volunteers.
I used projects below 65k articles as an example, as the chance
of
competing articles are pretty low.
- With respect to "paying per word" the problem is this would
require
> significant checks and balances to make sure people are taking
the
work
> seriously and not simple using Google translate for the 70 or so
languages
> in which it claims to work. We often had translations undergo a
second
> review and the volunteers at TWB have to pass certain tests to
be
> accepted.
I'n my original email I wrote "verified good translators". It is
as
simple as "Has the editor contributed other articles at the
project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <jmh649@gmail.com
wrote:
> We learned a few things during the medical translation project
which
> started back in 2011: > > 1) You must start with high quality content and thus all
articles
are
> extensively improved before being proposed for translation. > > 2) A lot of languages want "less" content than is present on EN
WP.
Thus
> we > moved to just improving and suggesting for translation the leads
of
the
> English articles. > > 3) The "Content Translation" tool developed by the WMF made
efforts
more
> efficient than handing around word documents. Would love to see
that
tool
> improved further such as having it support specific lists of
articles
that
> are deemed ready for translation by certain groups. Would also
love
the
> tool to have tracking metrics for these types of projects. > > 4) We used volunteer translators mostly associated with our
partner
> Translators Without Borders. One issue we found was that
languages
in
> which > their are lots of translators such as French, Spanish, and
Italian
there
> is > often already at least some content on many of the topics in
question.
The
> issue than becomes integration which needs an expert Wikipedia.
And
for
> languages in which we have little content there are often few
avaliable
> volunteers. > > 5) With respect to "paying per word" the problem is this would
require
> significant checks and balances to make sure people are taking
the
work
> seriously and not simple using Google translate for the 70 or so
languages
> in which it claims to work. We often had translations undergo a
second
> review and the volunteers at TWB have to pass certain tests to
be
> accepted. > > 6) I hired a coordinator for the translation project for a
couple
of
> years. > The translators at TWB did not want to become Wikipedians or
learn
how
to
> use our systems. The coordinator created account like TransSW001
(one
for
> each volunteer) and preloaded the article to be translated into
Content
> Translation. They than gave the volunteer translator the user
name
and
> password to the account. > > 7) Were are we at now? There are currently just over 1,000 leads
of
> articles that have been improved and are ready for translation.
This
> includes articles on the 440 medications that are on the WHO
Essential
> List. We have worked a bit in some 100 languages. The efforts
have
> resulted > in more than 5 million works translated and integrated into
different
> Wikipedias. The coordinator has unfortunately moved on to his
real
job
of
> teaching high school students. > > 8) The project continues but at a slower pace than before. The
Wikipedian
> and retired orthopedic surgeon Subas Chandra Rout has basically
single
> handedly translated nearly all 1,000 leads into Odia a language
spoken
by
> 40 million people in Eastern India. The amazing thing is that
for
many
of
> these topics this is the first and only information online about
it.
> Google > translate does not even claim to work in this language. Our
partnerships
> with WMTW and medical school in Taipai continue to translate
into
Chinese.
> There the students translate and than their translations are
reviewed
by
> their profs before being posted. They translate in groups using
hackpad
to
> make it more social. > > I am currently working to re invigorate the project :-) > James > > On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad <
jeblad@gmail.com
> wrote: > > > This discussion is going to be fun! =D > > > > A little more than seventy Wikipedia-projects has more than
65k
> articles, > > the remaining two hundred or so are pretty small. > > > > What if a base set of articles were opened for paid
translators?
There
> are > > several lists of such base sets. We have both the thousand
articles
from
> > "List of articles every Wikipedia should have"[1] and and the
ten
> thousand > > articles from the expanded list[2]. > > > > Lets say verified good translators was paid about $0.01 per
word
(about
> $1 > > for a 1k-article) for translating one of those articles into
another
> > language, with perhaps a higher pay for contributors in
high-cost
> > countries. The pay would also have to be higher for languages
that
lacks
> > good translation tools. > > > > I believe this would be an _enabling_ activity for the
communities,
as
> > without a base set of articles it won't be possible to build a > community at > > all. By not paying for new articles, and only translating > well-referenced > > articles, some of the disputes in the communities could be
avoided.
> Perhaps > > we should also identify good source articles, that would be a
help.
> > Translated articles should be above some minimum size, but
they
does
not
> > have to be full translations of the source article. > > > > A real problem is that our existing lists of good articles
other
> projects > > should have is pretty much biased towards Western World, so
they
need
a
> lot > > of adjustments. Perhaps such a project would identify our
inherit
bias?
> > > > [1] > > https://meta.wikimedia.org/wiki/List_of_articles_every_ > > Wikipedia_should_have > > [2] > > https://meta.wikimedia.org/wiki/List_of_articles_every_ > > Wikipedia_should_have/Expanded > > _______________________________________________ > > Wikimedia-l mailing list, guidelines at:
> > wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ > > wiki/Wikimedia-l > > New messages to: Wikimedia-l@lists.wikimedia.org > > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
,
> > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> > > > > -- > James Heilman > MD, CCFP-EM, Wikipedian > _______________________________________________ > Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wik
> i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik > i/Wikimedia-l > New messages to: Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
,
> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
this would be a good practical exercise to develop for WiR / WikiEd programs in universities where they can engage with International Students and local students studying additional languages as means of learning the written nuances of the individual languages. Any funding would be better utilised in enabling such programs where the flow on impact is more likely{fact} to be lasting. Though I can see value in using a gift/reward system for technically disadvantaged communities like the case presented about Swahili . The focus would need to be on basic health, hygiene, biology, science topics rather than more social or political topics.
On 25 February 2018 at 01:08, Vi to vituzzu.wiki@gmail.com wrote:
I'll reply to the most recent email just for laziness.
I'm doubtful for a series of reasons, most of were already expressed in a better way by others: *a remuneration in terms of quantity will weaken the quality of translations unless there's a strong mechanism of quality verification requiring a quantity of resources comparable to translations themselves; *articles are the result of a long process which reflects cultural identity of different communities, I'm not confident with transferring them to a different "weaker" cultures. My usage of "weaker" adjective only focuses about the strength of a cultural presence on the Internet; *articles to be translated are at high risk of reflecting the cultural identity (and biases) of the Western culture; *finally I think paid translators would hardly turn into stable Wikipedians.
IMHO some paid editing may be better exploited in order to digitalise texts of unrepresented cultures (wikisource) or preserving their vocabularies (wiktionary).
Also those languages which are secondary for all their speakers should be dealt with in a different fashion. I, for one, am a native speaker of specific variant of Sicilian, Sicilian is a secondary language to any of its speakers. Honestly, I'd find pointless to read the biography of Leonardo da Vinci in Sicilian while I can find thousands of books about him in Italian. Also I find this kind of translation creates a fake "literary" language totally detached from reality: there's no "encaustic painting" in Sicilian, still a Sicilian article about Leonardo will invent one.
As a general principle we should always collect, rather than create, knowledge.
Vito
2018-02-24 16:30 GMT+01:00 John Erling Blad jeblad@gmail.com:
My reply can be read as a bit more harsh than intended, it was merely a statement about my present experience about translators in general.
The problem with lack of contributors (and translators) in a specialized area is that there is a small community, and within this community some kind of selection is made. Each time a selection is repeated the
remaining
group shrinks. Specialize the selection sufficiently many times and there will be no contributors (or translators) left. It is simply a game of probabilities. Thus, to make such a project work it must have a sufficiently broad scope for the articles. Articles about public health services will probably work even for a pretty small language group, but specialized medical articles might create a problem. But then you find a retired orthopedic surgeon like Subas Chandra Rout…
On Sat, Feb 24, 2018 at 4:04 PM, James Heilman jmh649@gmail.com wrote:
I agree with John that it is very difficult to turn a translator into a
new
editor. I also agree with Jean-Philippe that it is key to have
involvement
of the local projects and preferable if they lead the efforts. Of the languages we worked in only one explicitly requested not to be
involved /
have translations from TWB.
James
On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad jeblad@gmail.com wrote:
You can turn it around; give added credits for translations from
small
language projects and into the larger ones, that is a lot more
interesting
than strictly translating from the larger language projects.
On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < jpbeland@wikimedia.ca
wrote:
I think the request for such projects should come from the
concerned
language projects, same for the list of articles. If not, in my
simple
opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <jeblad@gmail.com
wrote:
Should have added that the remaining points are somewhat less
interesting
in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles
should
also
be
pretty broad, not very narrow technical or medical, ie vertical
articles,
as the number of editors that can handle those will be pretty
small.
In particular: Do not believe you can turn a teanslator into a
new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
jeblad@gmail.com>
wrote:
> 1) You must start with high quality content and thus all
articles
are
>> extensively improved before being proposed for translation. > > > Note that to much pressure on "quality" can easily kill the
project.
> > 3) The "Content Translation" tool developed by the WMF made
efforts
more
>> efficient than handing around word documents. Would love to
see
that
tool >> improved further such as having it support specific lists of
articles
that >> are deemed ready for translation by certain groups. Would also
love
the
>> tool to have tracking metrics for these types of projects. > > > Didn't mention ContentTranslation, but it should be pretty
obvious.
> > 4) We used volunteer translators mostly associated with our
partner
>> Translators Without Borders. One issue we found was that
languages
in
>> which >> their are lots of translators such as French, Spanish, and
Italian
there
>> is >> often already at least some content on many of the topics in
question.
The >> issue than becomes integration which needs an expert
Wikipedia.
And
for
>> languages in which we have little content there are often few
avaliable
>> volunteers. > > > I used projects below 65k articles as an example, as the chance
of
> competing articles are pretty low. > > 5) With respect to "paying per word" the problem is this would
require
>> significant checks and balances to make sure people are taking
the
work
>> seriously and not simple using Google translate for the 70 or
so
languages >> in which it claims to work. We often had translations undergo
a
second
>> review and the volunteers at TWB have to pass certain tests to
be
>> accepted. > > > I'n my original email I wrote "verified good translators". It
is
as
> simple as "Has the editor contributed other articles at the
project?"
> > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <
jmh649@gmail.com
wrote:
> >> We learned a few things during the medical translation project
which
>> started back in 2011: >> >> 1) You must start with high quality content and thus all
articles
are
>> extensively improved before being proposed for translation. >> >> 2) A lot of languages want "less" content than is present on
EN
WP.
Thus
>> we >> moved to just improving and suggesting for translation the
leads
of
the
>> English articles. >> >> 3) The "Content Translation" tool developed by the WMF made
efforts
more
>> efficient than handing around word documents. Would love to
see
that
tool >> improved further such as having it support specific lists of
articles
that >> are deemed ready for translation by certain groups. Would also
love
the
>> tool to have tracking metrics for these types of projects. >> >> 4) We used volunteer translators mostly associated with our
partner
>> Translators Without Borders. One issue we found was that
languages
in
>> which >> their are lots of translators such as French, Spanish, and
Italian
there
>> is >> often already at least some content on many of the topics in
question.
The >> issue than becomes integration which needs an expert
Wikipedia.
And
for
>> languages in which we have little content there are often few
avaliable
>> volunteers. >> >> 5) With respect to "paying per word" the problem is this would
require
>> significant checks and balances to make sure people are taking
the
work
>> seriously and not simple using Google translate for the 70 or
so
languages >> in which it claims to work. We often had translations undergo
a
second
>> review and the volunteers at TWB have to pass certain tests to
be
>> accepted. >> >> 6) I hired a coordinator for the translation project for a
couple
of
>> years. >> The translators at TWB did not want to become Wikipedians or
learn
how
to >> use our systems. The coordinator created account like
TransSW001
(one
for >> each volunteer) and preloaded the article to be translated
into
Content
>> Translation. They than gave the volunteer translator the user
name
and
>> password to the account. >> >> 7) Were are we at now? There are currently just over 1,000
leads
of
>> articles that have been improved and are ready for
translation.
This
>> includes articles on the 440 medications that are on the WHO
Essential
>> List. We have worked a bit in some 100 languages. The efforts
have
>> resulted >> in more than 5 million works translated and integrated into
different
>> Wikipedias. The coordinator has unfortunately moved on to his
real
job
of >> teaching high school students. >> >> 8) The project continues but at a slower pace than before. The Wikipedian >> and retired orthopedic surgeon Subas Chandra Rout has
basically
single
>> handedly translated nearly all 1,000 leads into Odia a
language
spoken
by >> 40 million people in Eastern India. The amazing thing is that
for
many
of >> these topics this is the first and only information online
about
it.
>> Google >> translate does not even claim to work in this language. Our
partnerships
>> with WMTW and medical school in Taipai continue to translate
into
Chinese. >> There the students translate and than their translations are
reviewed
by
>> their profs before being posted. They translate in groups
using
hackpad
to >> make it more social. >> >> I am currently working to re invigorate the project :-) >> James >> >> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad <
jeblad@gmail.com
>> wrote: >> >> > This discussion is going to be fun! =D >> > >> > A little more than seventy Wikipedia-projects has more than
65k
>> articles, >> > the remaining two hundred or so are pretty small. >> > >> > What if a base set of articles were opened for paid
translators?
There
>> are >> > several lists of such base sets. We have both the thousand
articles
from >> > "List of articles every Wikipedia should have"[1] and and
the
ten
>> thousand >> > articles from the expanded list[2]. >> > >> > Lets say verified good translators was paid about $0.01 per
word
(about >> $1 >> > for a 1k-article) for translating one of those articles into
another
>> > language, with perhaps a higher pay for contributors in
high-cost
>> > countries. The pay would also have to be higher for
languages
that
lacks >> > good translation tools. >> > >> > I believe this would be an _enabling_ activity for the
communities,
as
>> > without a base set of articles it won't be possible to
build a
>> community at >> > all. By not paying for new articles, and only translating >> well-referenced >> > articles, some of the disputes in the communities could be
avoided.
>> Perhaps >> > we should also identify good source articles, that would be
a
help.
>> > Translated articles should be above some minimum size, but
they
does
not >> > have to be full translations of the source article. >> > >> > A real problem is that our existing lists of good articles
other
>> projects >> > should have is pretty much biased towards Western World, so
they
need
a >> lot >> > of adjustments. Perhaps such a project would identify our
inherit
bias? >> > >> > [1] >> > https://meta.wikimedia.org/wiki/List_of_articles_every_ >> > Wikipedia_should_have >> > [2] >> > https://meta.wikimedia.org/wiki/List_of_articles_every_ >> > Wikipedia_should_have/Expanded >> > _______________________________________________ >> > Wikimedia-l mailing list, guidelines at:
>> > wiki/Mailing_lists/Guidelines and
>> > wiki/Wikimedia-l >> > New messages to: Wikimedia-l@lists.wikimedia.org >> > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
, >> > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
>> >> >> >> >> -- >> James Heilman >> MD, CCFP-EM, Wikipedian >> _______________________________________________ >> Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wik
>> i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik >> i/Wikimedia-l >> New messages to: Wikimedia-l@lists.wikimedia.org >> Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
,
>> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
>> > > _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
This is not the same, and is more like the present grant system.
On Sat, Feb 24, 2018 at 8:05 PM, Gnangarra gnangarra@gmail.com wrote:
this would be a good practical exercise to develop for WiR / WikiEd programs in universities where they can engage with International Students and local students studying additional languages as means of learning the written nuances of the individual languages. Any funding would be better utilised in enabling such programs where the flow on impact is more likely{fact} to be lasting. Though I can see value in using a gift/reward system for technically disadvantaged communities like the case presented about Swahili . The focus would need to be on basic health, hygiene, biology, science topics rather than more social or political topics.
On 25 February 2018 at 01:08, Vi to vituzzu.wiki@gmail.com wrote:
I'll reply to the most recent email just for laziness.
I'm doubtful for a series of reasons, most of were already expressed in a better way by others: *a remuneration in terms of quantity will weaken the quality of translations unless there's a strong mechanism of quality verification requiring a quantity of resources comparable to translations themselves; *articles are the result of a long process which reflects cultural
identity
of different communities, I'm not confident with transferring them to a different "weaker" cultures. My usage of "weaker" adjective only focuses about the strength of a cultural presence on the Internet; *articles to be translated are at high risk of reflecting the cultural identity (and biases) of the Western culture; *finally I think paid translators would hardly turn into stable Wikipedians.
IMHO some paid editing may be better exploited in order to digitalise
texts
of unrepresented cultures (wikisource) or preserving their vocabularies (wiktionary).
Also those languages which are secondary for all their speakers should be dealt with in a different fashion. I, for one, am a native speaker of specific variant of Sicilian, Sicilian is a secondary language to any of its speakers. Honestly, I'd find pointless to read the biography of Leonardo da Vinci in Sicilian while I can find thousands of books about
him
in Italian. Also I find this kind of translation creates a fake
"literary"
language totally detached from reality: there's no "encaustic painting"
in
Sicilian, still a Sicilian article about Leonardo will invent one.
As a general principle we should always collect, rather than create, knowledge.
Vito
2018-02-24 16:30 GMT+01:00 John Erling Blad jeblad@gmail.com:
My reply can be read as a bit more harsh than intended, it was merely a statement about my present experience about translators in general.
The problem with lack of contributors (and translators) in a
specialized
area is that there is a small community, and within this community some kind of selection is made. Each time a selection is repeated the
remaining
group shrinks. Specialize the selection sufficiently many times and
there
will be no contributors (or translators) left. It is simply a game of probabilities. Thus, to make such a project work it must have a sufficiently broad scope for the articles. Articles about public health services will probably work even for a pretty small language group, but specialized medical articles might create a problem. But then you find a retired orthopedic surgeon like Subas Chandra Rout…
On Sat, Feb 24, 2018 at 4:04 PM, James Heilman jmh649@gmail.com
wrote:
I agree with John that it is very difficult to turn a translator
into a
new
editor. I also agree with Jean-Philippe that it is key to have
involvement
of the local projects and preferable if they lead the efforts. Of the languages we worked in only one explicitly requested not to be
involved /
have translations from TWB.
James
On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad jeblad@gmail.com wrote:
You can turn it around; give added credits for translations from
small
language projects and into the larger ones, that is a lot more
interesting
than strictly translating from the larger language projects.
On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < jpbeland@wikimedia.ca
wrote:
I think the request for such projects should come from the
concerned
language projects, same for the list of articles. If not, in my
simple
opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <
jeblad@gmail.com
wrote:
> Should have added that the remaining points are somewhat less
interesting
> in this context. Preloading a set of articles is a bad idea,
the
> translators should be able to chose for themselves. Articles
should
also
be > pretty broad, not very narrow technical or medical, ie vertical
articles,
> as the number of editors that can handle those will be pretty
small.
> > In particular: Do not believe you can turn a teanslator into a
new
editor! > You can although turn an existing editor into a translator. > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
jeblad@gmail.com>
> wrote: > > > 1) You must start with high quality content and thus all
articles
are
> >> extensively improved before being proposed for translation. > > > > > > Note that to much pressure on "quality" can easily kill the
project.
> > > > 3) The "Content Translation" tool developed by the WMF made
efforts
more > >> efficient than handing around word documents. Would love to
see
that
> tool > >> improved further such as having it support specific lists of
articles
> that > >> are deemed ready for translation by certain groups. Would
also
love
the > >> tool to have tracking metrics for these types of projects. > > > > > > Didn't mention ContentTranslation, but it should be pretty
obvious.
> > > > 4) We used volunteer translators mostly associated with our
partner
> >> Translators Without Borders. One issue we found was that
languages
in
> >> which > >> their are lots of translators such as French, Spanish, and
Italian
there > >> is > >> often already at least some content on many of the topics in
question.
> The > >> issue than becomes integration which needs an expert
Wikipedia.
And
for > >> languages in which we have little content there are often
few
avaliable > >> volunteers. > > > > > > I used projects below 65k articles as an example, as the
chance
of
> > competing articles are pretty low. > > > > 5) With respect to "paying per word" the problem is this
would
require
> >> significant checks and balances to make sure people are
taking
the
work > >> seriously and not simple using Google translate for the 70
or
so
> languages > >> in which it claims to work. We often had translations
undergo
a
second
> >> review and the volunteers at TWB have to pass certain tests
to
be
> >> accepted. > > > > > > I'n my original email I wrote "verified good translators". It
is
as
> > simple as "Has the editor contributed other articles at the
project?"
> > > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <
jmh649@gmail.com
wrote: > > > >> We learned a few things during the medical translation
project
which
> >> started back in 2011: > >> > >> 1) You must start with high quality content and thus all
articles
are
> >> extensively improved before being proposed for translation. > >> > >> 2) A lot of languages want "less" content than is present on
EN
WP.
Thus > >> we > >> moved to just improving and suggesting for translation the
leads
of
the > >> English articles. > >> > >> 3) The "Content Translation" tool developed by the WMF made
efforts
more > >> efficient than handing around word documents. Would love to
see
that
> tool > >> improved further such as having it support specific lists of
articles
> that > >> are deemed ready for translation by certain groups. Would
also
love
the > >> tool to have tracking metrics for these types of projects. > >> > >> 4) We used volunteer translators mostly associated with our
partner
> >> Translators Without Borders. One issue we found was that
languages
in
> >> which > >> their are lots of translators such as French, Spanish, and
Italian
there > >> is > >> often already at least some content on many of the topics in
question.
> The > >> issue than becomes integration which needs an expert
Wikipedia.
And
for > >> languages in which we have little content there are often
few
avaliable > >> volunteers. > >> > >> 5) With respect to "paying per word" the problem is this
would
require
> >> significant checks and balances to make sure people are
taking
the
work > >> seriously and not simple using Google translate for the 70
or
so
> languages > >> in which it claims to work. We often had translations
undergo
a
second
> >> review and the volunteers at TWB have to pass certain tests
to
be
> >> accepted. > >> > >> 6) I hired a coordinator for the translation project for a
couple
of
> >> years. > >> The translators at TWB did not want to become Wikipedians or
learn
how
> to > >> use our systems. The coordinator created account like
TransSW001
(one
> for > >> each volunteer) and preloaded the article to be translated
into
Content > >> Translation. They than gave the volunteer translator the
user
name
and
> >> password to the account. > >> > >> 7) Were are we at now? There are currently just over 1,000
leads
of
> >> articles that have been improved and are ready for
translation.
This
> >> includes articles on the 440 medications that are on the WHO
Essential
> >> List. We have worked a bit in some 100 languages. The
efforts
have
> >> resulted > >> in more than 5 million works translated and integrated into
different
> >> Wikipedias. The coordinator has unfortunately moved on to
his
real
job
> of > >> teaching high school students. > >> > >> 8) The project continues but at a slower pace than before.
The
> Wikipedian > >> and retired orthopedic surgeon Subas Chandra Rout has
basically
single
> >> handedly translated nearly all 1,000 leads into Odia a
language
spoken
> by > >> 40 million people in Eastern India. The amazing thing is
that
for
many
> of > >> these topics this is the first and only information online
about
it.
> >> Google > >> translate does not even claim to work in this language. Our partnerships > >> with WMTW and medical school in Taipai continue to translate
into
> Chinese. > >> There the students translate and than their translations are
reviewed
by > >> their profs before being posted. They translate in groups
using
hackpad > to > >> make it more social. > >> > >> I am currently working to re invigorate the project :-) > >> James > >> > >> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad <
jeblad@gmail.com
> >> wrote: > >> > >> > This discussion is going to be fun! =D > >> > > >> > A little more than seventy Wikipedia-projects has more
than
65k
> >> articles, > >> > the remaining two hundred or so are pretty small. > >> > > >> > What if a base set of articles were opened for paid
translators?
There > >> are > >> > several lists of such base sets. We have both the thousand
articles
> from > >> > "List of articles every Wikipedia should have"[1] and and
the
ten
> >> thousand > >> > articles from the expanded list[2]. > >> > > >> > Lets say verified good translators was paid about $0.01
per
word
> (about > >> $1 > >> > for a 1k-article) for translating one of those articles
into
another
> >> > language, with perhaps a higher pay for contributors in
high-cost
> >> > countries. The pay would also have to be higher for
languages
that
> lacks > >> > good translation tools. > >> > > >> > I believe this would be an _enabling_ activity for the
communities,
as > >> > without a base set of articles it won't be possible to
build a
> >> community at > >> > all. By not paying for new articles, and only translating > >> well-referenced > >> > articles, some of the disputes in the communities could be
avoided.
> >> Perhaps > >> > we should also identify good source articles, that would
be
a
help.
> >> > Translated articles should be above some minimum size, but
they
does
> not > >> > have to be full translations of the source article. > >> > > >> > A real problem is that our existing lists of good articles
other
> >> projects > >> > should have is pretty much biased towards Western World,
so
they
need > a > >> lot > >> > of adjustments. Perhaps such a project would identify our
inherit
> bias? > >> > > >> > [1] > >> > https://meta.wikimedia.org/wiki/List_of_articles_every_ > >> > Wikipedia_should_have > >> > [2] > >> > https://meta.wikimedia.org/wiki/List_of_articles_every_ > >> > Wikipedia_should_have/Expanded > >> > _______________________________________________ > >> > Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ > >> > wiki/Mailing_lists/Guidelines and
> >> > wiki/Wikimedia-l > >> > New messages to: Wikimedia-l@lists.wikimedia.org > >> > Unsubscribe: https://lists.wikimedia.org/ mailman/listinfo/wikimedia-l > , > >> > mailto:wikimedia-l-request@lists.wikimedia.org?subject= unsubscribe > >> > >> > >> > >> > >> -- > >> James Heilman > >> MD, CCFP-EM, Wikipedian > >> _______________________________________________ > >> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik > >> i/Mailing_lists/Guidelines and
https://meta.wikimedia.org/wik
> >> i/Wikimedia-l > >> New messages to: Wikimedia-l@lists.wikimedia.org > >> Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
, > >> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> >> > > > > > _______________________________________________ > Wikimedia-l mailing list, guidelines at: > https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and > https://meta.wikimedia.org/wiki/Wikimedia-l > New messages to: Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- GN. Noongarpedia: https://incubator.wikimedia.org/wiki/Wp/nys/Main_Page WMAU: http://www.wikimedia.org.au/wiki/User:Gnangarra Photo Gallery: http://gnangarra.redbubble.com Out now: A.Gaynor, P. Newman and P. Jennings (eds.), *Never Again: Reflections on Environmental Responsibility after Roe 8*, UWAP, 2017. Order here https://uwap.uwa.edu.au/products/never-again- reflections-on-environmental-responsibility-after-roe-8 . _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Seems like this is mostly about cultural ownership and appropriation. Not sure if it is possible to agree on this.
On Sat, Feb 24, 2018 at 6:08 PM, Vi to vituzzu.wiki@gmail.com wrote:
I'll reply to the most recent email just for laziness.
I'm doubtful for a series of reasons, most of were already expressed in a better way by others: *a remuneration in terms of quantity will weaken the quality of translations unless there's a strong mechanism of quality verification requiring a quantity of resources comparable to translations themselves; *articles are the result of a long process which reflects cultural identity of different communities, I'm not confident with transferring them to a different "weaker" cultures. My usage of "weaker" adjective only focuses about the strength of a cultural presence on the Internet; *articles to be translated are at high risk of reflecting the cultural identity (and biases) of the Western culture; *finally I think paid translators would hardly turn into stable Wikipedians.
IMHO some paid editing may be better exploited in order to digitalise texts of unrepresented cultures (wikisource) or preserving their vocabularies (wiktionary).
Also those languages which are secondary for all their speakers should be dealt with in a different fashion. I, for one, am a native speaker of specific variant of Sicilian, Sicilian is a secondary language to any of its speakers. Honestly, I'd find pointless to read the biography of Leonardo da Vinci in Sicilian while I can find thousands of books about him in Italian. Also I find this kind of translation creates a fake "literary" language totally detached from reality: there's no "encaustic painting" in Sicilian, still a Sicilian article about Leonardo will invent one.
As a general principle we should always collect, rather than create, knowledge.
Vito
2018-02-24 16:30 GMT+01:00 John Erling Blad jeblad@gmail.com:
My reply can be read as a bit more harsh than intended, it was merely a statement about my present experience about translators in general.
The problem with lack of contributors (and translators) in a specialized area is that there is a small community, and within this community some kind of selection is made. Each time a selection is repeated the
remaining
group shrinks. Specialize the selection sufficiently many times and there will be no contributors (or translators) left. It is simply a game of probabilities. Thus, to make such a project work it must have a sufficiently broad scope for the articles. Articles about public health services will probably work even for a pretty small language group, but specialized medical articles might create a problem. But then you find a retired orthopedic surgeon like Subas Chandra Rout…
On Sat, Feb 24, 2018 at 4:04 PM, James Heilman jmh649@gmail.com wrote:
I agree with John that it is very difficult to turn a translator into a
new
editor. I also agree with Jean-Philippe that it is key to have
involvement
of the local projects and preferable if they lead the efforts. Of the languages we worked in only one explicitly requested not to be
involved /
have translations from TWB.
James
On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad jeblad@gmail.com wrote:
You can turn it around; give added credits for translations from
small
language projects and into the larger ones, that is a lot more
interesting
than strictly translating from the larger language projects.
On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < jpbeland@wikimedia.ca
wrote:
I think the request for such projects should come from the
concerned
language projects, same for the list of articles. If not, in my
simple
opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <jeblad@gmail.com
wrote:
Should have added that the remaining points are somewhat less
interesting
in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles
should
also
be
pretty broad, not very narrow technical or medical, ie vertical
articles,
as the number of editors that can handle those will be pretty
small.
In particular: Do not believe you can turn a teanslator into a
new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
jeblad@gmail.com>
wrote:
> 1) You must start with high quality content and thus all
articles
are
>> extensively improved before being proposed for translation. > > > Note that to much pressure on "quality" can easily kill the
project.
> > 3) The "Content Translation" tool developed by the WMF made
efforts
more
>> efficient than handing around word documents. Would love to
see
that
tool >> improved further such as having it support specific lists of
articles
that >> are deemed ready for translation by certain groups. Would also
love
the
>> tool to have tracking metrics for these types of projects. > > > Didn't mention ContentTranslation, but it should be pretty
obvious.
> > 4) We used volunteer translators mostly associated with our
partner
>> Translators Without Borders. One issue we found was that
languages
in
>> which >> their are lots of translators such as French, Spanish, and
Italian
there
>> is >> often already at least some content on many of the topics in
question.
The >> issue than becomes integration which needs an expert
Wikipedia.
And
for
>> languages in which we have little content there are often few
avaliable
>> volunteers. > > > I used projects below 65k articles as an example, as the chance
of
> competing articles are pretty low. > > 5) With respect to "paying per word" the problem is this would
require
>> significant checks and balances to make sure people are taking
the
work
>> seriously and not simple using Google translate for the 70 or
so
languages >> in which it claims to work. We often had translations undergo
a
second
>> review and the volunteers at TWB have to pass certain tests to
be
>> accepted. > > > I'n my original email I wrote "verified good translators". It
is
as
> simple as "Has the editor contributed other articles at the
project?"
> > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <
jmh649@gmail.com
wrote:
> >> We learned a few things during the medical translation project
which
>> started back in 2011: >> >> 1) You must start with high quality content and thus all
articles
are
>> extensively improved before being proposed for translation. >> >> 2) A lot of languages want "less" content than is present on
EN
WP.
Thus
>> we >> moved to just improving and suggesting for translation the
leads
of
the
>> English articles. >> >> 3) The "Content Translation" tool developed by the WMF made
efforts
more
>> efficient than handing around word documents. Would love to
see
that
tool >> improved further such as having it support specific lists of
articles
that >> are deemed ready for translation by certain groups. Would also
love
the
>> tool to have tracking metrics for these types of projects. >> >> 4) We used volunteer translators mostly associated with our
partner
>> Translators Without Borders. One issue we found was that
languages
in
>> which >> their are lots of translators such as French, Spanish, and
Italian
there
>> is >> often already at least some content on many of the topics in
question.
The >> issue than becomes integration which needs an expert
Wikipedia.
And
for
>> languages in which we have little content there are often few
avaliable
>> volunteers. >> >> 5) With respect to "paying per word" the problem is this would
require
>> significant checks and balances to make sure people are taking
the
work
>> seriously and not simple using Google translate for the 70 or
so
languages >> in which it claims to work. We often had translations undergo
a
second
>> review and the volunteers at TWB have to pass certain tests to
be
>> accepted. >> >> 6) I hired a coordinator for the translation project for a
couple
of
>> years. >> The translators at TWB did not want to become Wikipedians or
learn
how
to >> use our systems. The coordinator created account like
TransSW001
(one
for >> each volunteer) and preloaded the article to be translated
into
Content
>> Translation. They than gave the volunteer translator the user
name
and
>> password to the account. >> >> 7) Were are we at now? There are currently just over 1,000
leads
of
>> articles that have been improved and are ready for
translation.
This
>> includes articles on the 440 medications that are on the WHO
Essential
>> List. We have worked a bit in some 100 languages. The efforts
have
>> resulted >> in more than 5 million works translated and integrated into
different
>> Wikipedias. The coordinator has unfortunately moved on to his
real
job
of >> teaching high school students. >> >> 8) The project continues but at a slower pace than before. The Wikipedian >> and retired orthopedic surgeon Subas Chandra Rout has
basically
single
>> handedly translated nearly all 1,000 leads into Odia a
language
spoken
by >> 40 million people in Eastern India. The amazing thing is that
for
many
of >> these topics this is the first and only information online
about
it.
>> Google >> translate does not even claim to work in this language. Our
partnerships
>> with WMTW and medical school in Taipai continue to translate
into
Chinese. >> There the students translate and than their translations are
reviewed
by
>> their profs before being posted. They translate in groups
using
hackpad
to >> make it more social. >> >> I am currently working to re invigorate the project :-) >> James >> >> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad <
jeblad@gmail.com
>> wrote: >> >> > This discussion is going to be fun! =D >> > >> > A little more than seventy Wikipedia-projects has more than
65k
>> articles, >> > the remaining two hundred or so are pretty small. >> > >> > What if a base set of articles were opened for paid
translators?
There
>> are >> > several lists of such base sets. We have both the thousand
articles
from >> > "List of articles every Wikipedia should have"[1] and and
the
ten
>> thousand >> > articles from the expanded list[2]. >> > >> > Lets say verified good translators was paid about $0.01 per
word
(about >> $1 >> > for a 1k-article) for translating one of those articles into
another
>> > language, with perhaps a higher pay for contributors in
high-cost
>> > countries. The pay would also have to be higher for
languages
that
lacks >> > good translation tools. >> > >> > I believe this would be an _enabling_ activity for the
communities,
as
>> > without a base set of articles it won't be possible to
build a
>> community at >> > all. By not paying for new articles, and only translating >> well-referenced >> > articles, some of the disputes in the communities could be
avoided.
>> Perhaps >> > we should also identify good source articles, that would be
a
help.
>> > Translated articles should be above some minimum size, but
they
does
not >> > have to be full translations of the source article. >> > >> > A real problem is that our existing lists of good articles
other
>> projects >> > should have is pretty much biased towards Western World, so
they
need
a >> lot >> > of adjustments. Perhaps such a project would identify our
inherit
bias? >> > >> > [1] >> > https://meta.wikimedia.org/wiki/List_of_articles_every_ >> > Wikipedia_should_have >> > [2] >> > https://meta.wikimedia.org/wiki/List_of_articles_every_ >> > Wikipedia_should_have/Expanded >> > _______________________________________________ >> > Wikimedia-l mailing list, guidelines at:
>> > wiki/Mailing_lists/Guidelines and
>> > wiki/Wikimedia-l >> > New messages to: Wikimedia-l@lists.wikimedia.org >> > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
, >> > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
>> >> >> >> >> -- >> James Heilman >> MD, CCFP-EM, Wikipedian >> _______________________________________________ >> Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wik
>> i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik >> i/Wikimedia-l >> New messages to: Wikimedia-l@lists.wikimedia.org >> Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
,
>> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
>> > > _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Cultural appropriation is something different, by "forcing" the contents in a minority language we would actually be at risk of implementing a form of "cultural colonialism" which is the opposite of a cultural appropriation.
NOTE: I refer to "the Western" in both cultural and "Wikipedian" sense: I mean cultures with a strong presence on the web plus developed and flourishing Wikipedia communities.
Helping minority languages with funds/workforce is not bad in my opinion, but I think a bottom-up process must be followed, with the "bottom" being as closer as possible to relevant linguistic/cultural communities. A Wikipedia full of "what the Westerns think is important" in a minority non-Western language would definitely fail project scopes.
This kind of problem almost does not arise with minority language associated to Western cultures since they share the same cultural backgrounds: back to my previous example the cultural background of Sicilian is substantially equal to Italian one. Still, as I already wrote, wikis in minority languages should focus on a certain aspect of wiki scope: Wiki has roughly two main scopes: 1) sharing knowledge in a certain language 2) also preserving the cultural heritage associated with different languages. For languages mainly spoken as first language the "sharing knowledge" aspect is predominant, while the second should take precedence in languages whose speakers are native speakers of a "bigger" language.
Vito
2018-02-24 22:58 GMT+01:00 John Erling Blad jeblad@gmail.com:
Seems like this is mostly about cultural ownership and appropriation. Not sure if it is possible to agree on this.
On Sat, Feb 24, 2018 at 6:08 PM, Vi to vituzzu.wiki@gmail.com wrote:
I'll reply to the most recent email just for laziness.
I'm doubtful for a series of reasons, most of were already expressed in a better way by others: *a remuneration in terms of quantity will weaken the quality of translations unless there's a strong mechanism of quality verification requiring a quantity of resources comparable to translations themselves; *articles are the result of a long process which reflects cultural
identity
of different communities, I'm not confident with transferring them to a different "weaker" cultures. My usage of "weaker" adjective only focuses about the strength of a cultural presence on the Internet; *articles to be translated are at high risk of reflecting the cultural identity (and biases) of the Western culture; *finally I think paid translators would hardly turn into stable Wikipedians.
IMHO some paid editing may be better exploited in order to digitalise
texts
of unrepresented cultures (wikisource) or preserving their vocabularies (wiktionary).
Also those languages which are secondary for all their speakers should be dealt with in a different fashion. I, for one, am a native speaker of specific variant of Sicilian, Sicilian is a secondary language to any of its speakers. Honestly, I'd find pointless to read the biography of Leonardo da Vinci in Sicilian while I can find thousands of books about
him
in Italian. Also I find this kind of translation creates a fake
"literary"
language totally detached from reality: there's no "encaustic painting"
in
Sicilian, still a Sicilian article about Leonardo will invent one.
As a general principle we should always collect, rather than create, knowledge.
Vito
2018-02-24 16:30 GMT+01:00 John Erling Blad jeblad@gmail.com:
My reply can be read as a bit more harsh than intended, it was merely a statement about my present experience about translators in general.
The problem with lack of contributors (and translators) in a
specialized
area is that there is a small community, and within this community some kind of selection is made. Each time a selection is repeated the
remaining
group shrinks. Specialize the selection sufficiently many times and
there
will be no contributors (or translators) left. It is simply a game of probabilities. Thus, to make such a project work it must have a sufficiently broad scope for the articles. Articles about public health services will probably work even for a pretty small language group, but specialized medical articles might create a problem. But then you find a retired orthopedic surgeon like Subas Chandra Rout…
On Sat, Feb 24, 2018 at 4:04 PM, James Heilman jmh649@gmail.com
wrote:
I agree with John that it is very difficult to turn a translator
into a
new
editor. I also agree with Jean-Philippe that it is key to have
involvement
of the local projects and preferable if they lead the efforts. Of the languages we worked in only one explicitly requested not to be
involved /
have translations from TWB.
James
On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad jeblad@gmail.com wrote:
You can turn it around; give added credits for translations from
small
language projects and into the larger ones, that is a lot more
interesting
than strictly translating from the larger language projects.
On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < jpbeland@wikimedia.ca
wrote:
I think the request for such projects should come from the
concerned
language projects, same for the list of articles. If not, in my
simple
opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <
jeblad@gmail.com
wrote:
> Should have added that the remaining points are somewhat less
interesting
> in this context. Preloading a set of articles is a bad idea,
the
> translators should be able to chose for themselves. Articles
should
also
be > pretty broad, not very narrow technical or medical, ie vertical
articles,
> as the number of editors that can handle those will be pretty
small.
> > In particular: Do not believe you can turn a teanslator into a
new
editor! > You can although turn an existing editor into a translator. > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
jeblad@gmail.com>
> wrote: > > > 1) You must start with high quality content and thus all
articles
are
> >> extensively improved before being proposed for translation. > > > > > > Note that to much pressure on "quality" can easily kill the
project.
> > > > 3) The "Content Translation" tool developed by the WMF made
efforts
more > >> efficient than handing around word documents. Would love to
see
that
> tool > >> improved further such as having it support specific lists of
articles
> that > >> are deemed ready for translation by certain groups. Would
also
love
the > >> tool to have tracking metrics for these types of projects. > > > > > > Didn't mention ContentTranslation, but it should be pretty
obvious.
> > > > 4) We used volunteer translators mostly associated with our
partner
> >> Translators Without Borders. One issue we found was that
languages
in
> >> which > >> their are lots of translators such as French, Spanish, and
Italian
there > >> is > >> often already at least some content on many of the topics in
question.
> The > >> issue than becomes integration which needs an expert
Wikipedia.
And
for > >> languages in which we have little content there are often
few
avaliable > >> volunteers. > > > > > > I used projects below 65k articles as an example, as the
chance
of
> > competing articles are pretty low. > > > > 5) With respect to "paying per word" the problem is this
would
require
> >> significant checks and balances to make sure people are
taking
the
work > >> seriously and not simple using Google translate for the 70
or
so
> languages > >> in which it claims to work. We often had translations
undergo
a
second
> >> review and the volunteers at TWB have to pass certain tests
to
be
> >> accepted. > > > > > > I'n my original email I wrote "verified good translators". It
is
as
> > simple as "Has the editor contributed other articles at the
project?"
> > > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <
jmh649@gmail.com
wrote: > > > >> We learned a few things during the medical translation
project
which
> >> started back in 2011: > >> > >> 1) You must start with high quality content and thus all
articles
are
> >> extensively improved before being proposed for translation. > >> > >> 2) A lot of languages want "less" content than is present on
EN
WP.
Thus > >> we > >> moved to just improving and suggesting for translation the
leads
of
the > >> English articles. > >> > >> 3) The "Content Translation" tool developed by the WMF made
efforts
more > >> efficient than handing around word documents. Would love to
see
that
> tool > >> improved further such as having it support specific lists of
articles
> that > >> are deemed ready for translation by certain groups. Would
also
love
the > >> tool to have tracking metrics for these types of projects. > >> > >> 4) We used volunteer translators mostly associated with our
partner
> >> Translators Without Borders. One issue we found was that
languages
in
> >> which > >> their are lots of translators such as French, Spanish, and
Italian
there > >> is > >> often already at least some content on many of the topics in
question.
> The > >> issue than becomes integration which needs an expert
Wikipedia.
And
for > >> languages in which we have little content there are often
few
avaliable > >> volunteers. > >> > >> 5) With respect to "paying per word" the problem is this
would
require
> >> significant checks and balances to make sure people are
taking
the
work > >> seriously and not simple using Google translate for the 70
or
so
> languages > >> in which it claims to work. We often had translations
undergo
a
second
> >> review and the volunteers at TWB have to pass certain tests
to
be
> >> accepted. > >> > >> 6) I hired a coordinator for the translation project for a
couple
of
> >> years. > >> The translators at TWB did not want to become Wikipedians or
learn
how
> to > >> use our systems. The coordinator created account like
TransSW001
(one
> for > >> each volunteer) and preloaded the article to be translated
into
Content > >> Translation. They than gave the volunteer translator the
user
name
and
> >> password to the account. > >> > >> 7) Were are we at now? There are currently just over 1,000
leads
of
> >> articles that have been improved and are ready for
translation.
This
> >> includes articles on the 440 medications that are on the WHO
Essential
> >> List. We have worked a bit in some 100 languages. The
efforts
have
> >> resulted > >> in more than 5 million works translated and integrated into
different
> >> Wikipedias. The coordinator has unfortunately moved on to
his
real
job
> of > >> teaching high school students. > >> > >> 8) The project continues but at a slower pace than before.
The
> Wikipedian > >> and retired orthopedic surgeon Subas Chandra Rout has
basically
single
> >> handedly translated nearly all 1,000 leads into Odia a
language
spoken
> by > >> 40 million people in Eastern India. The amazing thing is
that
for
many
> of > >> these topics this is the first and only information online
about
it.
> >> Google > >> translate does not even claim to work in this language. Our partnerships > >> with WMTW and medical school in Taipai continue to translate
into
> Chinese. > >> There the students translate and than their translations are
reviewed
by > >> their profs before being posted. They translate in groups
using
hackpad > to > >> make it more social. > >> > >> I am currently working to re invigorate the project :-) > >> James > >> > >> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad <
jeblad@gmail.com
> >> wrote: > >> > >> > This discussion is going to be fun! =D > >> > > >> > A little more than seventy Wikipedia-projects has more
than
65k
> >> articles, > >> > the remaining two hundred or so are pretty small. > >> > > >> > What if a base set of articles were opened for paid
translators?
There > >> are > >> > several lists of such base sets. We have both the thousand
articles
> from > >> > "List of articles every Wikipedia should have"[1] and and
the
ten
> >> thousand > >> > articles from the expanded list[2]. > >> > > >> > Lets say verified good translators was paid about $0.01
per
word
> (about > >> $1 > >> > for a 1k-article) for translating one of those articles
into
another
> >> > language, with perhaps a higher pay for contributors in
high-cost
> >> > countries. The pay would also have to be higher for
languages
that
> lacks > >> > good translation tools. > >> > > >> > I believe this would be an _enabling_ activity for the
communities,
as > >> > without a base set of articles it won't be possible to
build a
> >> community at > >> > all. By not paying for new articles, and only translating > >> well-referenced > >> > articles, some of the disputes in the communities could be
avoided.
> >> Perhaps > >> > we should also identify good source articles, that would
be
a
help.
> >> > Translated articles should be above some minimum size, but
they
does
> not > >> > have to be full translations of the source article. > >> > > >> > A real problem is that our existing lists of good articles
other
> >> projects > >> > should have is pretty much biased towards Western World,
so
they
need > a > >> lot > >> > of adjustments. Perhaps such a project would identify our
inherit
> bias? > >> > > >> > [1] > >> > https://meta.wikimedia.org/wiki/List_of_articles_every_ > >> > Wikipedia_should_have > >> > [2] > >> > https://meta.wikimedia.org/wiki/List_of_articles_every_ > >> > Wikipedia_should_have/Expanded > >> > _______________________________________________ > >> > Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ > >> > wiki/Mailing_lists/Guidelines and
> >> > wiki/Wikimedia-l > >> > New messages to: Wikimedia-l@lists.wikimedia.org > >> > Unsubscribe: https://lists.wikimedia.org/ mailman/listinfo/wikimedia-l > , > >> > mailto:wikimedia-l-request@lists.wikimedia.org?subject= unsubscribe > >> > >> > >> > >> > >> -- > >> James Heilman > >> MD, CCFP-EM, Wikipedian > >> _______________________________________________ > >> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik > >> i/Mailing_lists/Guidelines and
https://meta.wikimedia.org/wik
> >> i/Wikimedia-l > >> New messages to: Wikimedia-l@lists.wikimedia.org > >> Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
, > >> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> >> > > > > > _______________________________________________ > Wikimedia-l mailing list, guidelines at: > https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and > https://meta.wikimedia.org/wiki/Wikimedia-l > New messages to: Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Sorry, but this does not make sense. The core articles apply globally. There will although be articles in additions to a list of core articles, but I don't try to advocate any of those lists as the one and only list. Actually I have toyed with an idea of automatically create a list of core articles, and that would identify important articles no matter if they are from a big western language or a minority language.
The main problem is NOT that minority languages should have articles about the major cities and important philosophers, *the main problem is that minor languages can't get started because they lack content*!
On Sun, Feb 25, 2018 at 2:41 AM, Vi to vituzzu.wiki@gmail.com wrote:
Cultural appropriation is something different, by "forcing" the contents in a minority language we would actually be at risk of implementing a form of "cultural colonialism" which is the opposite of a cultural appropriation.
NOTE: I refer to "the Western" in both cultural and "Wikipedian" sense: I mean cultures with a strong presence on the web plus developed and flourishing Wikipedia communities.
Helping minority languages with funds/workforce is not bad in my opinion, but I think a bottom-up process must be followed, with the "bottom" being as closer as possible to relevant linguistic/cultural communities. A Wikipedia full of "what the Westerns think is important" in a minority non-Western language would definitely fail project scopes.
This kind of problem almost does not arise with minority language associated to Western cultures since they share the same cultural backgrounds: back to my previous example the cultural background of Sicilian is substantially equal to Italian one. Still, as I already wrote, wikis in minority languages should focus on a certain aspect of wiki scope: Wiki has roughly two main scopes: 1) sharing knowledge in a certain language 2) also preserving the cultural heritage associated with different languages. For languages mainly spoken as first language the "sharing knowledge" aspect is predominant, while the second should take precedence in languages whose speakers are native speakers of a "bigger" language.
Vito
2018-02-24 22:58 GMT+01:00 John Erling Blad jeblad@gmail.com:
Seems like this is mostly about cultural ownership and appropriation. Not sure if it is possible to agree on this.
On Sat, Feb 24, 2018 at 6:08 PM, Vi to vituzzu.wiki@gmail.com wrote:
I'll reply to the most recent email just for laziness.
I'm doubtful for a series of reasons, most of were already expressed
in a
better way by others: *a remuneration in terms of quantity will weaken the quality of translations unless there's a strong mechanism of quality verification requiring a quantity of resources comparable to translations
themselves;
*articles are the result of a long process which reflects cultural
identity
of different communities, I'm not confident with transferring them to a different "weaker" cultures. My usage of "weaker" adjective only
focuses
about the strength of a cultural presence on the Internet; *articles to be translated are at high risk of reflecting the cultural identity (and biases) of the Western culture; *finally I think paid translators would hardly turn into stable Wikipedians.
IMHO some paid editing may be better exploited in order to digitalise
texts
of unrepresented cultures (wikisource) or preserving their vocabularies (wiktionary).
Also those languages which are secondary for all their speakers should
be
dealt with in a different fashion. I, for one, am a native speaker of specific variant of Sicilian, Sicilian is a secondary language to any
of
its speakers. Honestly, I'd find pointless to read the biography of Leonardo da Vinci in Sicilian while I can find thousands of books about
him
in Italian. Also I find this kind of translation creates a fake
"literary"
language totally detached from reality: there's no "encaustic painting"
in
Sicilian, still a Sicilian article about Leonardo will invent one.
As a general principle we should always collect, rather than create, knowledge.
Vito
2018-02-24 16:30 GMT+01:00 John Erling Blad jeblad@gmail.com:
My reply can be read as a bit more harsh than intended, it was
merely a
statement about my present experience about translators in general.
The problem with lack of contributors (and translators) in a
specialized
area is that there is a small community, and within this community
some
kind of selection is made. Each time a selection is repeated the
remaining
group shrinks. Specialize the selection sufficiently many times and
there
will be no contributors (or translators) left. It is simply a game of probabilities. Thus, to make such a project work it must have a sufficiently broad scope for the articles. Articles about public
health
services will probably work even for a pretty small language group,
but
specialized medical articles might create a problem. But then you
find
a retired orthopedic surgeon like Subas Chandra Rout…
On Sat, Feb 24, 2018 at 4:04 PM, James Heilman jmh649@gmail.com
wrote:
I agree with John that it is very difficult to turn a translator
into a
new
editor. I also agree with Jean-Philippe that it is key to have
involvement
of the local projects and preferable if they lead the efforts. Of
the
languages we worked in only one explicitly requested not to be
involved /
have translations from TWB.
James
On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad <
jeblad@gmail.com>
wrote:
You can turn it around; give added credits for translations from
small
language projects and into the larger ones, that is a lot more
interesting
than strictly translating from the larger language projects.
On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < jpbeland@wikimedia.ca > wrote:
> I think the request for such projects should come from the
concerned
> language projects, same for the list of articles. If not, in my
simple
> opinion, it is a form of coloniasm again. > > Jean-Philippe Béland > Vice President, Wikimedia Canada > > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <
jeblad@gmail.com
wrote: > > > Should have added that the remaining points are somewhat less interesting > > in this context. Preloading a set of articles is a bad idea,
the
> > translators should be able to chose for themselves. Articles
should
also > be > > pretty broad, not very narrow technical or medical, ie
vertical
articles, > > as the number of editors that can handle those will be pretty
small.
> > > > In particular: Do not believe you can turn a teanslator into
a
new
> editor! > > You can although turn an existing editor into a translator. > > > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
jeblad@gmail.com>
> > wrote: > > > > > 1) You must start with high quality content and thus all
articles
are
> > >> extensively improved before being proposed for
translation.
> > > > > > > > > Note that to much pressure on "quality" can easily kill the
project.
> > > > > > 3) The "Content Translation" tool developed by the WMF made
efforts
> more > > >> efficient than handing around word documents. Would love
to
see
that
> > tool > > >> improved further such as having it support specific lists
of
articles > > that > > >> are deemed ready for translation by certain groups. Would
also
love
> the > > >> tool to have tracking metrics for these types of projects. > > > > > > > > > Didn't mention ContentTranslation, but it should be pretty
obvious.
> > > > > > 4) We used volunteer translators mostly associated with our
partner
> > >> Translators Without Borders. One issue we found was that
languages
in > > >> which > > >> their are lots of translators such as French, Spanish, and
Italian
> there > > >> is > > >> often already at least some content on many of the topics
in
question. > > The > > >> issue than becomes integration which needs an expert
Wikipedia.
And
> for > > >> languages in which we have little content there are often
few
> avaliable > > >> volunteers. > > > > > > > > > I used projects below 65k articles as an example, as the
chance
of
> > > competing articles are pretty low. > > > > > > 5) With respect to "paying per word" the problem is this
would
require > > >> significant checks and balances to make sure people are
taking
the
> work > > >> seriously and not simple using Google translate for the 70
or
so
> > languages > > >> in which it claims to work. We often had translations
undergo
a
second > > >> review and the volunteers at TWB have to pass certain
tests
to
be
> > >> accepted. > > > > > > > > > I'n my original email I wrote "verified good translators".
It
is
as
> > > simple as "Has the editor contributed other articles at the
project?"
> > > > > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <
jmh649@gmail.com
> wrote: > > > > > >> We learned a few things during the medical translation
project
which
> > >> started back in 2011: > > >> > > >> 1) You must start with high quality content and thus all
articles
are > > >> extensively improved before being proposed for
translation.
> > >> > > >> 2) A lot of languages want "less" content than is present
on
EN
WP.
> Thus > > >> we > > >> moved to just improving and suggesting for translation the
leads
of
> the > > >> English articles. > > >> > > >> 3) The "Content Translation" tool developed by the WMF
made
efforts
> more > > >> efficient than handing around word documents. Would love
to
see
that
> > tool > > >> improved further such as having it support specific lists
of
articles > > that > > >> are deemed ready for translation by certain groups. Would
also
love
> the > > >> tool to have tracking metrics for these types of projects. > > >> > > >> 4) We used volunteer translators mostly associated with
our
partner
> > >> Translators Without Borders. One issue we found was that
languages
in > > >> which > > >> their are lots of translators such as French, Spanish, and
Italian
> there > > >> is > > >> often already at least some content on many of the topics
in
question. > > The > > >> issue than becomes integration which needs an expert
Wikipedia.
And
> for > > >> languages in which we have little content there are often
few
> avaliable > > >> volunteers. > > >> > > >> 5) With respect to "paying per word" the problem is this
would
require > > >> significant checks and balances to make sure people are
taking
the
> work > > >> seriously and not simple using Google translate for the 70
or
so
> > languages > > >> in which it claims to work. We often had translations
undergo
a
second > > >> review and the volunteers at TWB have to pass certain
tests
to
be
> > >> accepted. > > >> > > >> 6) I hired a coordinator for the translation project for a
couple
of
> > >> years. > > >> The translators at TWB did not want to become Wikipedians
or
learn
how > > to > > >> use our systems. The coordinator created account like
TransSW001
(one > > for > > >> each volunteer) and preloaded the article to be translated
into
> Content > > >> Translation. They than gave the volunteer translator the
user
name
and > > >> password to the account. > > >> > > >> 7) Were are we at now? There are currently just over 1,000
leads
of
> > >> articles that have been improved and are ready for
translation.
This
> > >> includes articles on the 440 medications that are on the
WHO
Essential > > >> List. We have worked a bit in some 100 languages. The
efforts
have
> > >> resulted > > >> in more than 5 million works translated and integrated
into
different > > >> Wikipedias. The coordinator has unfortunately moved on to
his
real
job > > of > > >> teaching high school students. > > >> > > >> 8) The project continues but at a slower pace than before.
The
> > Wikipedian > > >> and retired orthopedic surgeon Subas Chandra Rout has
basically
single > > >> handedly translated nearly all 1,000 leads into Odia a
language
spoken > > by > > >> 40 million people in Eastern India. The amazing thing is
that
for
many > > of > > >> these topics this is the first and only information online
about
it.
> > >> Google > > >> translate does not even claim to work in this language.
Our
> partnerships > > >> with WMTW and medical school in Taipai continue to
translate
into
> > Chinese. > > >> There the students translate and than their translations
are
reviewed > by > > >> their profs before being posted. They translate in groups
using
> hackpad > > to > > >> make it more social. > > >> > > >> I am currently working to re invigorate the project :-) > > >> James > > >> > > >> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad <
jeblad@gmail.com
> > > >> wrote: > > >> > > >> > This discussion is going to be fun! =D > > >> > > > >> > A little more than seventy Wikipedia-projects has more
than
65k
> > >> articles, > > >> > the remaining two hundred or so are pretty small. > > >> > > > >> > What if a base set of articles were opened for paid
translators?
> There > > >> are > > >> > several lists of such base sets. We have both the
thousand
articles > > from > > >> > "List of articles every Wikipedia should have"[1] and
and
the
ten
> > >> thousand > > >> > articles from the expanded list[2]. > > >> > > > >> > Lets say verified good translators was paid about $0.01
per
word
> > (about > > >> $1 > > >> > for a 1k-article) for translating one of those articles
into
another > > >> > language, with perhaps a higher pay for contributors in
high-cost
> > >> > countries. The pay would also have to be higher for
languages
that
> > lacks > > >> > good translation tools. > > >> > > > >> > I believe this would be an _enabling_ activity for the communities, > as > > >> > without a base set of articles it won't be possible to
build a
> > >> community at > > >> > all. By not paying for new articles, and only
translating
> > >> well-referenced > > >> > articles, some of the disputes in the communities could
be
avoided. > > >> Perhaps > > >> > we should also identify good source articles, that would
be
a
help. > > >> > Translated articles should be above some minimum size,
but
they
does > > not > > >> > have to be full translations of the source article. > > >> > > > >> > A real problem is that our existing lists of good
articles
other
> > >> projects > > >> > should have is pretty much biased towards Western World,
so
they
> need > > a > > >> lot > > >> > of adjustments. Perhaps such a project would identify
our
inherit
> > bias? > > >> > > > >> > [1] > > >> > https://meta.wikimedia.org/wiki/List_of_articles_every_ > > >> > Wikipedia_should_have > > >> > [2] > > >> > https://meta.wikimedia.org/wiki/List_of_articles_every_ > > >> > Wikipedia_should_have/Expanded > > >> > _______________________________________________ > > >> > Wikimedia-l mailing list, guidelines at: > https://meta.wikimedia.org/ > > >> > wiki/Mailing_lists/Guidelines and
> > >> > wiki/Wikimedia-l > > >> > New messages to: Wikimedia-l@lists.wikimedia.org > > >> > Unsubscribe: https://lists.wikimedia.org/ > mailman/listinfo/wikimedia-l > > , > > >> > <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> unsubscribe> > > >> > > >> > > >> > > >> > > >> -- > > >> James Heilman > > >> MD, CCFP-EM, Wikipedian > > >> _______________________________________________ > > >> Wikimedia-l mailing list, guidelines at: > https://meta.wikimedia.org/wik > > >> i/Mailing_lists/Guidelines and
https://meta.wikimedia.org/wik
> > >> i/Wikimedia-l > > >> New messages to: Wikimedia-l@lists.wikimedia.org > > >> Unsubscribe: https://lists.wikimedia.org/ mailman/listinfo/wikimedia-l > , > > >> mailto:wikimedia-l-request@lists.wikimedia.org?subject= unsubscribe > > >> > > > > > > > > _______________________________________________ > > Wikimedia-l mailing list, guidelines at: > > https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and > > https://meta.wikimedia.org/wiki/Wikimedia-l > > New messages to: Wikimedia-l@lists.wikimedia.org > > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> _______________________________________________ > Wikimedia-l mailing list, guidelines at:
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ > wiki/Wikimedia-l > New messages to: Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> _______________________________________________ Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Any "global" list reflects (and I fear it will always reflect) the Weltanschauung of those cultures which are stronger on the web.
I'm deeply concerned about cultures being eaten up by globalization but attempts to preserve them should take into account the risk of ending up preserving just "our" view of these cultures.
I also agree with WereSpielChequers' comments about mixing paid and unpaid editing. What I think it can be done is a system of prizes/contests (maybe evaluated by paid experts) focused on attracting people on Wikisource and Wiktionaries, Wikipedia can follow if a critical mass is eventually reached.
Vito
2018-02-25 15:16 GMT+01:00 John Erling Blad jeblad@gmail.com:
Sorry, but this does not make sense. The core articles apply globally. There will although be articles in additions to a list of core articles, but I don't try to advocate any of those lists as the one and only list. Actually I have toyed with an idea of automatically create a list of core articles, and that would identify important articles no matter if they are from a big western language or a minority language.
The main problem is NOT that minority languages should have articles about the major cities and important philosophers, *the main problem is that minor languages can't get started because they lack content*!
On Sun, Feb 25, 2018 at 2:41 AM, Vi to vituzzu.wiki@gmail.com wrote:
Cultural appropriation is something different, by "forcing" the contents
in
a minority language we would actually be at risk of implementing a form
of
"cultural colonialism" which is the opposite of a cultural appropriation.
NOTE: I refer to "the Western" in both cultural and "Wikipedian" sense: I mean cultures with a strong presence on the web plus developed and flourishing Wikipedia communities.
Helping minority languages with funds/workforce is not bad in my opinion, but I think a bottom-up process must be followed, with the "bottom" being as closer as possible to relevant linguistic/cultural communities. A Wikipedia full of "what the Westerns think is important" in a minority non-Western language would definitely fail project scopes.
This kind of problem almost does not arise with minority language associated to Western cultures since they share the same cultural backgrounds: back to my previous example the cultural background of Sicilian is substantially equal to Italian one. Still, as I already
wrote,
wikis in minority languages should focus on a certain aspect of wiki
scope:
Wiki has roughly two main scopes: 1) sharing knowledge in a certain language 2) also preserving the cultural heritage associated with
different
languages. For languages mainly spoken as first language the "sharing knowledge" aspect is predominant, while the second should take precedence in languages whose speakers are native speakers of a "bigger" language.
Vito
2018-02-24 22:58 GMT+01:00 John Erling Blad jeblad@gmail.com:
Seems like this is mostly about cultural ownership and appropriation.
Not
sure if it is possible to agree on this.
On Sat, Feb 24, 2018 at 6:08 PM, Vi to vituzzu.wiki@gmail.com wrote:
I'll reply to the most recent email just for laziness.
I'm doubtful for a series of reasons, most of were already expressed
in a
better way by others: *a remuneration in terms of quantity will weaken the quality of translations unless there's a strong mechanism of quality
verification
requiring a quantity of resources comparable to translations
themselves;
*articles are the result of a long process which reflects cultural
identity
of different communities, I'm not confident with transferring them
to a
different "weaker" cultures. My usage of "weaker" adjective only
focuses
about the strength of a cultural presence on the Internet; *articles to be translated are at high risk of reflecting the
cultural
identity (and biases) of the Western culture; *finally I think paid translators would hardly turn into stable Wikipedians.
IMHO some paid editing may be better exploited in order to digitalise
texts
of unrepresented cultures (wikisource) or preserving their
vocabularies
(wiktionary).
Also those languages which are secondary for all their speakers
should
be
dealt with in a different fashion. I, for one, am a native speaker of specific variant of Sicilian, Sicilian is a secondary language to any
of
its speakers. Honestly, I'd find pointless to read the biography of Leonardo da Vinci in Sicilian while I can find thousands of books
about
him
in Italian. Also I find this kind of translation creates a fake
"literary"
language totally detached from reality: there's no "encaustic
painting"
in
Sicilian, still a Sicilian article about Leonardo will invent one.
As a general principle we should always collect, rather than create, knowledge.
Vito
2018-02-24 16:30 GMT+01:00 John Erling Blad jeblad@gmail.com:
My reply can be read as a bit more harsh than intended, it was
merely a
statement about my present experience about translators in general.
The problem with lack of contributors (and translators) in a
specialized
area is that there is a small community, and within this community
some
kind of selection is made. Each time a selection is repeated the
remaining
group shrinks. Specialize the selection sufficiently many times and
there
will be no contributors (or translators) left. It is simply a game
of
probabilities. Thus, to make such a project work it must have a sufficiently broad scope for the articles. Articles about public
health
services will probably work even for a pretty small language group,
but
specialized medical articles might create a problem. But then you
find
a retired orthopedic surgeon like Subas Chandra Rout…
On Sat, Feb 24, 2018 at 4:04 PM, James Heilman jmh649@gmail.com
wrote:
I agree with John that it is very difficult to turn a translator
into a
new
editor. I also agree with Jean-Philippe that it is key to have
involvement
of the local projects and preferable if they lead the efforts. Of
the
languages we worked in only one explicitly requested not to be
involved /
have translations from TWB.
James
On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad <
jeblad@gmail.com>
wrote:
> You can turn it around; give added credits for translations
from
small
> language projects and into the larger ones, that is a lot more interesting > than strictly translating from the larger language projects. > > On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < > jpbeland@wikimedia.ca > > wrote: > > > I think the request for such projects should come from the
concerned
> > language projects, same for the list of articles. If not, in
my
simple
> > opinion, it is a form of coloniasm again. > > > > Jean-Philippe Béland > > Vice President, Wikimedia Canada > > > > > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <
jeblad@gmail.com
> wrote: > > > > > Should have added that the remaining points are somewhat
less
> interesting > > > in this context. Preloading a set of articles is a bad
idea,
the
> > > translators should be able to chose for themselves.
Articles
should
> also > > be > > > pretty broad, not very narrow technical or medical, ie
vertical
> articles, > > > as the number of editors that can handle those will be
pretty
small.
> > > > > > In particular: Do not believe you can turn a teanslator
into
a
new
> > editor! > > > You can although turn an existing editor into a translator. > > > > > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
jeblad@gmail.com>
> > > wrote: > > > > > > > 1) You must start with high quality content and thus all
articles
are > > > >> extensively improved before being proposed for
translation.
> > > > > > > > > > > > Note that to much pressure on "quality" can easily kill
the
project. > > > > > > > > 3) The "Content Translation" tool developed by the WMF
made
efforts
> > more > > > >> efficient than handing around word documents. Would love
to
see
that > > > tool > > > >> improved further such as having it support specific
lists
of
> articles > > > that > > > >> are deemed ready for translation by certain groups.
Would
also
love > > the > > > >> tool to have tracking metrics for these types of
projects.
> > > > > > > > > > > > Didn't mention ContentTranslation, but it should be
pretty
obvious.
> > > > > > > > 4) We used volunteer translators mostly associated with
our
partner
> > > >> Translators Without Borders. One issue we found was that
languages
> in > > > >> which > > > >> their are lots of translators such as French, Spanish,
and
Italian
> > there > > > >> is > > > >> often already at least some content on many of the
topics
in
> question. > > > The > > > >> issue than becomes integration which needs an expert
Wikipedia.
And > > for > > > >> languages in which we have little content there are
often
few
> > avaliable > > > >> volunteers. > > > > > > > > > > > > I used projects below 65k articles as an example, as the
chance
of
> > > > competing articles are pretty low. > > > > > > > > 5) With respect to "paying per word" the problem is this
would
> require > > > >> significant checks and balances to make sure people are
taking
the
> > work > > > >> seriously and not simple using Google translate for the
70
or
so
> > > languages > > > >> in which it claims to work. We often had translations
undergo
a
> second > > > >> review and the volunteers at TWB have to pass certain
tests
to
be
> > > >> accepted. > > > > > > > > > > > > I'n my original email I wrote "verified good
translators".
It
is
as
> > > > simple as "Has the editor contributed other articles at
the
project?" > > > > > > > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <
jmh649@gmail.com
> > wrote: > > > > > > > >> We learned a few things during the medical translation
project
which > > > >> started back in 2011: > > > >> > > > >> 1) You must start with high quality content and thus all
articles
> are > > > >> extensively improved before being proposed for
translation.
> > > >> > > > >> 2) A lot of languages want "less" content than is
present
on
EN
WP. > > Thus > > > >> we > > > >> moved to just improving and suggesting for translation
the
leads
of > > the > > > >> English articles. > > > >> > > > >> 3) The "Content Translation" tool developed by the WMF
made
efforts > > more > > > >> efficient than handing around word documents. Would love
to
see
that > > > tool > > > >> improved further such as having it support specific
lists
of
> articles > > > that > > > >> are deemed ready for translation by certain groups.
Would
also
love > > the > > > >> tool to have tracking metrics for these types of
projects.
> > > >> > > > >> 4) We used volunteer translators mostly associated with
our
partner > > > >> Translators Without Borders. One issue we found was that
languages
> in > > > >> which > > > >> their are lots of translators such as French, Spanish,
and
Italian
> > there > > > >> is > > > >> often already at least some content on many of the
topics
in
> question. > > > The > > > >> issue than becomes integration which needs an expert
Wikipedia.
And > > for > > > >> languages in which we have little content there are
often
few
> > avaliable > > > >> volunteers. > > > >> > > > >> 5) With respect to "paying per word" the problem is this
would
> require > > > >> significant checks and balances to make sure people are
taking
the
> > work > > > >> seriously and not simple using Google translate for the
70
or
so
> > > languages > > > >> in which it claims to work. We often had translations
undergo
a
> second > > > >> review and the volunteers at TWB have to pass certain
tests
to
be
> > > >> accepted. > > > >> > > > >> 6) I hired a coordinator for the translation project
for a
couple
of > > > >> years. > > > >> The translators at TWB did not want to become
Wikipedians
or
learn
> how > > > to > > > >> use our systems. The coordinator created account like
TransSW001
> (one > > > for > > > >> each volunteer) and preloaded the article to be
translated
into
> > Content > > > >> Translation. They than gave the volunteer translator the
user
name
> and > > > >> password to the account. > > > >> > > > >> 7) Were are we at now? There are currently just over
1,000
leads
of > > > >> articles that have been improved and are ready for
translation.
This > > > >> includes articles on the 440 medications that are on the
WHO
> Essential > > > >> List. We have worked a bit in some 100 languages. The
efforts
have
> > > >> resulted > > > >> in more than 5 million works translated and integrated
into
> different > > > >> Wikipedias. The coordinator has unfortunately moved on
to
his
real
> job > > > of > > > >> teaching high school students. > > > >> > > > >> 8) The project continues but at a slower pace than
before.
The
> > > Wikipedian > > > >> and retired orthopedic surgeon Subas Chandra Rout has
basically
> single > > > >> handedly translated nearly all 1,000 leads into Odia a
language
> spoken > > > by > > > >> 40 million people in Eastern India. The amazing thing is
that
for
> many > > > of > > > >> these topics this is the first and only information
online
about
it. > > > >> Google > > > >> translate does not even claim to work in this language.
Our
> > partnerships > > > >> with WMTW and medical school in Taipai continue to
translate
into
> > > Chinese. > > > >> There the students translate and than their translations
are
> reviewed > > by > > > >> their profs before being posted. They translate in
groups
using
> > hackpad > > > to > > > >> make it more social. > > > >> > > > >> I am currently working to re invigorate the project :-) > > > >> James > > > >> > > > >> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad < jeblad@gmail.com > > > > > >> wrote: > > > >> > > > >> > This discussion is going to be fun! =D > > > >> > > > > >> > A little more than seventy Wikipedia-projects has more
than
65k
> > > >> articles, > > > >> > the remaining two hundred or so are pretty small. > > > >> > > > > >> > What if a base set of articles were opened for paid
translators?
> > There > > > >> are > > > >> > several lists of such base sets. We have both the
thousand
> articles > > > from > > > >> > "List of articles every Wikipedia should have"[1] and
and
the
ten > > > >> thousand > > > >> > articles from the expanded list[2]. > > > >> > > > > >> > Lets say verified good translators was paid about
$0.01
per
word
> > > (about > > > >> $1 > > > >> > for a 1k-article) for translating one of those
articles
into
> another > > > >> > language, with perhaps a higher pay for contributors
in
high-cost > > > >> > countries. The pay would also have to be higher for
languages
that > > > lacks > > > >> > good translation tools. > > > >> > > > > >> > I believe this would be an _enabling_ activity for the > communities, > > as > > > >> > without a base set of articles it won't be possible to
build a
> > > >> community at > > > >> > all. By not paying for new articles, and only
translating
> > > >> well-referenced > > > >> > articles, some of the disputes in the communities
could
be
> avoided. > > > >> Perhaps > > > >> > we should also identify good source articles, that
would
be
a
> help. > > > >> > Translated articles should be above some minimum size,
but
they
> does > > > not > > > >> > have to be full translations of the source article. > > > >> > > > > >> > A real problem is that our existing lists of good
articles
other
> > > >> projects > > > >> > should have is pretty much biased towards Western
World,
so
they
> > need > > > a > > > >> lot > > > >> > of adjustments. Perhaps such a project would identify
our
inherit > > > bias? > > > >> > > > > >> > [1] > > > >> > https://meta.wikimedia.org/
wiki/List_of_articles_every_
> > > >> > Wikipedia_should_have > > > >> > [2] > > > >> > https://meta.wikimedia.org/
wiki/List_of_articles_every_
> > > >> > Wikipedia_should_have/Expanded > > > >> > _______________________________________________ > > > >> > Wikimedia-l mailing list, guidelines at: > > https://meta.wikimedia.org/ > > > >> > wiki/Mailing_lists/Guidelines and
> > > >> > wiki/Wikimedia-l > > > >> > New messages to: Wikimedia-l@lists.wikimedia.org > > > >> > Unsubscribe: https://lists.wikimedia.org/ > > mailman/listinfo/wikimedia-l > > > , > > > >> > <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> > unsubscribe> > > > >> > > > >> > > > >> > > > >> > > > >> -- > > > >> James Heilman > > > >> MD, CCFP-EM, Wikipedian > > > >> _______________________________________________ > > > >> Wikimedia-l mailing list, guidelines at: > > https://meta.wikimedia.org/wik > > > >> i/Mailing_lists/Guidelines and
https://meta.wikimedia.org/wik
> > > >> i/Wikimedia-l > > > >> New messages to: Wikimedia-l@lists.wikimedia.org > > > >> Unsubscribe: https://lists.wikimedia.org/ > mailman/listinfo/wikimedia-l > > , > > > >> <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> unsubscribe> > > > >> > > > > > > > > > > > _______________________________________________ > > > Wikimedia-l mailing list, guidelines at: > > > https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
and
> > > https://meta.wikimedia.org/wiki/Wikimedia-l > > > New messages to: Wikimedia-l@lists.wikimedia.org > > > Unsubscribe: https://lists.wikimedia.org/ mailman/listinfo/wikimedia-l, > > > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> > _______________________________________________ > > Wikimedia-l mailing list, guidelines at:
> > wiki/Mailing_lists/Guidelines and
> > wiki/Wikimedia-l > > New messages to: Wikimedia-l@lists.wikimedia.org > > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> > > _______________________________________________ > Wikimedia-l mailing list, guidelines at:
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ > wiki/Wikimedia-l > New messages to: Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
>
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Hoi, I have been involved in a translation project with professional translators translating featured articles of the English Wikipedia. The choice for featured articles was done because we expected that the content would not be in dispute. We found different. Several of the translated articles were not accepted.. one of them was about World War II.
I have also toyed with the idea of content that is not available in the language of a Wikipedia (including English). Translation is one solution an other solution is generating basic information from the data available at Wikidata. The benefit is not only to our readers; they will at least be informed up to a point and another benefit will be the quality of the Wikipedia involved. One problem that will be fixed is the one of false friends, when red links are linked to Wikidata, the information provided will always be implicitly correct. Another possibility is to provide the text of a sister Wikipedia.
We can do a better job by providing the sum of all knowledge that is available to us. Thanks, GerardM
On 25 February 2018 at 15:16, John Erling Blad jeblad@gmail.com wrote:
Sorry, but this does not make sense. The core articles apply globally. There will although be articles in additions to a list of core articles, but I don't try to advocate any of those lists as the one and only list. Actually I have toyed with an idea of automatically create a list of core articles, and that would identify important articles no matter if they are from a big western language or a minority language.
The main problem is NOT that minority languages should have articles about the major cities and important philosophers, *the main problem is that minor languages can't get started because they lack content*!
On Sun, Feb 25, 2018 at 2:41 AM, Vi to vituzzu.wiki@gmail.com wrote:
Cultural appropriation is something different, by "forcing" the contents
in
a minority language we would actually be at risk of implementing a form
of
"cultural colonialism" which is the opposite of a cultural appropriation.
NOTE: I refer to "the Western" in both cultural and "Wikipedian" sense: I mean cultures with a strong presence on the web plus developed and flourishing Wikipedia communities.
Helping minority languages with funds/workforce is not bad in my opinion, but I think a bottom-up process must be followed, with the "bottom" being as closer as possible to relevant linguistic/cultural communities. A Wikipedia full of "what the Westerns think is important" in a minority non-Western language would definitely fail project scopes.
This kind of problem almost does not arise with minority language associated to Western cultures since they share the same cultural backgrounds: back to my previous example the cultural background of Sicilian is substantially equal to Italian one. Still, as I already
wrote,
wikis in minority languages should focus on a certain aspect of wiki
scope:
Wiki has roughly two main scopes: 1) sharing knowledge in a certain language 2) also preserving the cultural heritage associated with
different
languages. For languages mainly spoken as first language the "sharing knowledge" aspect is predominant, while the second should take precedence in languages whose speakers are native speakers of a "bigger" language.
Vito
2018-02-24 22:58 GMT+01:00 John Erling Blad jeblad@gmail.com:
Seems like this is mostly about cultural ownership and appropriation.
Not
sure if it is possible to agree on this.
On Sat, Feb 24, 2018 at 6:08 PM, Vi to vituzzu.wiki@gmail.com wrote:
I'll reply to the most recent email just for laziness.
I'm doubtful for a series of reasons, most of were already expressed
in a
better way by others: *a remuneration in terms of quantity will weaken the quality of translations unless there's a strong mechanism of quality
verification
requiring a quantity of resources comparable to translations
themselves;
*articles are the result of a long process which reflects cultural
identity
of different communities, I'm not confident with transferring them
to a
different "weaker" cultures. My usage of "weaker" adjective only
focuses
about the strength of a cultural presence on the Internet; *articles to be translated are at high risk of reflecting the
cultural
identity (and biases) of the Western culture; *finally I think paid translators would hardly turn into stable Wikipedians.
IMHO some paid editing may be better exploited in order to digitalise
texts
of unrepresented cultures (wikisource) or preserving their
vocabularies
(wiktionary).
Also those languages which are secondary for all their speakers
should
be
dealt with in a different fashion. I, for one, am a native speaker of specific variant of Sicilian, Sicilian is a secondary language to any
of
its speakers. Honestly, I'd find pointless to read the biography of Leonardo da Vinci in Sicilian while I can find thousands of books
about
him
in Italian. Also I find this kind of translation creates a fake
"literary"
language totally detached from reality: there's no "encaustic
painting"
in
Sicilian, still a Sicilian article about Leonardo will invent one.
As a general principle we should always collect, rather than create, knowledge.
Vito
2018-02-24 16:30 GMT+01:00 John Erling Blad jeblad@gmail.com:
My reply can be read as a bit more harsh than intended, it was
merely a
statement about my present experience about translators in general.
The problem with lack of contributors (and translators) in a
specialized
area is that there is a small community, and within this community
some
kind of selection is made. Each time a selection is repeated the
remaining
group shrinks. Specialize the selection sufficiently many times and
there
will be no contributors (or translators) left. It is simply a game
of
probabilities. Thus, to make such a project work it must have a sufficiently broad scope for the articles. Articles about public
health
services will probably work even for a pretty small language group,
but
specialized medical articles might create a problem. But then you
find
a retired orthopedic surgeon like Subas Chandra Rout…
On Sat, Feb 24, 2018 at 4:04 PM, James Heilman jmh649@gmail.com
wrote:
I agree with John that it is very difficult to turn a translator
into a
new
editor. I also agree with Jean-Philippe that it is key to have
involvement
of the local projects and preferable if they lead the efforts. Of
the
languages we worked in only one explicitly requested not to be
involved /
have translations from TWB.
James
On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad <
jeblad@gmail.com>
wrote:
> You can turn it around; give added credits for translations
from
small
> language projects and into the larger ones, that is a lot more interesting > than strictly translating from the larger language projects. > > On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < > jpbeland@wikimedia.ca > > wrote: > > > I think the request for such projects should come from the
concerned
> > language projects, same for the list of articles. If not, in
my
simple
> > opinion, it is a form of coloniasm again. > > > > Jean-Philippe Béland > > Vice President, Wikimedia Canada > > > > > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <
jeblad@gmail.com
> wrote: > > > > > Should have added that the remaining points are somewhat
less
> interesting > > > in this context. Preloading a set of articles is a bad
idea,
the
> > > translators should be able to chose for themselves.
Articles
should
> also > > be > > > pretty broad, not very narrow technical or medical, ie
vertical
> articles, > > > as the number of editors that can handle those will be
pretty
small.
> > > > > > In particular: Do not believe you can turn a teanslator
into
a
new
> > editor! > > > You can although turn an existing editor into a translator. > > > > > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
jeblad@gmail.com>
> > > wrote: > > > > > > > 1) You must start with high quality content and thus all
articles
are > > > >> extensively improved before being proposed for
translation.
> > > > > > > > > > > > Note that to much pressure on "quality" can easily kill
the
project. > > > > > > > > 3) The "Content Translation" tool developed by the WMF
made
efforts
> > more > > > >> efficient than handing around word documents. Would love
to
see
that > > > tool > > > >> improved further such as having it support specific
lists
of
> articles > > > that > > > >> are deemed ready for translation by certain groups.
Would
also
love > > the > > > >> tool to have tracking metrics for these types of
projects.
> > > > > > > > > > > > Didn't mention ContentTranslation, but it should be
pretty
obvious.
> > > > > > > > 4) We used volunteer translators mostly associated with
our
partner
> > > >> Translators Without Borders. One issue we found was that
languages
> in > > > >> which > > > >> their are lots of translators such as French, Spanish,
and
Italian
> > there > > > >> is > > > >> often already at least some content on many of the
topics
in
> question. > > > The > > > >> issue than becomes integration which needs an expert
Wikipedia.
And > > for > > > >> languages in which we have little content there are
often
few
> > avaliable > > > >> volunteers. > > > > > > > > > > > > I used projects below 65k articles as an example, as the
chance
of
> > > > competing articles are pretty low. > > > > > > > > 5) With respect to "paying per word" the problem is this
would
> require > > > >> significant checks and balances to make sure people are
taking
the
> > work > > > >> seriously and not simple using Google translate for the
70
or
so
> > > languages > > > >> in which it claims to work. We often had translations
undergo
a
> second > > > >> review and the volunteers at TWB have to pass certain
tests
to
be
> > > >> accepted. > > > > > > > > > > > > I'n my original email I wrote "verified good
translators".
It
is
as
> > > > simple as "Has the editor contributed other articles at
the
project?" > > > > > > > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <
jmh649@gmail.com
> > wrote: > > > > > > > >> We learned a few things during the medical translation
project
which > > > >> started back in 2011: > > > >> > > > >> 1) You must start with high quality content and thus all
articles
> are > > > >> extensively improved before being proposed for
translation.
> > > >> > > > >> 2) A lot of languages want "less" content than is
present
on
EN
WP. > > Thus > > > >> we > > > >> moved to just improving and suggesting for translation
the
leads
of > > the > > > >> English articles. > > > >> > > > >> 3) The "Content Translation" tool developed by the WMF
made
efforts > > more > > > >> efficient than handing around word documents. Would love
to
see
that > > > tool > > > >> improved further such as having it support specific
lists
of
> articles > > > that > > > >> are deemed ready for translation by certain groups.
Would
also
love > > the > > > >> tool to have tracking metrics for these types of
projects.
> > > >> > > > >> 4) We used volunteer translators mostly associated with
our
partner > > > >> Translators Without Borders. One issue we found was that
languages
> in > > > >> which > > > >> their are lots of translators such as French, Spanish,
and
Italian
> > there > > > >> is > > > >> often already at least some content on many of the
topics
in
> question. > > > The > > > >> issue than becomes integration which needs an expert
Wikipedia.
And > > for > > > >> languages in which we have little content there are
often
few
> > avaliable > > > >> volunteers. > > > >> > > > >> 5) With respect to "paying per word" the problem is this
would
> require > > > >> significant checks and balances to make sure people are
taking
the
> > work > > > >> seriously and not simple using Google translate for the
70
or
so
> > > languages > > > >> in which it claims to work. We often had translations
undergo
a
> second > > > >> review and the volunteers at TWB have to pass certain
tests
to
be
> > > >> accepted. > > > >> > > > >> 6) I hired a coordinator for the translation project
for a
couple
of > > > >> years. > > > >> The translators at TWB did not want to become
Wikipedians
or
learn
> how > > > to > > > >> use our systems. The coordinator created account like
TransSW001
> (one > > > for > > > >> each volunteer) and preloaded the article to be
translated
into
> > Content > > > >> Translation. They than gave the volunteer translator the
user
name
> and > > > >> password to the account. > > > >> > > > >> 7) Were are we at now? There are currently just over
1,000
leads
of > > > >> articles that have been improved and are ready for
translation.
This > > > >> includes articles on the 440 medications that are on the
WHO
> Essential > > > >> List. We have worked a bit in some 100 languages. The
efforts
have
> > > >> resulted > > > >> in more than 5 million works translated and integrated
into
> different > > > >> Wikipedias. The coordinator has unfortunately moved on
to
his
real
> job > > > of > > > >> teaching high school students. > > > >> > > > >> 8) The project continues but at a slower pace than
before.
The
> > > Wikipedian > > > >> and retired orthopedic surgeon Subas Chandra Rout has
basically
> single > > > >> handedly translated nearly all 1,000 leads into Odia a
language
> spoken > > > by > > > >> 40 million people in Eastern India. The amazing thing is
that
for
> many > > > of > > > >> these topics this is the first and only information
online
about
it. > > > >> Google > > > >> translate does not even claim to work in this language.
Our
> > partnerships > > > >> with WMTW and medical school in Taipai continue to
translate
into
> > > Chinese. > > > >> There the students translate and than their translations
are
> reviewed > > by > > > >> their profs before being posted. They translate in
groups
using
> > hackpad > > > to > > > >> make it more social. > > > >> > > > >> I am currently working to re invigorate the project :-) > > > >> James > > > >> > > > >> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad < jeblad@gmail.com > > > > > >> wrote: > > > >> > > > >> > This discussion is going to be fun! =D > > > >> > > > > >> > A little more than seventy Wikipedia-projects has more
than
65k
> > > >> articles, > > > >> > the remaining two hundred or so are pretty small. > > > >> > > > > >> > What if a base set of articles were opened for paid
translators?
> > There > > > >> are > > > >> > several lists of such base sets. We have both the
thousand
> articles > > > from > > > >> > "List of articles every Wikipedia should have"[1] and
and
the
ten > > > >> thousand > > > >> > articles from the expanded list[2]. > > > >> > > > > >> > Lets say verified good translators was paid about
$0.01
per
word
> > > (about > > > >> $1 > > > >> > for a 1k-article) for translating one of those
articles
into
> another > > > >> > language, with perhaps a higher pay for contributors
in
high-cost > > > >> > countries. The pay would also have to be higher for
languages
that > > > lacks > > > >> > good translation tools. > > > >> > > > > >> > I believe this would be an _enabling_ activity for the > communities, > > as > > > >> > without a base set of articles it won't be possible to
build a
> > > >> community at > > > >> > all. By not paying for new articles, and only
translating
> > > >> well-referenced > > > >> > articles, some of the disputes in the communities
could
be
> avoided. > > > >> Perhaps > > > >> > we should also identify good source articles, that
would
be
a
> help. > > > >> > Translated articles should be above some minimum size,
but
they
> does > > > not > > > >> > have to be full translations of the source article. > > > >> > > > > >> > A real problem is that our existing lists of good
articles
other
> > > >> projects > > > >> > should have is pretty much biased towards Western
World,
so
they
> > need > > > a > > > >> lot > > > >> > of adjustments. Perhaps such a project would identify
our
inherit > > > bias? > > > >> > > > > >> > [1] > > > >> > https://meta.wikimedia.org/
wiki/List_of_articles_every_
> > > >> > Wikipedia_should_have > > > >> > [2] > > > >> > https://meta.wikimedia.org/
wiki/List_of_articles_every_
> > > >> > Wikipedia_should_have/Expanded > > > >> > _______________________________________________ > > > >> > Wikimedia-l mailing list, guidelines at: > > https://meta.wikimedia.org/ > > > >> > wiki/Mailing_lists/Guidelines and
> > > >> > wiki/Wikimedia-l > > > >> > New messages to: Wikimedia-l@lists.wikimedia.org > > > >> > Unsubscribe: https://lists.wikimedia.org/ > > mailman/listinfo/wikimedia-l > > > , > > > >> > <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> > unsubscribe> > > > >> > > > >> > > > >> > > > >> > > > >> -- > > > >> James Heilman > > > >> MD, CCFP-EM, Wikipedian > > > >> _______________________________________________ > > > >> Wikimedia-l mailing list, guidelines at: > > https://meta.wikimedia.org/wik > > > >> i/Mailing_lists/Guidelines and
https://meta.wikimedia.org/wik
> > > >> i/Wikimedia-l > > > >> New messages to: Wikimedia-l@lists.wikimedia.org > > > >> Unsubscribe: https://lists.wikimedia.org/ > mailman/listinfo/wikimedia-l > > , > > > >> <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> unsubscribe> > > > >> > > > > > > > > > > > _______________________________________________ > > > Wikimedia-l mailing list, guidelines at: > > > https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
and
> > > https://meta.wikimedia.org/wiki/Wikimedia-l > > > New messages to: Wikimedia-l@lists.wikimedia.org > > > Unsubscribe: https://lists.wikimedia.org/ mailman/listinfo/wikimedia-l, > > > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> > _______________________________________________ > > Wikimedia-l mailing list, guidelines at:
> > wiki/Mailing_lists/Guidelines and
> > wiki/Wikimedia-l > > New messages to: Wikimedia-l@lists.wikimedia.org > > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> > > _______________________________________________ > Wikimedia-l mailing list, guidelines at:
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ > wiki/Wikimedia-l > New messages to: Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
>
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
WWII is not an universal truth. If some small country claim the Nazis was the good guys, then they are simply wrong.
Yes there are a lot of projects where information diverge, but usually that is because someone added material that somehow seems more appropriate for readers in that specific language. Although sometimes the content is really wrong, and that happen on all projects.
On Mon, Feb 26, 2018 at 12:51 PM, Gerard Meijssen <gerard.meijssen@gmail.com
wrote:
Hoi, I have been involved in a translation project with professional translators translating featured articles of the English Wikipedia. The choice for featured articles was done because we expected that the content would not be in dispute. We found different. Several of the translated articles were not accepted.. one of them was about World War II.
I have also toyed with the idea of content that is not available in the language of a Wikipedia (including English). Translation is one solution an other solution is generating basic information from the data available at Wikidata. The benefit is not only to our readers; they will at least be informed up to a point and another benefit will be the quality of the Wikipedia involved. One problem that will be fixed is the one of false friends, when red links are linked to Wikidata, the information provided will always be implicitly correct. Another possibility is to provide the text of a sister Wikipedia.
We can do a better job by providing the sum of all knowledge that is available to us. Thanks, GerardM
On 25 February 2018 at 15:16, John Erling Blad jeblad@gmail.com wrote:
Sorry, but this does not make sense. The core articles apply globally. There will although be articles in additions to a list of core articles, but I don't try to advocate any of those lists as the one and only list. Actually I have toyed with an idea of automatically create a list of core articles, and that would identify important articles no matter if they
are
from a big western language or a minority language.
The main problem is NOT that minority languages should have articles
about
the major cities and important philosophers, *the main problem is that minor languages can't get started because they lack content*!
On Sun, Feb 25, 2018 at 2:41 AM, Vi to vituzzu.wiki@gmail.com wrote:
Cultural appropriation is something different, by "forcing" the
contents
in
a minority language we would actually be at risk of implementing a form
of
"cultural colonialism" which is the opposite of a cultural
appropriation.
NOTE: I refer to "the Western" in both cultural and "Wikipedian"
sense: I
mean cultures with a strong presence on the web plus developed and flourishing Wikipedia communities.
Helping minority languages with funds/workforce is not bad in my
opinion,
but I think a bottom-up process must be followed, with the "bottom"
being
as closer as possible to relevant linguistic/cultural communities. A Wikipedia full of "what the Westerns think is important" in a minority non-Western language would definitely fail project scopes.
This kind of problem almost does not arise with minority language associated to Western cultures since they share the same cultural backgrounds: back to my previous example the cultural background of Sicilian is substantially equal to Italian one. Still, as I already
wrote,
wikis in minority languages should focus on a certain aspect of wiki
scope:
Wiki has roughly two main scopes: 1) sharing knowledge in a certain language 2) also preserving the cultural heritage associated with
different
languages. For languages mainly spoken as first language the "sharing knowledge" aspect is predominant, while the second should take
precedence
in languages whose speakers are native speakers of a "bigger" language.
Vito
2018-02-24 22:58 GMT+01:00 John Erling Blad jeblad@gmail.com:
Seems like this is mostly about cultural ownership and appropriation.
Not
sure if it is possible to agree on this.
On Sat, Feb 24, 2018 at 6:08 PM, Vi to vituzzu.wiki@gmail.com
wrote:
I'll reply to the most recent email just for laziness.
I'm doubtful for a series of reasons, most of were already
expressed
in a
better way by others: *a remuneration in terms of quantity will weaken the quality of translations unless there's a strong mechanism of quality
verification
requiring a quantity of resources comparable to translations
themselves;
*articles are the result of a long process which reflects cultural
identity
of different communities, I'm not confident with transferring them
to a
different "weaker" cultures. My usage of "weaker" adjective only
focuses
about the strength of a cultural presence on the Internet; *articles to be translated are at high risk of reflecting the
cultural
identity (and biases) of the Western culture; *finally I think paid translators would hardly turn into stable Wikipedians.
IMHO some paid editing may be better exploited in order to
digitalise
texts
of unrepresented cultures (wikisource) or preserving their
vocabularies
(wiktionary).
Also those languages which are secondary for all their speakers
should
be
dealt with in a different fashion. I, for one, am a native speaker
of
specific variant of Sicilian, Sicilian is a secondary language to
any
of
its speakers. Honestly, I'd find pointless to read the biography of Leonardo da Vinci in Sicilian while I can find thousands of books
about
him
in Italian. Also I find this kind of translation creates a fake
"literary"
language totally detached from reality: there's no "encaustic
painting"
in
Sicilian, still a Sicilian article about Leonardo will invent one.
As a general principle we should always collect, rather than
create,
knowledge.
Vito
2018-02-24 16:30 GMT+01:00 John Erling Blad jeblad@gmail.com:
My reply can be read as a bit more harsh than intended, it was
merely a
statement about my present experience about translators in
general.
The problem with lack of contributors (and translators) in a
specialized
area is that there is a small community, and within this
community
some
kind of selection is made. Each time a selection is repeated the
remaining
group shrinks. Specialize the selection sufficiently many times
and
there
will be no contributors (or translators) left. It is simply a
game
of
probabilities. Thus, to make such a project work it must have a sufficiently broad scope for the articles. Articles about public
health
services will probably work even for a pretty small language
group,
but
specialized medical articles might create a problem. But then you
find
a retired orthopedic surgeon like Subas Chandra Rout…
On Sat, Feb 24, 2018 at 4:04 PM, James Heilman <jmh649@gmail.com
wrote:
> I agree with John that it is very difficult to turn a
translator
into a
new > editor. I also agree with Jean-Philippe that it is key to have involvement > of the local projects and preferable if they lead the efforts.
Of
the
> languages we worked in only one explicitly requested not to be
involved /
> have translations from TWB. > > James > > On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad <
jeblad@gmail.com>
> wrote: > > > You can turn it around; give added credits for translations
from
small
> > language projects and into the larger ones, that is a lot
more
> interesting > > than strictly translating from the larger language projects. > > > > On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < > > jpbeland@wikimedia.ca > > > wrote: > > > > > I think the request for such projects should come from the
concerned
> > > language projects, same for the list of articles. If not,
in
my
simple > > > opinion, it is a form of coloniasm again. > > > > > > Jean-Philippe Béland > > > Vice President, Wikimedia Canada > > > > > > > > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <
jeblad@gmail.com
> > wrote: > > > > > > > Should have added that the remaining points are somewhat
less
> > interesting > > > > in this context. Preloading a set of articles is a bad
idea,
the
> > > > translators should be able to chose for themselves.
Articles
should
> > also > > > be > > > > pretty broad, not very narrow technical or medical, ie
vertical
> > articles, > > > > as the number of editors that can handle those will be
pretty
small. > > > > > > > > In particular: Do not believe you can turn a teanslator
into
a
new
> > > editor! > > > > You can although turn an existing editor into a
translator.
> > > > > > > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad < jeblad@gmail.com> > > > > wrote: > > > > > > > > > 1) You must start with high quality content and thus
all
articles
> are > > > > >> extensively improved before being proposed for
translation.
> > > > > > > > > > > > > > > Note that to much pressure on "quality" can easily kill
the
> project. > > > > > > > > > > 3) The "Content Translation" tool developed by the WMF
made
efforts > > > more > > > > >> efficient than handing around word documents. Would
love
to
see
> that > > > > tool > > > > >> improved further such as having it support specific
lists
of
> > articles > > > > that > > > > >> are deemed ready for translation by certain groups.
Would
also
> love > > > the > > > > >> tool to have tracking metrics for these types of
projects.
> > > > > > > > > > > > > > > Didn't mention ContentTranslation, but it should be
pretty
obvious. > > > > > > > > > > 4) We used volunteer translators mostly associated with
our
partner > > > > >> Translators Without Borders. One issue we found was
that
languages > > in > > > > >> which > > > > >> their are lots of translators such as French, Spanish,
and
Italian > > > there > > > > >> is > > > > >> often already at least some content on many of the
topics
in
> > question. > > > > The > > > > >> issue than becomes integration which needs an expert
Wikipedia.
> And > > > for > > > > >> languages in which we have little content there are
often
few
> > > avaliable > > > > >> volunteers. > > > > > > > > > > > > > > > I used projects below 65k articles as an example, as
the
chance
of > > > > > competing articles are pretty low. > > > > > > > > > > 5) With respect to "paying per word" the problem is
this
would
> > require > > > > >> significant checks and balances to make sure people
are
taking
the > > > work > > > > >> seriously and not simple using Google translate for
the
70
or
so
> > > > languages > > > > >> in which it claims to work. We often had translations
undergo
a
> > second > > > > >> review and the volunteers at TWB have to pass certain
tests
to
be > > > > >> accepted. > > > > > > > > > > > > > > > I'n my original email I wrote "verified good
translators".
It
is
as > > > > > simple as "Has the editor contributed other articles at
the
> project?" > > > > > > > > > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <
jmh649@gmail.com
> > > > wrote: > > > > > > > > > >> We learned a few things during the medical translation
project
> which > > > > >> started back in 2011: > > > > >> > > > > >> 1) You must start with high quality content and thus
all
articles > > are > > > > >> extensively improved before being proposed for
translation.
> > > > >> > > > > >> 2) A lot of languages want "less" content than is
present
on
EN
> WP. > > > Thus > > > > >> we > > > > >> moved to just improving and suggesting for translation
the
leads
> of > > > the > > > > >> English articles. > > > > >> > > > > >> 3) The "Content Translation" tool developed by the WMF
made
> efforts > > > more > > > > >> efficient than handing around word documents. Would
love
to
see
> that > > > > tool > > > > >> improved further such as having it support specific
lists
of
> > articles > > > > that > > > > >> are deemed ready for translation by certain groups.
Would
also
> love > > > the > > > > >> tool to have tracking metrics for these types of
projects.
> > > > >> > > > > >> 4) We used volunteer translators mostly associated
with
our
> partner > > > > >> Translators Without Borders. One issue we found was
that
languages > > in > > > > >> which > > > > >> their are lots of translators such as French, Spanish,
and
Italian > > > there > > > > >> is > > > > >> often already at least some content on many of the
topics
in
> > question. > > > > The > > > > >> issue than becomes integration which needs an expert
Wikipedia.
> And > > > for > > > > >> languages in which we have little content there are
often
few
> > > avaliable > > > > >> volunteers. > > > > >> > > > > >> 5) With respect to "paying per word" the problem is
this
would
> > require > > > > >> significant checks and balances to make sure people
are
taking
the > > > work > > > > >> seriously and not simple using Google translate for
the
70
or
so
> > > > languages > > > > >> in which it claims to work. We often had translations
undergo
a
> > second > > > > >> review and the volunteers at TWB have to pass certain
tests
to
be > > > > >> accepted. > > > > >> > > > > >> 6) I hired a coordinator for the translation project
for a
couple > of > > > > >> years. > > > > >> The translators at TWB did not want to become
Wikipedians
or
learn > > how > > > > to > > > > >> use our systems. The coordinator created account like
TransSW001
> > (one > > > > for > > > > >> each volunteer) and preloaded the article to be
translated
into
> > > Content > > > > >> Translation. They than gave the volunteer translator
the
user
name > > and > > > > >> password to the account. > > > > >> > > > > >> 7) Were are we at now? There are currently just over
1,000
leads
> of > > > > >> articles that have been improved and are ready for
translation.
> This > > > > >> includes articles on the 440 medications that are on
the
WHO
> > Essential > > > > >> List. We have worked a bit in some 100 languages. The
efforts
have > > > > >> resulted > > > > >> in more than 5 million works translated and integrated
into
> > different > > > > >> Wikipedias. The coordinator has unfortunately moved on
to
his
real > > job > > > > of > > > > >> teaching high school students. > > > > >> > > > > >> 8) The project continues but at a slower pace than
before.
The
> > > > Wikipedian > > > > >> and retired orthopedic surgeon Subas Chandra Rout has
basically
> > single > > > > >> handedly translated nearly all 1,000 leads into Odia a
language
> > spoken > > > > by > > > > >> 40 million people in Eastern India. The amazing thing
is
that
for > > many > > > > of > > > > >> these topics this is the first and only information
online
about
> it. > > > > >> Google > > > > >> translate does not even claim to work in this
language.
Our
> > > partnerships > > > > >> with WMTW and medical school in Taipai continue to
translate
into > > > > Chinese. > > > > >> There the students translate and than their
translations
are
> > reviewed > > > by > > > > >> their profs before being posted. They translate in
groups
using
> > > hackpad > > > > to > > > > >> make it more social. > > > > >> > > > > >> I am currently working to re invigorate the project
:-)
> > > > >> James > > > > >> > > > > >> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad < > jeblad@gmail.com > > > > > > > >> wrote: > > > > >> > > > > >> > This discussion is going to be fun! =D > > > > >> > > > > > >> > A little more than seventy Wikipedia-projects has
more
than
65k > > > > >> articles, > > > > >> > the remaining two hundred or so are pretty small. > > > > >> > > > > > >> > What if a base set of articles were opened for paid translators? > > > There > > > > >> are > > > > >> > several lists of such base sets. We have both the
thousand
> > articles > > > > from > > > > >> > "List of articles every Wikipedia should have"[1]
and
and
the
> ten > > > > >> thousand > > > > >> > articles from the expanded list[2]. > > > > >> > > > > > >> > Lets say verified good translators was paid about
$0.01
per
word > > > > (about > > > > >> $1 > > > > >> > for a 1k-article) for translating one of those
articles
into
> > another > > > > >> > language, with perhaps a higher pay for contributors
in
> high-cost > > > > >> > countries. The pay would also have to be higher for
languages
> that > > > > lacks > > > > >> > good translation tools. > > > > >> > > > > > >> > I believe this would be an _enabling_ activity for
the
> > communities, > > > as > > > > >> > without a base set of articles it won't be possible
to
build a
> > > > >> community at > > > > >> > all. By not paying for new articles, and only
translating
> > > > >> well-referenced > > > > >> > articles, some of the disputes in the communities
could
be
> > avoided. > > > > >> Perhaps > > > > >> > we should also identify good source articles, that
would
be
a
> > help. > > > > >> > Translated articles should be above some minimum
size,
but
they > > does > > > > not > > > > >> > have to be full translations of the source article. > > > > >> > > > > > >> > A real problem is that our existing lists of good
articles
other > > > > >> projects > > > > >> > should have is pretty much biased towards Western
World,
so
they > > > need > > > > a > > > > >> lot > > > > >> > of adjustments. Perhaps such a project would
identify
our
> inherit > > > > bias? > > > > >> > > > > > >> > [1] > > > > >> > https://meta.wikimedia.org/
wiki/List_of_articles_every_
> > > > >> > Wikipedia_should_have > > > > >> > [2] > > > > >> > https://meta.wikimedia.org/
wiki/List_of_articles_every_
> > > > >> > Wikipedia_should_have/Expanded > > > > >> > _______________________________________________ > > > > >> > Wikimedia-l mailing list, guidelines at: > > > https://meta.wikimedia.org/ > > > > >> > wiki/Mailing_lists/Guidelines and
> > > > >> > wiki/Wikimedia-l > > > > >> > New messages to: Wikimedia-l@lists.wikimedia.org > > > > >> > Unsubscribe: https://lists.wikimedia.org/ > > > mailman/listinfo/wikimedia-l > > > > , > > > > >> > <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> > > unsubscribe> > > > > >> > > > > >> > > > > >> > > > > >> > > > > >> -- > > > > >> James Heilman > > > > >> MD, CCFP-EM, Wikipedian > > > > >> _______________________________________________ > > > > >> Wikimedia-l mailing list, guidelines at: > > > https://meta.wikimedia.org/wik > > > > >> i/Mailing_lists/Guidelines and
https://meta.wikimedia.org/wik
> > > > >> i/Wikimedia-l > > > > >> New messages to: Wikimedia-l@lists.wikimedia.org > > > > >> Unsubscribe: https://lists.wikimedia.org/ > > mailman/listinfo/wikimedia-l > > > , > > > > >> <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> > unsubscribe> > > > > >> > > > > > > > > > > > > > > _______________________________________________ > > > > Wikimedia-l mailing list, guidelines at: > > > > https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
and
> > > > https://meta.wikimedia.org/wiki/Wikimedia-l > > > > New messages to: Wikimedia-l@lists.wikimedia.org > > > > Unsubscribe: https://lists.wikimedia.org/ > mailman/listinfo/wikimedia-l, > > > > mailto:wikimedia-l-request@lists.wikimedia.org?subject= unsubscribe > > > _______________________________________________ > > > Wikimedia-l mailing list, guidelines at:
> > > wiki/Mailing_lists/Guidelines and
> > > wiki/Wikimedia-l > > > New messages to: Wikimedia-l@lists.wikimedia.org > > > Unsubscribe: https://lists.wikimedia.org/ mailman/listinfo/wikimedia-l, > > > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> > > > > _______________________________________________ > > Wikimedia-l mailing list, guidelines at:
> > wiki/Mailing_lists/Guidelines and
> > wiki/Wikimedia-l > > New messages to: Wikimedia-l@lists.wikimedia.org > > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> > > > > > -- > James Heilman > MD, CCFP-EM, Wikipedian > _______________________________________________ > Wikimedia-l mailing list, guidelines at:
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ > wiki/Wikimedia-l > New messages to: Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> _______________________________________________ Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Indeed. We can all agree that it's OK for a lot of reason to have differences in content between projects. What these differences are is a separate discussion.
These differences often come up when discussing translation projects in Wikipedia, and it's important to recognize them, but it's also important not to treat them as a blocker or to let them be too much of a distraction.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore
2018-02-27 11:40 GMT+02:00 John Erling Blad jeblad@gmail.com:
WWII is not an universal truth. If some small country claim the Nazis was the good guys, then they are simply wrong.
Yes there are a lot of projects where information diverge, but usually that is because someone added material that somehow seems more appropriate for readers in that specific language. Although sometimes the content is really wrong, and that happen on all projects.
On Mon, Feb 26, 2018 at 12:51 PM, Gerard Meijssen < gerard.meijssen@gmail.com
wrote:
Hoi, I have been involved in a translation project with professional
translators
translating featured articles of the English Wikipedia. The choice for featured articles was done because we expected that the content would not be in dispute. We found different. Several of the translated articles
were
not accepted.. one of them was about World War II.
I have also toyed with the idea of content that is not available in the language of a Wikipedia (including English). Translation is one solution
an
other solution is generating basic information from the data available at Wikidata. The benefit is not only to our readers; they will at least be informed up to a point and another benefit will be the quality of the Wikipedia involved. One problem that will be fixed is the one of false friends, when red links are linked to Wikidata, the information provided will always be implicitly correct. Another possibility is to provide the text of a sister Wikipedia.
We can do a better job by providing the sum of all knowledge that is available to us. Thanks, GerardM
On 25 February 2018 at 15:16, John Erling Blad jeblad@gmail.com wrote:
Sorry, but this does not make sense. The core articles apply globally. There will although be articles in additions to a list of core
articles,
but I don't try to advocate any of those lists as the one and only
list.
Actually I have toyed with an idea of automatically create a list of
core
articles, and that would identify important articles no matter if they
are
from a big western language or a minority language.
The main problem is NOT that minority languages should have articles
about
the major cities and important philosophers, *the main problem is that minor languages can't get started because they lack content*!
On Sun, Feb 25, 2018 at 2:41 AM, Vi to vituzzu.wiki@gmail.com wrote:
Cultural appropriation is something different, by "forcing" the
contents
in
a minority language we would actually be at risk of implementing a
form
of
"cultural colonialism" which is the opposite of a cultural
appropriation.
NOTE: I refer to "the Western" in both cultural and "Wikipedian"
sense: I
mean cultures with a strong presence on the web plus developed and flourishing Wikipedia communities.
Helping minority languages with funds/workforce is not bad in my
opinion,
but I think a bottom-up process must be followed, with the "bottom"
being
as closer as possible to relevant linguistic/cultural communities. A Wikipedia full of "what the Westerns think is important" in a
minority
non-Western language would definitely fail project scopes.
This kind of problem almost does not arise with minority language associated to Western cultures since they share the same cultural backgrounds: back to my previous example the cultural background of Sicilian is substantially equal to Italian one. Still, as I already
wrote,
wikis in minority languages should focus on a certain aspect of wiki
scope:
Wiki has roughly two main scopes: 1) sharing knowledge in a certain language 2) also preserving the cultural heritage associated with
different
languages. For languages mainly spoken as first language the "sharing knowledge" aspect is predominant, while the second should take
precedence
in languages whose speakers are native speakers of a "bigger"
language.
Vito
2018-02-24 22:58 GMT+01:00 John Erling Blad jeblad@gmail.com:
Seems like this is mostly about cultural ownership and
appropriation.
Not
sure if it is possible to agree on this.
On Sat, Feb 24, 2018 at 6:08 PM, Vi to vituzzu.wiki@gmail.com
wrote:
I'll reply to the most recent email just for laziness.
I'm doubtful for a series of reasons, most of were already
expressed
in a
better way by others: *a remuneration in terms of quantity will weaken the quality of translations unless there's a strong mechanism of quality
verification
requiring a quantity of resources comparable to translations
themselves;
*articles are the result of a long process which reflects
cultural
identity
of different communities, I'm not confident with transferring
them
to a
different "weaker" cultures. My usage of "weaker" adjective only
focuses
about the strength of a cultural presence on the Internet; *articles to be translated are at high risk of reflecting the
cultural
identity (and biases) of the Western culture; *finally I think paid translators would hardly turn into stable Wikipedians.
IMHO some paid editing may be better exploited in order to
digitalise
texts
of unrepresented cultures (wikisource) or preserving their
vocabularies
(wiktionary).
Also those languages which are secondary for all their speakers
should
be
dealt with in a different fashion. I, for one, am a native
speaker
of
specific variant of Sicilian, Sicilian is a secondary language to
any
of
its speakers. Honestly, I'd find pointless to read the biography
of
Leonardo da Vinci in Sicilian while I can find thousands of books
about
him
in Italian. Also I find this kind of translation creates a fake
"literary"
language totally detached from reality: there's no "encaustic
painting"
in
Sicilian, still a Sicilian article about Leonardo will invent
one.
As a general principle we should always collect, rather than
create,
knowledge.
Vito
2018-02-24 16:30 GMT+01:00 John Erling Blad jeblad@gmail.com:
> My reply can be read as a bit more harsh than intended, it was
merely a
> statement about my present experience about translators in
general.
> > The problem with lack of contributors (and translators) in a
specialized
> area is that there is a small community, and within this
community
some
> kind of selection is made. Each time a selection is repeated
the
remaining > group shrinks. Specialize the selection sufficiently many times
and
there
> will be no contributors (or translators) left. It is simply a
game
of
> probabilities. Thus, to make such a project work it must have a > sufficiently broad scope for the articles. Articles about
public
health
> services will probably work even for a pretty small language
group,
but
> specialized medical articles might create a problem. But then
you
find
> a retired > orthopedic surgeon like Subas Chandra Rout… > > On Sat, Feb 24, 2018 at 4:04 PM, James Heilman <
jmh649@gmail.com
wrote:
> > > I agree with John that it is very difficult to turn a
translator
into a
> new > > editor. I also agree with Jean-Philippe that it is key to
have
> involvement > > of the local projects and preferable if they lead the
efforts.
Of
the
> > languages we worked in only one explicitly requested not to
be
involved / > > have translations from TWB. > > > > James > > > > On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad <
jeblad@gmail.com>
> > wrote: > > > > > You can turn it around; give added credits for translations
from
small > > > language projects and into the larger ones, that is a lot
more
> > interesting > > > than strictly translating from the larger language
projects.
> > > > > > On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < > > > jpbeland@wikimedia.ca > > > > wrote: > > > > > > > I think the request for such projects should come from
the
concerned > > > > language projects, same for the list of articles. If not,
in
my
> simple > > > > opinion, it is a form of coloniasm again. > > > > > > > > Jean-Philippe Béland > > > > Vice President, Wikimedia Canada > > > > > > > > > > > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <
jeblad@gmail.com
> > > > wrote: > > > > > > > > > Should have added that the remaining points are
somewhat
less
> > > interesting > > > > > in this context. Preloading a set of articles is a bad
idea,
the
> > > > > translators should be able to chose for themselves.
Articles
should > > > also > > > > be > > > > > pretty broad, not very narrow technical or medical, ie
vertical
> > > articles, > > > > > as the number of editors that can handle those will be
pretty
> small. > > > > > > > > > > In particular: Do not believe you can turn a teanslator
into
a
new > > > > editor! > > > > > You can although turn an existing editor into a
translator.
> > > > > > > > > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad < > jeblad@gmail.com> > > > > > wrote: > > > > > > > > > > > 1) You must start with high quality content and thus
all
articles > > are > > > > > >> extensively improved before being proposed for
translation.
> > > > > > > > > > > > > > > > > > Note that to much pressure on "quality" can easily
kill
the
> > project. > > > > > > > > > > > > 3) The "Content Translation" tool developed by the
WMF
made
> efforts > > > > more > > > > > >> efficient than handing around word documents. Would
love
to
see > > that > > > > > tool > > > > > >> improved further such as having it support specific
lists
of
> > > articles > > > > > that > > > > > >> are deemed ready for translation by certain groups.
Would
also
> > love > > > > the > > > > > >> tool to have tracking metrics for these types of
projects.
> > > > > > > > > > > > > > > > > > Didn't mention ContentTranslation, but it should be
pretty
> obvious. > > > > > > > > > > > > 4) We used volunteer translators mostly associated
with
our
> partner > > > > > >> Translators Without Borders. One issue we found was
that
> languages > > > in > > > > > >> which > > > > > >> their are lots of translators such as French,
Spanish,
and
> Italian > > > > there > > > > > >> is > > > > > >> often already at least some content on many of the
topics
in
> > > question. > > > > > The > > > > > >> issue than becomes integration which needs an expert Wikipedia. > > And > > > > for > > > > > >> languages in which we have little content there are
often
few
> > > > avaliable > > > > > >> volunteers. > > > > > > > > > > > > > > > > > > I used projects below 65k articles as an example, as
the
chance
> of > > > > > > competing articles are pretty low. > > > > > > > > > > > > 5) With respect to "paying per word" the problem is
this
would
> > > require > > > > > >> significant checks and balances to make sure people
are
taking
> the > > > > work > > > > > >> seriously and not simple using Google translate for
the
70
or
so > > > > > languages > > > > > >> in which it claims to work. We often had
translations
undergo
a > > > second > > > > > >> review and the volunteers at TWB have to pass
certain
tests
to
> be > > > > > >> accepted. > > > > > > > > > > > > > > > > > > I'n my original email I wrote "verified good
translators".
It
is > as > > > > > > simple as "Has the editor contributed other articles
at
the
> > project?" > > > > > > > > > > > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman < jmh649@gmail.com > > > > > > wrote: > > > > > > > > > > > >> We learned a few things during the medical
translation
project
> > which > > > > > >> started back in 2011: > > > > > >> > > > > > >> 1) You must start with high quality content and thus
all
> articles > > > are > > > > > >> extensively improved before being proposed for
translation.
> > > > > >> > > > > > >> 2) A lot of languages want "less" content than is
present
on
EN > > WP. > > > > Thus > > > > > >> we > > > > > >> moved to just improving and suggesting for
translation
the
leads > > of > > > > the > > > > > >> English articles. > > > > > >> > > > > > >> 3) The "Content Translation" tool developed by the
WMF
made
> > efforts > > > > more > > > > > >> efficient than handing around word documents. Would
love
to
see > > that > > > > > tool > > > > > >> improved further such as having it support specific
lists
of
> > > articles > > > > > that > > > > > >> are deemed ready for translation by certain groups.
Would
also
> > love > > > > the > > > > > >> tool to have tracking metrics for these types of
projects.
> > > > > >> > > > > > >> 4) We used volunteer translators mostly associated
with
our
> > partner > > > > > >> Translators Without Borders. One issue we found was
that
> languages > > > in > > > > > >> which > > > > > >> their are lots of translators such as French,
Spanish,
and
> Italian > > > > there > > > > > >> is > > > > > >> often already at least some content on many of the
topics
in
> > > question. > > > > > The > > > > > >> issue than becomes integration which needs an expert Wikipedia. > > And > > > > for > > > > > >> languages in which we have little content there are
often
few
> > > > avaliable > > > > > >> volunteers. > > > > > >> > > > > > >> 5) With respect to "paying per word" the problem is
this
would
> > > require > > > > > >> significant checks and balances to make sure people
are
taking
> the > > > > work > > > > > >> seriously and not simple using Google translate for
the
70
or
so > > > > > languages > > > > > >> in which it claims to work. We often had
translations
undergo
a > > > second > > > > > >> review and the volunteers at TWB have to pass
certain
tests
to
> be > > > > > >> accepted. > > > > > >> > > > > > >> 6) I hired a coordinator for the translation project
for a
> couple > > of > > > > > >> years. > > > > > >> The translators at TWB did not want to become
Wikipedians
or
> learn > > > how > > > > > to > > > > > >> use our systems. The coordinator created account
like
TransSW001 > > > (one > > > > > for > > > > > >> each volunteer) and preloaded the article to be
translated
into > > > > Content > > > > > >> Translation. They than gave the volunteer translator
the
user
> name > > > and > > > > > >> password to the account. > > > > > >> > > > > > >> 7) Were are we at now? There are currently just over
1,000
leads > > of > > > > > >> articles that have been improved and are ready for translation. > > This > > > > > >> includes articles on the 440 medications that are on
the
WHO
> > > Essential > > > > > >> List. We have worked a bit in some 100 languages.
The
efforts
> have > > > > > >> resulted > > > > > >> in more than 5 million works translated and
integrated
into
> > > different > > > > > >> Wikipedias. The coordinator has unfortunately moved
on
to
his
> real > > > job > > > > > of > > > > > >> teaching high school students. > > > > > >> > > > > > >> 8) The project continues but at a slower pace than
before.
The
> > > > > Wikipedian > > > > > >> and retired orthopedic surgeon Subas Chandra Rout
has
basically > > > single > > > > > >> handedly translated nearly all 1,000 leads into
Odia a
language > > > spoken > > > > > by > > > > > >> 40 million people in Eastern India. The amazing
thing
is
that
> for > > > many > > > > > of > > > > > >> these topics this is the first and only information
online
about > > it. > > > > > >> Google > > > > > >> translate does not even claim to work in this
language.
Our
> > > > partnerships > > > > > >> with WMTW and medical school in Taipai continue to
translate
> into > > > > > Chinese. > > > > > >> There the students translate and than their
translations
are
> > > reviewed > > > > by > > > > > >> their profs before being posted. They translate in
groups
using > > > > hackpad > > > > > to > > > > > >> make it more social. > > > > > >> > > > > > >> I am currently working to re invigorate the project
:-)
> > > > > >> James > > > > > >> > > > > > >> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad < > > jeblad@gmail.com > > > > > > > > > >> wrote: > > > > > >> > > > > > >> > This discussion is going to be fun! =D > > > > > >> > > > > > > >> > A little more than seventy Wikipedia-projects has
more
than
> 65k > > > > > >> articles, > > > > > >> > the remaining two hundred or so are pretty small. > > > > > >> > > > > > > >> > What if a base set of articles were opened for
paid
> translators? > > > > There > > > > > >> are > > > > > >> > several lists of such base sets. We have both the
thousand
> > > articles > > > > > from > > > > > >> > "List of articles every Wikipedia should have"[1]
and
and
the > > ten > > > > > >> thousand > > > > > >> > articles from the expanded list[2]. > > > > > >> > > > > > > >> > Lets say verified good translators was paid about
$0.01
per
> word > > > > > (about > > > > > >> $1 > > > > > >> > for a 1k-article) for translating one of those
articles
into
> > > another > > > > > >> > language, with perhaps a higher pay for
contributors
in
> > high-cost > > > > > >> > countries. The pay would also have to be higher
for
languages > > that > > > > > lacks > > > > > >> > good translation tools. > > > > > >> > > > > > > >> > I believe this would be an _enabling_ activity for
the
> > > communities, > > > > as > > > > > >> > without a base set of articles it won't be
possible
to
build a > > > > > >> community at > > > > > >> > all. By not paying for new articles, and only
translating
> > > > > >> well-referenced > > > > > >> > articles, some of the disputes in the communities
could
be
> > > avoided. > > > > > >> Perhaps > > > > > >> > we should also identify good source articles, that
would
be
a > > > help. > > > > > >> > Translated articles should be above some minimum
size,
but
> they > > > does > > > > > not > > > > > >> > have to be full translations of the source
article.
> > > > > >> > > > > > > >> > A real problem is that our existing lists of good
articles
> other > > > > > >> projects > > > > > >> > should have is pretty much biased towards Western
World,
so
> they > > > > need > > > > > a > > > > > >> lot > > > > > >> > of adjustments. Perhaps such a project would
identify
our
> > inherit > > > > > bias? > > > > > >> > > > > > > >> > [1] > > > > > >> > https://meta.wikimedia.org/
wiki/List_of_articles_every_
> > > > > >> > Wikipedia_should_have > > > > > >> > [2] > > > > > >> > https://meta.wikimedia.org/
wiki/List_of_articles_every_
> > > > > >> > Wikipedia_should_have/Expanded > > > > > >> > _______________________________________________ > > > > > >> > Wikimedia-l mailing list, guidelines at: > > > > https://meta.wikimedia.org/ > > > > > >> > wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ > > > > > >> > wiki/Wikimedia-l > > > > > >> > New messages to: Wikimedia-l@lists.wikimedia.org > > > > > >> > Unsubscribe: https://lists.wikimedia.org/ > > > > mailman/listinfo/wikimedia-l > > > > > , > > > > > >> > <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> > > > unsubscribe> > > > > > >> > > > > > >> > > > > > >> > > > > > >> > > > > > >> -- > > > > > >> James Heilman > > > > > >> MD, CCFP-EM, Wikipedian > > > > > >> _______________________________________________ > > > > > >> Wikimedia-l mailing list, guidelines at: > > > > https://meta.wikimedia.org/wik > > > > > >> i/Mailing_lists/Guidelines and
https://meta.wikimedia.org/wik
> > > > > >> i/Wikimedia-l > > > > > >> New messages to: Wikimedia-l@lists.wikimedia.org > > > > > >> Unsubscribe: https://lists.wikimedia.org/ > > > mailman/listinfo/wikimedia-l > > > > , > > > > > >> <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> > > unsubscribe> > > > > > >> > > > > > > > > > > > > > > > > > _______________________________________________ > > > > > Wikimedia-l mailing list, guidelines at: > > > > > https://meta.wikimedia.org/
wiki/Mailing_lists/Guidelines
and
> > > > > https://meta.wikimedia.org/wiki/Wikimedia-l > > > > > New messages to: Wikimedia-l@lists.wikimedia.org > > > > > Unsubscribe: https://lists.wikimedia.org/ > > mailman/listinfo/wikimedia-l, > > > > > <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> unsubscribe> > > > > _______________________________________________ > > > > Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ > > > > wiki/Mailing_lists/Guidelines and
> > > > wiki/Wikimedia-l > > > > New messages to: Wikimedia-l@lists.wikimedia.org > > > > Unsubscribe: https://lists.wikimedia.org/ > mailman/listinfo/wikimedia-l, > > > > mailto:wikimedia-l-request@lists.wikimedia.org?subject= unsubscribe > > > > > > > _______________________________________________ > > > Wikimedia-l mailing list, guidelines at:
> > > wiki/Mailing_lists/Guidelines and
> > > wiki/Wikimedia-l > > > New messages to: Wikimedia-l@lists.wikimedia.org > > > Unsubscribe: https://lists.wikimedia.org/ mailman/listinfo/wikimedia-l, > > > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> > > > > > > > > > > -- > > James Heilman > > MD, CCFP-EM, Wikipedian > > _______________________________________________ > > Wikimedia-l mailing list, guidelines at:
> > wiki/Mailing_lists/Guidelines and
> > wiki/Wikimedia-l > > New messages to: Wikimedia-l@lists.wikimedia.org > > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> > > _______________________________________________ > Wikimedia-l mailing list, guidelines at:
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ > wiki/Wikimedia-l > New messages to: Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> _______________________________________________ Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
WWII is not an universal truth. If some small country claim the Nazis was the good guys, then they are simply wrong.
No even thats not entirely true for some countries WWII in Europe was foot note, for others WWII was the trigger for escaping colonial rule. Languages related to individual cultures do have different perspectives on events even on en.wp some FA can tend to have bias to US/UK perspective on events. There is no one truth for history its all about perspectives, about the significance of differing events, and the impact those events had.
Even when it comes to less disputed topics like biota there can differences, take Kangaroo there is referred to an Aboriginal Australian word but in reality there are over 300 different Australian Languages and each has their own name for a kangaroo. They each also have different knowledge and information simply because of the different environmental conditions.
Paid translations is not the ideal format, it even has flaws if money is to be spent then making tools and support projects that enable translations. Translations risk being interpreted at paternalism with a colonial language deciding how an indigenous language should speak about a subject.
On 27 February 2018 at 17:40, John Erling Blad jeblad@gmail.com wrote:
WWII is not an universal truth. If some small country claim the Nazis was the good guys, then they are simply wrong.
Yes there are a lot of projects where information diverge, but usually that is because someone added material that somehow seems more appropriate for readers in that specific language. Although sometimes the content is really wrong, and that happen on all projects.
On Mon, Feb 26, 2018 at 12:51 PM, Gerard Meijssen < gerard.meijssen@gmail.com
wrote:
Hoi, I have been involved in a translation project with professional
translators
translating featured articles of the English Wikipedia. The choice for featured articles was done because we expected that the content would not be in dispute. We found different. Several of the translated articles
were
not accepted.. one of them was about World War II.
I have also toyed with the idea of content that is not available in the language of a Wikipedia (including English). Translation is one solution
an
other solution is generating basic information from the data available at Wikidata. The benefit is not only to our readers; they will at least be informed up to a point and another benefit will be the quality of the Wikipedia involved. One problem that will be fixed is the one of false friends, when red links are linked to Wikidata, the information provided will always be implicitly correct. Another possibility is to provide the text of a sister Wikipedia.
We can do a better job by providing the sum of all knowledge that is available to us. Thanks, GerardM
On 25 February 2018 at 15:16, John Erling Blad jeblad@gmail.com wrote:
Sorry, but this does not make sense. The core articles apply globally. There will although be articles in additions to a list of core
articles,
but I don't try to advocate any of those lists as the one and only
list.
Actually I have toyed with an idea of automatically create a list of
core
articles, and that would identify important articles no matter if they
are
from a big western language or a minority language.
The main problem is NOT that minority languages should have articles
about
the major cities and important philosophers, *the main problem is that minor languages can't get started because they lack content*!
On Sun, Feb 25, 2018 at 2:41 AM, Vi to vituzzu.wiki@gmail.com wrote:
Cultural appropriation is something different, by "forcing" the
contents
in
a minority language we would actually be at risk of implementing a
form
of
"cultural colonialism" which is the opposite of a cultural
appropriation.
NOTE: I refer to "the Western" in both cultural and "Wikipedian"
sense: I
mean cultures with a strong presence on the web plus developed and flourishing Wikipedia communities.
Helping minority languages with funds/workforce is not bad in my
opinion,
but I think a bottom-up process must be followed, with the "bottom"
being
as closer as possible to relevant linguistic/cultural communities. A Wikipedia full of "what the Westerns think is important" in a
minority
non-Western language would definitely fail project scopes.
This kind of problem almost does not arise with minority language associated to Western cultures since they share the same cultural backgrounds: back to my previous example the cultural background of Sicilian is substantially equal to Italian one. Still, as I already
wrote,
wikis in minority languages should focus on a certain aspect of wiki
scope:
Wiki has roughly two main scopes: 1) sharing knowledge in a certain language 2) also preserving the cultural heritage associated with
different
languages. For languages mainly spoken as first language the "sharing knowledge" aspect is predominant, while the second should take
precedence
in languages whose speakers are native speakers of a "bigger"
language.
Vito
2018-02-24 22:58 GMT+01:00 John Erling Blad jeblad@gmail.com:
Seems like this is mostly about cultural ownership and
appropriation.
Not
sure if it is possible to agree on this.
On Sat, Feb 24, 2018 at 6:08 PM, Vi to vituzzu.wiki@gmail.com
wrote:
I'll reply to the most recent email just for laziness.
I'm doubtful for a series of reasons, most of were already
expressed
in a
better way by others: *a remuneration in terms of quantity will weaken the quality of translations unless there's a strong mechanism of quality
verification
requiring a quantity of resources comparable to translations
themselves;
*articles are the result of a long process which reflects
cultural
identity
of different communities, I'm not confident with transferring
them
to a
different "weaker" cultures. My usage of "weaker" adjective only
focuses
about the strength of a cultural presence on the Internet; *articles to be translated are at high risk of reflecting the
cultural
identity (and biases) of the Western culture; *finally I think paid translators would hardly turn into stable Wikipedians.
IMHO some paid editing may be better exploited in order to
digitalise
texts
of unrepresented cultures (wikisource) or preserving their
vocabularies
(wiktionary).
Also those languages which are secondary for all their speakers
should
be
dealt with in a different fashion. I, for one, am a native
speaker
of
specific variant of Sicilian, Sicilian is a secondary language to
any
of
its speakers. Honestly, I'd find pointless to read the biography
of
Leonardo da Vinci in Sicilian while I can find thousands of books
about
him
in Italian. Also I find this kind of translation creates a fake
"literary"
language totally detached from reality: there's no "encaustic
painting"
in
Sicilian, still a Sicilian article about Leonardo will invent
one.
As a general principle we should always collect, rather than
create,
knowledge.
Vito
2018-02-24 16:30 GMT+01:00 John Erling Blad jeblad@gmail.com:
> My reply can be read as a bit more harsh than intended, it was
merely a
> statement about my present experience about translators in
general.
> > The problem with lack of contributors (and translators) in a
specialized
> area is that there is a small community, and within this
community
some
> kind of selection is made. Each time a selection is repeated
the
remaining > group shrinks. Specialize the selection sufficiently many times
and
there
> will be no contributors (or translators) left. It is simply a
game
of
> probabilities. Thus, to make such a project work it must have a > sufficiently broad scope for the articles. Articles about
public
health
> services will probably work even for a pretty small language
group,
but
> specialized medical articles might create a problem. But then
you
find
> a retired > orthopedic surgeon like Subas Chandra Rout… > > On Sat, Feb 24, 2018 at 4:04 PM, James Heilman <
jmh649@gmail.com
wrote:
> > > I agree with John that it is very difficult to turn a
translator
into a
> new > > editor. I also agree with Jean-Philippe that it is key to
have
> involvement > > of the local projects and preferable if they lead the
efforts.
Of
the
> > languages we worked in only one explicitly requested not to
be
involved / > > have translations from TWB. > > > > James > > > > On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad <
jeblad@gmail.com>
> > wrote: > > > > > You can turn it around; give added credits for translations
from
small > > > language projects and into the larger ones, that is a lot
more
> > interesting > > > than strictly translating from the larger language
projects.
> > > > > > On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < > > > jpbeland@wikimedia.ca > > > > wrote: > > > > > > > I think the request for such projects should come from
the
concerned > > > > language projects, same for the list of articles. If not,
in
my
> simple > > > > opinion, it is a form of coloniasm again. > > > > > > > > Jean-Philippe Béland > > > > Vice President, Wikimedia Canada > > > > > > > > > > > > On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <
jeblad@gmail.com
> > > > wrote: > > > > > > > > > Should have added that the remaining points are
somewhat
less
> > > interesting > > > > > in this context. Preloading a set of articles is a bad
idea,
the
> > > > > translators should be able to chose for themselves.
Articles
should > > > also > > > > be > > > > > pretty broad, not very narrow technical or medical, ie
vertical
> > > articles, > > > > > as the number of editors that can handle those will be
pretty
> small. > > > > > > > > > > In particular: Do not believe you can turn a teanslator
into
a
new > > > > editor! > > > > > You can although turn an existing editor into a
translator.
> > > > > > > > > > On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad < > jeblad@gmail.com> > > > > > wrote: > > > > > > > > > > > 1) You must start with high quality content and thus
all
articles > > are > > > > > >> extensively improved before being proposed for
translation.
> > > > > > > > > > > > > > > > > > Note that to much pressure on "quality" can easily
kill
the
> > project. > > > > > > > > > > > > 3) The "Content Translation" tool developed by the
WMF
made
> efforts > > > > more > > > > > >> efficient than handing around word documents. Would
love
to
see > > that > > > > > tool > > > > > >> improved further such as having it support specific
lists
of
> > > articles > > > > > that > > > > > >> are deemed ready for translation by certain groups.
Would
also
> > love > > > > the > > > > > >> tool to have tracking metrics for these types of
projects.
> > > > > > > > > > > > > > > > > > Didn't mention ContentTranslation, but it should be
pretty
> obvious. > > > > > > > > > > > > 4) We used volunteer translators mostly associated
with
our
> partner > > > > > >> Translators Without Borders. One issue we found was
that
> languages > > > in > > > > > >> which > > > > > >> their are lots of translators such as French,
Spanish,
and
> Italian > > > > there > > > > > >> is > > > > > >> often already at least some content on many of the
topics
in
> > > question. > > > > > The > > > > > >> issue than becomes integration which needs an expert Wikipedia. > > And > > > > for > > > > > >> languages in which we have little content there are
often
few
> > > > avaliable > > > > > >> volunteers. > > > > > > > > > > > > > > > > > > I used projects below 65k articles as an example, as
the
chance
> of > > > > > > competing articles are pretty low. > > > > > > > > > > > > 5) With respect to "paying per word" the problem is
this
would
> > > require > > > > > >> significant checks and balances to make sure people
are
taking
> the > > > > work > > > > > >> seriously and not simple using Google translate for
the
70
or
so > > > > > languages > > > > > >> in which it claims to work. We often had
translations
undergo
a > > > second > > > > > >> review and the volunteers at TWB have to pass
certain
tests
to
> be > > > > > >> accepted. > > > > > > > > > > > > > > > > > > I'n my original email I wrote "verified good
translators".
It
is > as > > > > > > simple as "Has the editor contributed other articles
at
the
> > project?" > > > > > > > > > > > > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman < jmh649@gmail.com > > > > > > wrote: > > > > > > > > > > > >> We learned a few things during the medical
translation
project
> > which > > > > > >> started back in 2011: > > > > > >> > > > > > >> 1) You must start with high quality content and thus
all
> articles > > > are > > > > > >> extensively improved before being proposed for
translation.
> > > > > >> > > > > > >> 2) A lot of languages want "less" content than is
present
on
EN > > WP. > > > > Thus > > > > > >> we > > > > > >> moved to just improving and suggesting for
translation
the
leads > > of > > > > the > > > > > >> English articles. > > > > > >> > > > > > >> 3) The "Content Translation" tool developed by the
WMF
made
> > efforts > > > > more > > > > > >> efficient than handing around word documents. Would
love
to
see > > that > > > > > tool > > > > > >> improved further such as having it support specific
lists
of
> > > articles > > > > > that > > > > > >> are deemed ready for translation by certain groups.
Would
also
> > love > > > > the > > > > > >> tool to have tracking metrics for these types of
projects.
> > > > > >> > > > > > >> 4) We used volunteer translators mostly associated
with
our
> > partner > > > > > >> Translators Without Borders. One issue we found was
that
> languages > > > in > > > > > >> which > > > > > >> their are lots of translators such as French,
Spanish,
and
> Italian > > > > there > > > > > >> is > > > > > >> often already at least some content on many of the
topics
in
> > > question. > > > > > The > > > > > >> issue than becomes integration which needs an expert Wikipedia. > > And > > > > for > > > > > >> languages in which we have little content there are
often
few
> > > > avaliable > > > > > >> volunteers. > > > > > >> > > > > > >> 5) With respect to "paying per word" the problem is
this
would
> > > require > > > > > >> significant checks and balances to make sure people
are
taking
> the > > > > work > > > > > >> seriously and not simple using Google translate for
the
70
or
so > > > > > languages > > > > > >> in which it claims to work. We often had
translations
undergo
a > > > second > > > > > >> review and the volunteers at TWB have to pass
certain
tests
to
> be > > > > > >> accepted. > > > > > >> > > > > > >> 6) I hired a coordinator for the translation project
for a
> couple > > of > > > > > >> years. > > > > > >> The translators at TWB did not want to become
Wikipedians
or
> learn > > > how > > > > > to > > > > > >> use our systems. The coordinator created account
like
TransSW001 > > > (one > > > > > for > > > > > >> each volunteer) and preloaded the article to be
translated
into > > > > Content > > > > > >> Translation. They than gave the volunteer translator
the
user
> name > > > and > > > > > >> password to the account. > > > > > >> > > > > > >> 7) Were are we at now? There are currently just over
1,000
leads > > of > > > > > >> articles that have been improved and are ready for translation. > > This > > > > > >> includes articles on the 440 medications that are on
the
WHO
> > > Essential > > > > > >> List. We have worked a bit in some 100 languages.
The
efforts
> have > > > > > >> resulted > > > > > >> in more than 5 million works translated and
integrated
into
> > > different > > > > > >> Wikipedias. The coordinator has unfortunately moved
on
to
his
> real > > > job > > > > > of > > > > > >> teaching high school students. > > > > > >> > > > > > >> 8) The project continues but at a slower pace than
before.
The
> > > > > Wikipedian > > > > > >> and retired orthopedic surgeon Subas Chandra Rout
has
basically > > > single > > > > > >> handedly translated nearly all 1,000 leads into
Odia a
language > > > spoken > > > > > by > > > > > >> 40 million people in Eastern India. The amazing
thing
is
that
> for > > > many > > > > > of > > > > > >> these topics this is the first and only information
online
about > > it. > > > > > >> Google > > > > > >> translate does not even claim to work in this
language.
Our
> > > > partnerships > > > > > >> with WMTW and medical school in Taipai continue to
translate
> into > > > > > Chinese. > > > > > >> There the students translate and than their
translations
are
> > > reviewed > > > > by > > > > > >> their profs before being posted. They translate in
groups
using > > > > hackpad > > > > > to > > > > > >> make it more social. > > > > > >> > > > > > >> I am currently working to re invigorate the project
:-)
> > > > > >> James > > > > > >> > > > > > >> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad < > > jeblad@gmail.com > > > > > > > > > >> wrote: > > > > > >> > > > > > >> > This discussion is going to be fun! =D > > > > > >> > > > > > > >> > A little more than seventy Wikipedia-projects has
more
than
> 65k > > > > > >> articles, > > > > > >> > the remaining two hundred or so are pretty small. > > > > > >> > > > > > > >> > What if a base set of articles were opened for
paid
> translators? > > > > There > > > > > >> are > > > > > >> > several lists of such base sets. We have both the
thousand
> > > articles > > > > > from > > > > > >> > "List of articles every Wikipedia should have"[1]
and
and
the > > ten > > > > > >> thousand > > > > > >> > articles from the expanded list[2]. > > > > > >> > > > > > > >> > Lets say verified good translators was paid about
$0.01
per
> word > > > > > (about > > > > > >> $1 > > > > > >> > for a 1k-article) for translating one of those
articles
into
> > > another > > > > > >> > language, with perhaps a higher pay for
contributors
in
> > high-cost > > > > > >> > countries. The pay would also have to be higher
for
languages > > that > > > > > lacks > > > > > >> > good translation tools. > > > > > >> > > > > > > >> > I believe this would be an _enabling_ activity for
the
> > > communities, > > > > as > > > > > >> > without a base set of articles it won't be
possible
to
build a > > > > > >> community at > > > > > >> > all. By not paying for new articles, and only
translating
> > > > > >> well-referenced > > > > > >> > articles, some of the disputes in the communities
could
be
> > > avoided. > > > > > >> Perhaps > > > > > >> > we should also identify good source articles, that
would
be
a > > > help. > > > > > >> > Translated articles should be above some minimum
size,
but
> they > > > does > > > > > not > > > > > >> > have to be full translations of the source
article.
> > > > > >> > > > > > > >> > A real problem is that our existing lists of good
articles
> other > > > > > >> projects > > > > > >> > should have is pretty much biased towards Western
World,
so
> they > > > > need > > > > > a > > > > > >> lot > > > > > >> > of adjustments. Perhaps such a project would
identify
our
> > inherit > > > > > bias? > > > > > >> > > > > > > >> > [1] > > > > > >> > https://meta.wikimedia.org/
wiki/List_of_articles_every_
> > > > > >> > Wikipedia_should_have > > > > > >> > [2] > > > > > >> > https://meta.wikimedia.org/
wiki/List_of_articles_every_
> > > > > >> > Wikipedia_should_have/Expanded > > > > > >> > _______________________________________________ > > > > > >> > Wikimedia-l mailing list, guidelines at: > > > > https://meta.wikimedia.org/ > > > > > >> > wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ > > > > > >> > wiki/Wikimedia-l > > > > > >> > New messages to: Wikimedia-l@lists.wikimedia.org > > > > > >> > Unsubscribe: https://lists.wikimedia.org/ > > > > mailman/listinfo/wikimedia-l > > > > > , > > > > > >> > <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> > > > unsubscribe> > > > > > >> > > > > > >> > > > > > >> > > > > > >> > > > > > >> -- > > > > > >> James Heilman > > > > > >> MD, CCFP-EM, Wikipedian > > > > > >> _______________________________________________ > > > > > >> Wikimedia-l mailing list, guidelines at: > > > > https://meta.wikimedia.org/wik > > > > > >> i/Mailing_lists/Guidelines and
https://meta.wikimedia.org/wik
> > > > > >> i/Wikimedia-l > > > > > >> New messages to: Wikimedia-l@lists.wikimedia.org > > > > > >> Unsubscribe: https://lists.wikimedia.org/ > > > mailman/listinfo/wikimedia-l > > > > , > > > > > >> <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> > > unsubscribe> > > > > > >> > > > > > > > > > > > > > > > > > _______________________________________________ > > > > > Wikimedia-l mailing list, guidelines at: > > > > > https://meta.wikimedia.org/
wiki/Mailing_lists/Guidelines
and
> > > > > https://meta.wikimedia.org/wiki/Wikimedia-l > > > > > New messages to: Wikimedia-l@lists.wikimedia.org > > > > > Unsubscribe: https://lists.wikimedia.org/ > > mailman/listinfo/wikimedia-l, > > > > > <mailto:wikimedia-l-request@lists.wikimedia.org
?subject=
> unsubscribe> > > > > _______________________________________________ > > > > Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ > > > > wiki/Mailing_lists/Guidelines and
> > > > wiki/Wikimedia-l > > > > New messages to: Wikimedia-l@lists.wikimedia.org > > > > Unsubscribe: https://lists.wikimedia.org/ > mailman/listinfo/wikimedia-l, > > > > mailto:wikimedia-l-request@lists.wikimedia.org?subject= unsubscribe > > > > > > > _______________________________________________ > > > Wikimedia-l mailing list, guidelines at:
> > > wiki/Mailing_lists/Guidelines and
> > > wiki/Wikimedia-l > > > New messages to: Wikimedia-l@lists.wikimedia.org > > > Unsubscribe: https://lists.wikimedia.org/ mailman/listinfo/wikimedia-l, > > > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> > > > > > > > > > > -- > > James Heilman > > MD, CCFP-EM, Wikipedian > > _______________________________________________ > > Wikimedia-l mailing list, guidelines at:
> > wiki/Mailing_lists/Guidelines and
> > wiki/Wikimedia-l > > New messages to: Wikimedia-l@lists.wikimedia.org > > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> > > _______________________________________________ > Wikimedia-l mailing list, guidelines at:
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ > wiki/Wikimedia-l > New messages to: Wikimedia-l@lists.wikimedia.org > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
> _______________________________________________ Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I agree with the last part of Vito's message. For languages where '''all''' the speakers speak another lingua franca, I think such process does not have real value. The speakers will always go read in the bigger language because the article is most likely to be better. The advantages of having their own Wikipedia is to be able to express knowledge in their own way according to their own culture.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 12:09 PM Vi to vituzzu.wiki@gmail.com wrote:
I'll reply to the most recent email just for laziness.
I'm doubtful for a series of reasons, most of were already expressed in a better way by others: *a remuneration in terms of quantity will weaken the quality of translations unless there's a strong mechanism of quality verification requiring a quantity of resources comparable to translations themselves; *articles are the result of a long process which reflects cultural identity of different communities, I'm not confident with transferring them to a different "weaker" cultures. My usage of "weaker" adjective only focuses about the strength of a cultural presence on the Internet; *articles to be translated are at high risk of reflecting the cultural identity (and biases) of the Western culture; *finally I think paid translators would hardly turn into stable Wikipedians.
IMHO some paid editing may be better exploited in order to digitalise texts of unrepresented cultures (wikisource) or preserving their vocabularies (wiktionary).
Also those languages which are secondary for all their speakers should be dealt with in a different fashion. I, for one, am a native speaker of specific variant of Sicilian, Sicilian is a secondary language to any of its speakers. Honestly, I'd find pointless to read the biography of Leonardo da Vinci in Sicilian while I can find thousands of books about him in Italian. Also I find this kind of translation creates a fake "literary" language totally detached from reality: there's no "encaustic painting" in Sicilian, still a Sicilian article about Leonardo will invent one.
As a general principle we should always collect, rather than create, knowledge.
Vito
2018-02-24 16:30 GMT+01:00 John Erling Blad jeblad@gmail.com:
My reply can be read as a bit more harsh than intended, it was merely a statement about my present experience about translators in general.
The problem with lack of contributors (and translators) in a specialized area is that there is a small community, and within this community some kind of selection is made. Each time a selection is repeated the
remaining
group shrinks. Specialize the selection sufficiently many times and there will be no contributors (or translators) left. It is simply a game of probabilities. Thus, to make such a project work it must have a sufficiently broad scope for the articles. Articles about public health services will probably work even for a pretty small language group, but specialized medical articles might create a problem. But then you find a retired orthopedic surgeon like Subas Chandra Rout…
On Sat, Feb 24, 2018 at 4:04 PM, James Heilman jmh649@gmail.com wrote:
I agree with John that it is very difficult to turn a translator into a
new
editor. I also agree with Jean-Philippe that it is key to have
involvement
of the local projects and preferable if they lead the efforts. Of the languages we worked in only one explicitly requested not to be
involved /
have translations from TWB.
James
On Sat, Feb 24, 2018 at 7:59 AM, John Erling Blad jeblad@gmail.com wrote:
You can turn it around; give added credits for translations from
small
language projects and into the larger ones, that is a lot more
interesting
than strictly translating from the larger language projects.
On Sat, Feb 24, 2018 at 3:55 PM, Jean-Philippe Béland < jpbeland@wikimedia.ca
wrote:
I think the request for such projects should come from the
concerned
language projects, same for the list of articles. If not, in my
simple
opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad <jeblad@gmail.com
wrote:
Should have added that the remaining points are somewhat less
interesting
in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles
should
also
be
pretty broad, not very narrow technical or medical, ie vertical
articles,
as the number of editors that can handle those will be pretty
small.
In particular: Do not believe you can turn a teanslator into a
new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
jeblad@gmail.com>
wrote:
> 1) You must start with high quality content and thus all
articles
are
>> extensively improved before being proposed for translation. > > > Note that to much pressure on "quality" can easily kill the
project.
> > 3) The "Content Translation" tool developed by the WMF made
efforts
more
>> efficient than handing around word documents. Would love to
see
that
tool >> improved further such as having it support specific lists of
articles
that >> are deemed ready for translation by certain groups. Would also
love
the
>> tool to have tracking metrics for these types of projects. > > > Didn't mention ContentTranslation, but it should be pretty
obvious.
> > 4) We used volunteer translators mostly associated with our
partner
>> Translators Without Borders. One issue we found was that
languages
in
>> which >> their are lots of translators such as French, Spanish, and
Italian
there
>> is >> often already at least some content on many of the topics in
question.
The >> issue than becomes integration which needs an expert
Wikipedia.
And
for
>> languages in which we have little content there are often few
avaliable
>> volunteers. > > > I used projects below 65k articles as an example, as the chance
of
> competing articles are pretty low. > > 5) With respect to "paying per word" the problem is this would
require
>> significant checks and balances to make sure people are taking
the
work
>> seriously and not simple using Google translate for the 70 or
so
languages >> in which it claims to work. We often had translations undergo
a
second
>> review and the volunteers at TWB have to pass certain tests to
be
>> accepted. > > > I'n my original email I wrote "verified good translators". It
is
as
> simple as "Has the editor contributed other articles at the
project?"
> > On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <
jmh649@gmail.com
wrote:
> >> We learned a few things during the medical translation project
which
>> started back in 2011: >> >> 1) You must start with high quality content and thus all
articles
are
>> extensively improved before being proposed for translation. >> >> 2) A lot of languages want "less" content than is present on
EN
WP.
Thus
>> we >> moved to just improving and suggesting for translation the
leads
of
the
>> English articles. >> >> 3) The "Content Translation" tool developed by the WMF made
efforts
more
>> efficient than handing around word documents. Would love to
see
that
tool >> improved further such as having it support specific lists of
articles
that >> are deemed ready for translation by certain groups. Would also
love
the
>> tool to have tracking metrics for these types of projects. >> >> 4) We used volunteer translators mostly associated with our
partner
>> Translators Without Borders. One issue we found was that
languages
in
>> which >> their are lots of translators such as French, Spanish, and
Italian
there
>> is >> often already at least some content on many of the topics in
question.
The >> issue than becomes integration which needs an expert
Wikipedia.
And
for
>> languages in which we have little content there are often few
avaliable
>> volunteers. >> >> 5) With respect to "paying per word" the problem is this would
require
>> significant checks and balances to make sure people are taking
the
work
>> seriously and not simple using Google translate for the 70 or
so
languages >> in which it claims to work. We often had translations undergo
a
second
>> review and the volunteers at TWB have to pass certain tests to
be
>> accepted. >> >> 6) I hired a coordinator for the translation project for a
couple
of
>> years. >> The translators at TWB did not want to become Wikipedians or
learn
how
to >> use our systems. The coordinator created account like
TransSW001
(one
for >> each volunteer) and preloaded the article to be translated
into
Content
>> Translation. They than gave the volunteer translator the user
name
and
>> password to the account. >> >> 7) Were are we at now? There are currently just over 1,000
leads
of
>> articles that have been improved and are ready for
translation.
This
>> includes articles on the 440 medications that are on the WHO
Essential
>> List. We have worked a bit in some 100 languages. The efforts
have
>> resulted >> in more than 5 million works translated and integrated into
different
>> Wikipedias. The coordinator has unfortunately moved on to his
real
job
of >> teaching high school students. >> >> 8) The project continues but at a slower pace than before. The Wikipedian >> and retired orthopedic surgeon Subas Chandra Rout has
basically
single
>> handedly translated nearly all 1,000 leads into Odia a
language
spoken
by >> 40 million people in Eastern India. The amazing thing is that
for
many
of >> these topics this is the first and only information online
about
it.
>> Google >> translate does not even claim to work in this language. Our
partnerships
>> with WMTW and medical school in Taipai continue to translate
into
Chinese. >> There the students translate and than their translations are
reviewed
by
>> their profs before being posted. They translate in groups
using
hackpad
to >> make it more social. >> >> I am currently working to re invigorate the project :-) >> James >> >> On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad <
jeblad@gmail.com
>> wrote: >> >> > This discussion is going to be fun! =D >> > >> > A little more than seventy Wikipedia-projects has more than
65k
>> articles, >> > the remaining two hundred or so are pretty small. >> > >> > What if a base set of articles were opened for paid
translators?
There
>> are >> > several lists of such base sets. We have both the thousand
articles
from >> > "List of articles every Wikipedia should have"[1] and and
the
ten
>> thousand >> > articles from the expanded list[2]. >> > >> > Lets say verified good translators was paid about $0.01 per
word
(about >> $1 >> > for a 1k-article) for translating one of those articles into
another
>> > language, with perhaps a higher pay for contributors in
high-cost
>> > countries. The pay would also have to be higher for
languages
that
lacks >> > good translation tools. >> > >> > I believe this would be an _enabling_ activity for the
communities,
as
>> > without a base set of articles it won't be possible to
build a
>> community at >> > all. By not paying for new articles, and only translating >> well-referenced >> > articles, some of the disputes in the communities could be
avoided.
>> Perhaps >> > we should also identify good source articles, that would be
a
help.
>> > Translated articles should be above some minimum size, but
they
does
not >> > have to be full translations of the source article. >> > >> > A real problem is that our existing lists of good articles
other
>> projects >> > should have is pretty much biased towards Western World, so
they
need
a >> lot >> > of adjustments. Perhaps such a project would identify our
inherit
bias? >> > >> > [1] >> > https://meta.wikimedia.org/wiki/List_of_articles_every_ >> > Wikipedia_should_have >> > [2] >> > https://meta.wikimedia.org/wiki/List_of_articles_every_ >> > Wikipedia_should_have/Expanded >> > _______________________________________________ >> > Wikimedia-l mailing list, guidelines at:
>> > wiki/Mailing_lists/Guidelines and
>> > wiki/Wikimedia-l >> > New messages to: Wikimedia-l@lists.wikimedia.org >> > Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
, >> > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
>> >> >> >> >> -- >> James Heilman >> MD, CCFP-EM, Wikipedian >> _______________________________________________ >> Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wik
>> i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik >> i/Wikimedia-l >> New messages to: Wikimedia-l@lists.wikimedia.org >> Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l
,
>> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
>> > > _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at:
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org
?subject=unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Le 24/02/2018 à 18:08, Vi to a écrit :
*finally I think paid translators would hardly turn into stable Wikipedians.
I think this misses an important point that is, we don't need the initial translator to turn into a sustaining editor, we need the article to evolve with call to action incentives. And articles which don't exist at all – even as a stub – or don't meet an audience of potential contributors will never catch such an evolving cycle.
2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz < psychoslave@culture-libre.org>:
Le 24/02/2018 à 18:08, Vi to a écrit :
*finally I think paid translators would hardly turn into stable Wikipedians.
I think this misses an important point that is, we don't need the initial
translator to turn into a sustaining editor, we need the article to evolve with call to action incentives. And articles which don't exist at all – even as a stub – or don't meet an audience of potential contributors will never catch such an evolving cycle.
This is one of the issues with what I alluded to in my earlier email in this thread: the privilege that the "big" languages have. It's the privilege of already having other encyclopedias, textbooks, public education, etc., in this language. A lot of languages don't have these things. When you speak a language that has had these things before Wikipedia came along, it's hard to perceive the world like a person who speaks a language that doesn't perceives it.
If you define the purpose of paying somebody to translate as "turning the paid translator" into a sustaining editor, then this is indeed likely to fail.
But if you define the purpose differently, it may succeed. For example, you may define the purpose as one or more of the following: * Demonstrating that it's possible to write an encyclopedia in that language * Creating basic encyclopedic terminology and style in that language * Creating a bunch of basic articles that would appear in interlanguage links in Wikipedias from bigger languages (English, French, etc.) * Creating a bunch of basic articles that would appear in search results from internet search engines
The existence of these things may bring in people who will become volunteer sustaining editors.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore
I see Amir's points, which are pretty reasonable, but I fear this would suit languages with a significant presence on the web.
Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating basic encyclopedic terminology and style in that language", if we want to preserve a language we shouldn't create a thing.
By the way I was wondering my concerns about cultural colonization may be addressed -for wikis which has some contents (let's say at least 1000 articles)- by starting expanding existing articles instead of translating new ones. This would solve the problem of choosing what to translate though would leave problems about the perspective contents are created.
Vito
2018-02-27 12:31 GMT+01:00 Amir E. Aharoni amir.aharoni@mail.huji.ac.il:
2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz < psychoslave@culture-libre.org>:
Le 24/02/2018 à 18:08, Vi to a écrit :
*finally I think paid translators would hardly turn into stable Wikipedians.
I think this misses an important point that is, we don't need the
initial
translator to turn into a sustaining editor, we need the article to
evolve
with call to action incentives. And articles which don't exist at all – even as a stub – or don't meet an audience of potential contributors will never catch such an evolving cycle.
This is one of the issues with what I alluded to in my earlier email in this thread: the privilege that the "big" languages have. It's the privilege of already having other encyclopedias, textbooks, public education, etc., in this language. A lot of languages don't have these things. When you speak a language that has had these things before Wikipedia came along, it's hard to perceive the world like a person who speaks a language that doesn't perceives it.
If you define the purpose of paying somebody to translate as "turning the paid translator" into a sustaining editor, then this is indeed likely to fail.
But if you define the purpose differently, it may succeed. For example, you may define the purpose as one or more of the following:
- Demonstrating that it's possible to write an encyclopedia in that
language
- Creating basic encyclopedic terminology and style in that language
- Creating a bunch of basic articles that would appear in interlanguage
links in Wikipedias from bigger languages (English, French, etc.)
- Creating a bunch of basic articles that would appear in search results
from internet search engines
The existence of these things may bring in people who will become volunteer sustaining editors.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
2018-02-27 13:42 GMT+02:00 Vi to vituzzu.wiki@gmail.com:
I see Amir's points, which are pretty reasonable, but I fear this would suit languages with a significant presence on the web.
Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating basic encyclopedic terminology and style in that language", if we want to preserve a language we shouldn't create a thing.
By the way I was wondering my concerns about cultural colonization may be addressed -for wikis which has some contents (let's say at least 1000 articles)- by starting expanding existing articles instead of translating new ones. This would solve the problem of choosing what to translate though would leave problems about the perspective contents are created.
Vito
It's a very common mistake to think that the purpose of Wikipedias in "small" languages is language preservation or revitalization.
Sometimes it is, but there is something much bigger: There are many languages that 1. are alive in speech (and possibly in writing) 2. are not in danger of extinction 3. have a large number of monolingual speakers (let's say 100,000+)
If there is no substantial Wikipedia in such a language, these people can't read Wikipedia in *any language* because they are monolingual. Most likely they cannot read any any encyclopedia in any language. They need a Wikipedia not in order to preserve the language, but to have access to *any* encyclopedic knowledge.
I speak a revitalized language, and I'm very well aware of its history. Language preservation and revitalization are lovely things. But it's not the main point of what Wikimedia does.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore
"Amir E. Aharoni" amir.aharoni@mail.huji.ac.il wrote:
[…]
Sometimes it is, but there is something much bigger: There are many languages that
- are alive in speech (and possibly in writing)
- are not in danger of extinction
- have a large number of monolingual speakers (let's say 100,000+)
If there is no substantial Wikipedia in such a language, these people can't read Wikipedia in *any language* because they are monolingual. Most likely they cannot read any any encyclopedia in any language. They need a Wikipedia not in order to preserve the language, but to have access to *any* encyclopedic knowledge.
I speak a revitalized language, and I'm very well aware of its history. Language preservation and revitalization are lovely things. But it's not the main point of what Wikimedia does.
"Need a Wikipedia" sounds like a great idea when you are selling Wikipedias, but for progress, betterment of humani- ty, sustainable development, etc. I think teaching those monolingual speakers a second language (for example English) is far preferable as it not only enables them to access to a few hundred or thousand articles someone paid to have trans- lated, but all articles of the English Wikipedia, plus every English article, every English book, every English blog, ev- ery English video on the InterNet.
It also grows them not only intellectually, but also removes economical barriers for trading with other groups.
Tim
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore
2018-02-27 18:04 GMT+02:00 Tim Landscheidt tim@tim-landscheidt.de:
"Amir E. Aharoni" amir.aharoni@mail.huji.ac.il wrote:
[…]
Sometimes it is, but there is something much bigger: There are many languages that
- are alive in speech (and possibly in writing)
- are not in danger of extinction
- have a large number of monolingual speakers (let's say 100,000+)
If there is no substantial Wikipedia in such a language, these people
can't
read Wikipedia in *any language* because they are monolingual. Most
likely
they cannot read any any encyclopedia in any language. They need a Wikipedia not in order to preserve the language, but to have access to *any* encyclopedic knowledge.
I speak a revitalized language, and I'm very well aware of its history. Language preservation and revitalization are lovely things. But it's not the main point of what Wikimedia does.
"Need a Wikipedia" sounds like a great idea when you are selling Wikipedias, but for progress, betterment of humani- ty, sustainable development, etc. I think teaching those monolingual speakers a second language (for example English) is far preferable as it not only enables them to access to a few hundred or thousand articles someone paid to have trans- lated, but all articles of the English Wikipedia, plus every English article, every English book, every English blog, ev- ery English video on the InterNet.
It also grows them not only intellectually, but also removes economical barriers for trading with other groups.
Tim
... Yeah, it's a tempting thought. Without English we wouldn't be able to
have this conversation, and do thousands of other things.
And yet, that's exactly what we as Wikimedia are not supposed to do, for reasons that mathieu stumpf guntz suggests: not only what is written in a language is knowledge; language itself is also knowledge.
On a more practical and less ideological note, I should note that even though I didn't run the numbers, I strongly suspect that translating 10,000 articles to 100 languages is considerably cheaper than teaching 7 billion people English.
Amir,
I agree with everything you said, especially that languages are knowledge in themselves, but I must say that Wikimedia is not doing much in an effort to teach languages to people. Why isn't there more effort at the WMF or as a movement to try to develop a platform to teach languages?
Jean-Philippe Béland Vice President and Programs Coordinator, Wikimedia Canada Coordinator, Wikimedians of North American Indigenous Languages User Group
Le 27/02/2018 à 18:51, Jean-Philippe Béland a écrit :
Amir,
I agree with everything you said, especially that languages are knowledge in themselves, but I must say that Wikimedia is not doing much in an effort to teach languages to people. Why isn't there more effort at the WMF or as a movement to try to develop a platform to teach languages?
I totaly support this idea. Right now there are a lot of digital solutions to learn new languages, but I'm not aware of any which is doing it with free knowledge activism in mind.
I think we could even make some programs like "start to learn, try to translate some existing free material selected according to your current level, get feedback from someone who master the language" pipeline.
Jean-Philippe Béland Vice President and Programs Coordinator, Wikimedia Canada Coordinator, Wikimedians of North American Indigenous Languages User Group _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
"Amir E. Aharoni" amir.aharoni@mail.huji.ac.il wrote:
[…]
On a more practical and less ideological note, I should note that even though I didn't run the numbers, I strongly suspect that translating 10,000 articles to 100 languages is considerably cheaper than teaching 7 billion people English.
Definitely, but your argument was:
| […]
| If there is no substantial Wikipedia in such a language, these people can't | read Wikipedia in *any language* because they are monolingual. Most likely | they cannot read any any encyclopedia in any language. They need a | Wikipedia not in order to preserve the language, but to have access to | *any* encyclopedic knowledge.
| […]
A large part of humanity *has* access to a reasonably main- tained Wikipedia in a language they understand, not to speak of traditional encyclopedias in schools and libraries.
Then of course there is the more fundamental problem: If those 100,000 monolingual speakers do not speak other lan- guages, have no access to encyclopedias, etc., how do they interact with a computer now, which web sites do they visit, etc.?
I just have a very hard time to imagine a community of 100,000 people under those circumstances who are only held back by not having access to a Wikipedia. On the contrary, this reminds me very much of traditional development prac- tices where third world countries always seem to urgently need to buy what first world countries have to sell. IMHO, there is a considerable risk that this creates unhealthy de- pendencies.
Tim
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore
2018-02-28 1:03 GMT+02:00 Tim Landscheidt tim@tim-landscheidt.de:
Then of course there is the more fundamental problem: If those 100,000 monolingual speakers do not speak other lan- guages, have no access to encyclopedias, etc., how do they interact with a computer now, which web sites do they visit, etc.?
Quite possibly, they don't visit any websites.
Can Wikipedia be a first website in a given language? Of course.
Who if not Wikipedia? In a lot of languages, the first, and sometimes the only written work is a translation of the Bible or of the UDHR. (Reminder: The Bible was the first work that was published in a lot of European languages, too.) These are usually made by some kind of a funded initiative that comes from religious or human rights organizations. Why shouldn't it be a translation of 10,000 Wikipedia articles? Why shouldn't it be an initiative from Wikimedia or another educational organization?
I just have a very hard time to imagine a community of 100,000 people under those circumstances who are only held back by not having access to a Wikipedia. On the contrary, this reminds me very much of traditional development prac- tices where third world countries always seem to urgently need to buy what first world countries have to sell. IMHO, there is a considerable risk that this creates unhealthy de- pendencies.
Hey, if people don't want it, they don't have to read it, but I suspect that if you *let* people read useful information about geography, medicine, public policy, economics, etc., they will use it.
But in very simplified terms, I see it as a competition between UN, JW, Facebook, and Wikimedia, and Wikimedia is hardly even participating. UN is a fine organization, but not very useful in people's daily life. Religious materials' contribution to development of publishing and literacy throughout history can't be denied, but the usefulness of their content can be questioned. Facebook is useful to a lot of people, and it can be localized easily, but it would be kind of depressing if that's the only thing that people do in their language. And Facebook is very actively trying to reach to the farthest corners of the world and get people connected.
And this leaves Wikimedia, which is hardly doing anything proactive to get its materials *actually* written in more languages. We are making *technologies* for translation—Wikidata, Content Translation, and more—and they are used by thousands of translators to write in dozens of languages, but we are not doing anything proactive to expand the coverage of languages beyond the usual suspects: the 70 or so languages that John Erling mentioned in the email that started this thread. The ~70 big languages take care of themselves. We've been saying that the rest of the languages can take care of themselves, but that is naïve.
Le 27/02/2018 à 12:42, Vi to a écrit :
I see Amir's points, which are pretty reasonable, but I fear this would suit languages with a significant presence on the web.
Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating basic encyclopedic terminology and style in that language", if we want to preserve a language we shouldn't create a thing.
I think that here the term "preserving" is misinterpreted. It's not about stuff it to put it in a nothing-should-move-anymore museum. It's about preserving actual use of diverse language as diachronic phenomena, ie as evolving objects.
On this regard, even largest language communities are seeing their use changing at an increasing pace, as recognize institutions like Académie française (not quite your average neologismophilic neo-punk band).
I think it's also good to recall that there are places where there is not yet a a high bandwith reliable internet (or internet at all), but that computer are accessibles. For example Libraries Without Borders[2] are providing computer boxes, which do include some Wikimedia material if I'm not mistaken. Although I'm not enough informed on their actions, but it would interesting to be in contact with them if it's not already the case. Making encyclopedia shared through travelling USB key would be surely possible for example, but that just a sketched idea.
On the other hand, should we recall that we are losing language diversity at an increasing pace?[3] And of course when a language die, it's whole culture which go with it like a bush medicine engraved in aboriginal vocabulary.[4] So really it's not about bringing knowledge to communities with less geopolitcally influence, it's about giving mankind a chance to loose as few as possible of valuable knowledge by diffusing it omnidirectly.
[1] Parce qu’il doit être tout à la fois le greffier de l’usage, le témoin de l’histoire et celui du changement le Dictionnaire de l’Académie aura donc presque doublé de volume. En consacrant ainsi un très grand nombre de mots nouveaux, l’Académie répond aux exigences du temps mais elle se montre fidèle aussi à sa tradition. http://www.academie-francaise.fr/la-langue-francaise-langue-de-la-modernite-... [2] https://www.librarieswithoutborders.org/ [3] http://www.ibtimes.co.uk/unesco-half-worlds-languages-will-disappear-by-2100... [4] https://en.wikipedia.org/wiki/Bush_medicine
By the way I was wondering my concerns about cultural colonization may be addressed -for wikis which has some contents (let's say at least 1000 articles)- by starting expanding existing articles instead of translating new ones. This would solve the problem of choosing what to translate though would leave problems about the perspective contents are created.
Vito
2018-02-27 12:31 GMT+01:00 Amir E. Aharoni amir.aharoni@mail.huji.ac.il:
2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz < psychoslave@culture-libre.org>:
Le 24/02/2018 à 18:08, Vi to a écrit :
*finally I think paid translators would hardly turn into stable Wikipedians.
I think this misses an important point that is, we don't need the
initial
translator to turn into a sustaining editor, we need the article to
evolve
with call to action incentives. And articles which don't exist at all – even as a stub – or don't meet an audience of potential contributors will never catch such an evolving cycle.
This is one of the issues with what I alluded to in my earlier email in this thread: the privilege that the "big" languages have. It's the privilege of already having other encyclopedias, textbooks, public education, etc., in this language. A lot of languages don't have these things. When you speak a language that has had these things before Wikipedia came along, it's hard to perceive the world like a person who speaks a language that doesn't perceives it.
If you define the purpose of paying somebody to translate as "turning the paid translator" into a sustaining editor, then this is indeed likely to fail.
But if you define the purpose differently, it may succeed. For example, you may define the purpose as one or more of the following:
- Demonstrating that it's possible to write an encyclopedia in that
language
- Creating basic encyclopedic terminology and style in that language
- Creating a bunch of basic articles that would appear in interlanguage
links in Wikipedias from bigger languages (English, French, etc.)
- Creating a bunch of basic articles that would appear in search results
from internet search engines
The existence of these things may bring in people who will become volunteer sustaining editors.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
wikipedia offline https://en.wikipedia.org/wiki/Kiwix
On Tue, Feb 27, 2018 at 9:50 AM, mathieu stumpf guntz < psychoslave@culture-libre.org> wrote:
Le 27/02/2018 à 12:42, Vi to a écrit :
I see Amir's points, which are pretty reasonable, but I fear this would suit languages with a significant presence on the web.
Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating basic encyclopedic terminology and style in that language", if we want to preserve a language we shouldn't create a thing.
I think that here the term "preserving" is misinterpreted. It's not about stuff it to put it in a nothing-should-move-anymore museum. It's about preserving actual use of diverse language as diachronic phenomena, ie as evolving objects.
On this regard, even largest language communities are seeing their use changing at an increasing pace, as recognize institutions like Académie française (not quite your average neologismophilic neo-punk band).
I think it's also good to recall that there are places where there is not yet a a high bandwith reliable internet (or internet at all), but that computer are accessibles. For example Libraries Without Borders[2] are providing computer boxes, which do include some Wikimedia material if I'm not mistaken. Although I'm not enough informed on their actions, but it would interesting to be in contact with them if it's not already the case. Making encyclopedia shared through travelling USB key would be surely possible for example, but that just a sketched idea.
On the other hand, should we recall that we are losing language diversity at an increasing pace?[3] And of course when a language die, it's whole culture which go with it like a bush medicine engraved in aboriginal vocabulary.[4] So really it's not about bringing knowledge to communities with less geopolitcally influence, it's about giving mankind a chance to loose as few as possible of valuable knowledge by diffusing it omnidirectly.
[1] Parce qu’il doit être tout à la fois le greffier de l’usage, le témoin de l’histoire et celui du changement le Dictionnaire de l’Académie aura donc presque doublé de volume. En consacrant ainsi un très grand nombre de mots nouveaux, l’Académie répond aux exigences du temps mais elle se montre fidèle aussi à sa tradition. http://www.academie-francaise. fr/la-langue-francaise-langue-de-la-modernite-seance-publique-annuelle [2] https://www.librarieswithoutborders.org/ [3] http://www.ibtimes.co.uk/unesco-half-worlds-languages-will- disappear-by-2100-1498154 [4] https://en.wikipedia.org/wiki/Bush_medicine
By the way I was wondering my concerns about cultural colonization may be
addressed -for wikis which has some contents (let's say at least 1000 articles)- by starting expanding existing articles instead of translating new ones. This would solve the problem of choosing what to translate though would leave problems about the perspective contents are created.
Vito
2018-02-27 12:31 GMT+01:00 Amir E. Aharoni amir.aharoni@mail.huji.ac.il :
2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz <
psychoslave@culture-libre.org>:
Le 24/02/2018 à 18:08, Vi to a écrit :
*finally I think paid translators would hardly turn into stable
Wikipedians.
I think this misses an important point that is, we don't need the
initial
translator to turn into a sustaining editor, we need the article to
evolve
with call to action incentives. And articles which don't exist at all – even as a stub – or don't meet an audience of potential contributors will never catch such an evolving cycle.
This is one of the issues with what I alluded to in my earlier email in this thread: the privilege that the "big" languages have. It's the privilege of already having other encyclopedias, textbooks, public education, etc., in this language. A lot of languages don't have these things. When you speak a language that has had these things before Wikipedia came along, it's hard to perceive the world like a person who speaks a language that doesn't perceives it.
If you define the purpose of paying somebody to translate as "turning the paid translator" into a sustaining editor, then this is indeed likely to fail.
But if you define the purpose differently, it may succeed. For example, you may define the purpose as one or more of the following:
- Demonstrating that it's possible to write an encyclopedia in that
language
- Creating basic encyclopedic terminology and style in that language
- Creating a bunch of basic articles that would appear in interlanguage
links in Wikipedias from bigger languages (English, French, etc.)
- Creating a bunch of basic articles that would appear in search results
from internet search engines
The existence of these things may bring in people who will become volunteer sustaining editors.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
If the people creating the basic encyclopaedic terminology and style in the language are native speakers, then it would not be a thing imposed from outside. It would be a development within the language, just like it was with the languages that already have encyclopaedias. The basic encyclopaedic terminology and style in languages that have then also had to be created before it existed, it just happened earlier. Living languages evolve to deal with the realities of the present. Those which don’t, tend to die out as they become less useful. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of Vi to Sent: Tuesday, February 27, 2018 1:43 PM To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I see Amir's points, which are pretty reasonable, but I fear this would suit languages with a significant presence on the web.
Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating basic encyclopedic terminology and style in that language", if we want to preserve a language we shouldn't create a thing.
By the way I was wondering my concerns about cultural colonization may be addressed -for wikis which has some contents (let's say at least 1000 articles)- by starting expanding existing articles instead of translating new ones. This would solve the problem of choosing what to translate though would leave problems about the perspective contents are created.
Vito
2018-02-27 12:31 GMT+01:00 Amir E. Aharoni amir.aharoni@mail.huji.ac.il:
2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz < psychoslave@culture-libre.org>:
Le 24/02/2018 à 18:08, Vi to a écrit :
*finally I think paid translators would hardly turn into stable Wikipedians.
I think this misses an important point that is, we don't need the
initial
translator to turn into a sustaining editor, we need the article to
evolve
with call to action incentives. And articles which don't exist at all – even as a stub – or don't meet an audience of potential contributors will never catch such an evolving cycle.
This is one of the issues with what I alluded to in my earlier email in this thread: the privilege that the "big" languages have. It's the privilege of already having other encyclopedias, textbooks, public education, etc., in this language. A lot of languages don't have these things. When you speak a language that has had these things before Wikipedia came along, it's hard to perceive the world like a person who speaks a language that doesn't perceives it.
If you define the purpose of paying somebody to translate as "turning the paid translator" into a sustaining editor, then this is indeed likely to fail.
But if you define the purpose differently, it may succeed. For example, you may define the purpose as one or more of the following:
- Demonstrating that it's possible to write an encyclopedia in that
language
- Creating basic encyclopedic terminology and style in that language
- Creating a bunch of basic articles that would appear in
interlanguage links in Wikipedias from bigger languages (English, French, etc.)
- Creating a bunch of basic articles that would appear in search
results from internet search engines
The existence of these things may bring in people who will become volunteer sustaining editors.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
_______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
(This thread is getting terribly interesting)
I generally think Wikipedia should be a strictly non interfering observer for various aspects, language included. I fear if a wiki tries to set a model for a language it may be a model which doesn't represent the reality of that language: small wikis are often monopolized by a few users. That's not a fault per se but it may introduce a significant bias in linguistic models used.
About one of Amir's emails I think a "small" Wikipedia edition is sign of a series of situations, one of the most common of is an endangered language. While planning should differentiate between endangered and non endangered language I think most of problems we have to face are related to languages endangered at various levels.
On a more practical and less ideological note, I should note that even though I didn't run the numbers, I strongly suspect that translating 10,000 articles to 100 languages is considerably cheaper than teaching 7 billion people English.
I don't why but I tend to second your suspects :p
Vito
2018-02-27 16:53 GMT+01:00 Peter Southwood peter.southwood@telkomsa.net:
If the people creating the basic encyclopaedic terminology and style in the language are native speakers, then it would not be a thing imposed from outside. It would be a development within the language, just like it was with the languages that already have encyclopaedias. The basic encyclopaedic terminology and style in languages that have then also had to be created before it existed, it just happened earlier. Living languages evolve to deal with the realities of the present. Those which don’t, tend to die out as they become less useful. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of Vi to Sent: Tuesday, February 27, 2018 1:43 PM To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I see Amir's points, which are pretty reasonable, but I fear this would suit languages with a significant presence on the web.
Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating basic encyclopedic terminology and style in that language", if we want to preserve a language we shouldn't create a thing.
By the way I was wondering my concerns about cultural colonization may be addressed -for wikis which has some contents (let's say at least 1000 articles)- by starting expanding existing articles instead of translating new ones. This would solve the problem of choosing what to translate though would leave problems about the perspective contents are created.
Vito
2018-02-27 12:31 GMT+01:00 Amir E. Aharoni amir.aharoni@mail.huji.ac.il:
2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz < psychoslave@culture-libre.org>:
Le 24/02/2018 à 18:08, Vi to a écrit :
*finally I think paid translators would hardly turn into stable Wikipedians.
I think this misses an important point that is, we don't need the
initial
translator to turn into a sustaining editor, we need the article to
evolve
with call to action incentives. And articles which don't exist at all – even as a stub – or don't meet an audience of potential contributors will never catch such an evolving cycle.
This is one of the issues with what I alluded to in my earlier email in this thread: the privilege that the "big" languages have. It's the privilege of already having other encyclopedias, textbooks, public education, etc., in this language. A lot of languages don't have these things. When you speak a language that has had these things before Wikipedia came along, it's hard to perceive the world like a person who speaks a language that doesn't perceives it.
If you define the purpose of paying somebody to translate as "turning the paid translator" into a sustaining editor, then this is indeed likely to fail.
But if you define the purpose differently, it may succeed. For example, you may define the purpose as one or more of the following:
- Demonstrating that it's possible to write an encyclopedia in that
language
- Creating basic encyclopedic terminology and style in that language
- Creating a bunch of basic articles that would appear in
interlanguage links in Wikipedias from bigger languages (English, French, etc.)
- Creating a bunch of basic articles that would appear in search
results from internet search engines
The existence of these things may bring in people who will become volunteer sustaining editors.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Does it make sense to have more articles in a language than can be curated by the volunteers who speak that language? This has already happened on the Englisg-language Wikipedia where the five million articles have simply overwhelmed the capability of the few thousand active contributors to self-organise and curate -- for example, there are about one million articles without adequate sources, and thousands of unsourced BLP; there are copyvio cleanups that will not complete, if ever, before 2030. An army of hand-coded bots is just about keeping on top of vandalism. How does that scale to projects where the number of native speaker contributors is in the dozens rather than the thousands?
On Tue, Feb 27, 2018 at 4:17 PM, Vi to vituzzu.wiki@gmail.com wrote:
(This thread is getting terribly interesting)
I generally think Wikipedia should be a strictly non interfering observer for various aspects, language included. I fear if a wiki tries to set a model for a language it may be a model which doesn't represent the reality of that language: small wikis are often monopolized by a few users. That's not a fault per se but it may introduce a significant bias in linguistic models used.
About one of Amir's emails I think a "small" Wikipedia edition is sign of a series of situations, one of the most common of is an endangered language. While planning should differentiate between endangered and non endangered language I think most of problems we have to face are related to languages endangered at various levels.
On a more practical and less ideological note, I should note that even though I didn't run the numbers, I strongly suspect that translating 10,000 articles to 100 languages is considerably cheaper than teaching 7 billion people English.
I don't why but I tend to second your suspects :p
Vito
2018-02-27 16:53 GMT+01:00 Peter Southwood peter.southwood@telkomsa.net:
If the people creating the basic encyclopaedic terminology and style in the language are native speakers, then it would not be a thing imposed
from
outside. It would be a development within the language, just like it was with the languages that already have encyclopaedias. The basic encyclopaedic terminology and style in languages that have then also had
to
be created before it existed, it just happened earlier. Living languages evolve to deal with the realities of the present. Those which don’t, tend to die out as they become less useful. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of Vi to Sent: Tuesday, February 27, 2018 1:43 PM To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I see Amir's points, which are pretty reasonable, but I fear this would suit languages with a significant presence on the web.
Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating basic encyclopedic terminology and style in that language", if
we
want to preserve a language we shouldn't create a thing.
By the way I was wondering my concerns about cultural colonization may be addressed -for wikis which has some contents (let's say at least 1000 articles)- by starting expanding existing articles instead of translating new ones. This would solve the problem of choosing what to translate
though
would leave problems about the perspective contents are created.
Vito
2018-02-27 12:31 GMT+01:00 Amir E. Aharoni <amir.aharoni@mail.huji.ac.il :
2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz < psychoslave@culture-libre.org>:
Le 24/02/2018 à 18:08, Vi to a écrit :
*finally I think paid translators would hardly turn into stable Wikipedians.
I think this misses an important point that is, we don't need the
initial
translator to turn into a sustaining editor, we need the article to
evolve
with call to action incentives. And articles which don't exist at all – even as a stub – or don't meet an audience of potential contributors will never catch such an evolving cycle.
This is one of the issues with what I alluded to in my earlier email in this thread: the privilege that the "big" languages have. It's the privilege of already having other encyclopedias, textbooks, public education, etc., in this language. A lot of languages don't have these things. When you speak a language that has had these things before Wikipedia came along, it's hard to perceive the world like a person who speaks a language that doesn't perceives it.
If you define the purpose of paying somebody to translate as "turning the paid translator" into a sustaining editor, then this is indeed likely to fail.
But if you define the purpose differently, it may succeed. For example, you may define the purpose as one or more of the following:
- Demonstrating that it's possible to write an encyclopedia in that
language
- Creating basic encyclopedic terminology and style in that language
- Creating a bunch of basic articles that would appear in
interlanguage links in Wikipedias from bigger languages (English, French, etc.)
- Creating a bunch of basic articles that would appear in search
results from internet search engines
The existence of these things may bring in people who will become volunteer sustaining editors.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Yes, it makes sense.
This is yet another thing that is a challenge, not a blocker.
English speakers have a useful result from Wikipedia coming up in almost every Google search. *Suspected* copyvio issues are an acceptable price to pay for this privilege. (Particularly bad copyvio issues are handled through OTRS.)
People who speak many other languages don't have the privilege of such high availability of useful knowledge. So first, let's not imagine problems that will prevent them from getting this. People who currently don't have this wealth of information in their language wish that they had such a problem (even if not consciously).
Besides, when people start getting useful search results in their language, they will read the articles, and some of them will become editors and the community will grow. It happened in English in 2002, and it can happen in other languages.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore
2018-02-27 19:49 GMT+02:00 Renée Bagslint reneebagslint@gmail.com:
Does it make sense to have more articles in a language than can be curated by the volunteers who speak that language? This has already happened on the Englisg-language Wikipedia where the five million articles have simply overwhelmed the capability of the few thousand active contributors to self-organise and curate -- for example, there are about one million articles without adequate sources, and thousands of unsourced BLP; there are copyvio cleanups that will not complete, if ever, before 2030. An army of hand-coded bots is just about keeping on top of vandalism. How does that scale to projects where the number of native speaker contributors is in the dozens rather than the thousands?
On Tue, Feb 27, 2018 at 4:17 PM, Vi to vituzzu.wiki@gmail.com wrote:
(This thread is getting terribly interesting)
I generally think Wikipedia should be a strictly non interfering observer for various aspects, language included. I fear if a wiki tries to set a model for a language it may be a model which doesn't represent the
reality
of that language: small wikis are often monopolized by a few users.
That's
not a fault per se but it may introduce a significant bias in linguistic models used.
About one of Amir's emails I think a "small" Wikipedia edition is sign
of a
series of situations, one of the most common of is an endangered
language.
While planning should differentiate between endangered and non endangered language I think most of problems we have to face are related to
languages
endangered at various levels.
On a more practical and less ideological note, I should note that even though I didn't run the numbers, I strongly suspect that translating
10,000
articles to 100 languages is considerably cheaper than teaching 7 billion people English.
I don't why but I tend to second your suspects :p
Vito
2018-02-27 16:53 GMT+01:00 Peter Southwood <peter.southwood@telkomsa.net :
If the people creating the basic encyclopaedic terminology and style in the language are native speakers, then it would not be a thing imposed
from
outside. It would be a development within the language, just like it
was
with the languages that already have encyclopaedias. The basic encyclopaedic terminology and style in languages that have then also
had
to
be created before it existed, it just happened earlier. Living
languages
evolve to deal with the realities of the present. Those which don’t,
tend
to die out as they become less useful. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of Vi to Sent: Tuesday, February 27, 2018 1:43 PM To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I see Amir's points, which are pretty reasonable, but I fear this would suit languages with a significant presence on the web.
Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating basic encyclopedic terminology and style in that language",
if
we
want to preserve a language we shouldn't create a thing.
By the way I was wondering my concerns about cultural colonization may
be
addressed -for wikis which has some contents (let's say at least 1000 articles)- by starting expanding existing articles instead of
translating
new ones. This would solve the problem of choosing what to translate
though
would leave problems about the perspective contents are created.
Vito
2018-02-27 12:31 GMT+01:00 Amir E. Aharoni <
amir.aharoni@mail.huji.ac.il
:
2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz < psychoslave@culture-libre.org>:
Le 24/02/2018 à 18:08, Vi to a écrit :
*finally I think paid translators would hardly turn into stable Wikipedians.
I think this misses an important point that is, we don't need the
initial
translator to turn into a sustaining editor, we need the article to
evolve
with call to action incentives. And articles which don't exist at all – even as a stub – or don't meet an audience of potential contributors will never catch such an evolving cycle.
This is one of the issues with what I alluded to in my earlier email in this thread: the privilege that the "big" languages have. It's the privilege of already having other encyclopedias, textbooks, public education, etc., in this language. A lot of languages don't have
these
things. When you speak a language that has had these things before Wikipedia came along, it's hard to perceive the world like a person who speaks a language that doesn't perceives it.
If you define the purpose of paying somebody to translate as "turning the paid translator" into a sustaining editor, then this is indeed likely to fail.
But if you define the purpose differently, it may succeed. For example, you may define the purpose as one or more of the following:
- Demonstrating that it's possible to write an encyclopedia in that
language
- Creating basic encyclopedic terminology and style in that language
- Creating a bunch of basic articles that would appear in
interlanguage links in Wikipedias from bigger languages (English, French, etc.)
- Creating a bunch of basic articles that would appear in search
results from internet search engines
The existence of these things may bring in people who will become volunteer sustaining editors.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to
live
in peace.” – T. Moore ______________________________
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
We do need better tools to curate the existing articles, but that is not a blocker for new ways to create and edit articles.
For example, what if we could simply select a sentence, create a query on some search engine, and then have an ai-bot crawl the result to see if one of the hits can be used as a source? Turned around, the ai-bot could check the sentences in an article and flag those it can't verify, thus guiding the editor to back those sentences with references. That would off-load the bulk of the work on sourcing articles.
Just an idea.
On Tue, Feb 27, 2018 at 6:49 PM, Renée Bagslint reneebagslint@gmail.com wrote:
Does it make sense to have more articles in a language than can be curated by the volunteers who speak that language? This has already happened on the Englisg-language Wikipedia where the five million articles have simply overwhelmed the capability of the few thousand active contributors to self-organise and curate -- for example, there are about one million articles without adequate sources, and thousands of unsourced BLP; there are copyvio cleanups that will not complete, if ever, before 2030. An army of hand-coded bots is just about keeping on top of vandalism. How does that scale to projects where the number of native speaker contributors is in the dozens rather than the thousands?
On Tue, Feb 27, 2018 at 4:17 PM, Vi to vituzzu.wiki@gmail.com wrote:
(This thread is getting terribly interesting)
I generally think Wikipedia should be a strictly non interfering observer for various aspects, language included. I fear if a wiki tries to set a model for a language it may be a model which doesn't represent the
reality
of that language: small wikis are often monopolized by a few users.
That's
not a fault per se but it may introduce a significant bias in linguistic models used.
About one of Amir's emails I think a "small" Wikipedia edition is sign
of a
series of situations, one of the most common of is an endangered
language.
While planning should differentiate between endangered and non endangered language I think most of problems we have to face are related to
languages
endangered at various levels.
On a more practical and less ideological note, I should note that even though I didn't run the numbers, I strongly suspect that translating
10,000
articles to 100 languages is considerably cheaper than teaching 7 billion people English.
I don't why but I tend to second your suspects :p
Vito
2018-02-27 16:53 GMT+01:00 Peter Southwood <peter.southwood@telkomsa.net :
If the people creating the basic encyclopaedic terminology and style in the language are native speakers, then it would not be a thing imposed
from
outside. It would be a development within the language, just like it
was
with the languages that already have encyclopaedias. The basic encyclopaedic terminology and style in languages that have then also
had
to
be created before it existed, it just happened earlier. Living
languages
evolve to deal with the realities of the present. Those which don’t,
tend
to die out as they become less useful. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of Vi to Sent: Tuesday, February 27, 2018 1:43 PM To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I see Amir's points, which are pretty reasonable, but I fear this would suit languages with a significant presence on the web.
Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating basic encyclopedic terminology and style in that language",
if
we
want to preserve a language we shouldn't create a thing.
By the way I was wondering my concerns about cultural colonization may
be
addressed -for wikis which has some contents (let's say at least 1000 articles)- by starting expanding existing articles instead of
translating
new ones. This would solve the problem of choosing what to translate
though
would leave problems about the perspective contents are created.
Vito
2018-02-27 12:31 GMT+01:00 Amir E. Aharoni <
amir.aharoni@mail.huji.ac.il
:
2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz < psychoslave@culture-libre.org>:
Le 24/02/2018 à 18:08, Vi to a écrit :
*finally I think paid translators would hardly turn into stable Wikipedians.
I think this misses an important point that is, we don't need the
initial
translator to turn into a sustaining editor, we need the article to
evolve
with call to action incentives. And articles which don't exist at all – even as a stub – or don't meet an audience of potential contributors will never catch such an evolving cycle.
This is one of the issues with what I alluded to in my earlier email in this thread: the privilege that the "big" languages have. It's the privilege of already having other encyclopedias, textbooks, public education, etc., in this language. A lot of languages don't have
these
things. When you speak a language that has had these things before Wikipedia came along, it's hard to perceive the world like a person who speaks a language that doesn't perceives it.
If you define the purpose of paying somebody to translate as "turning the paid translator" into a sustaining editor, then this is indeed likely to fail.
But if you define the purpose differently, it may succeed. For example, you may define the purpose as one or more of the following:
- Demonstrating that it's possible to write an encyclopedia in that
language
- Creating basic encyclopedic terminology and style in that language
- Creating a bunch of basic articles that would appear in
interlanguage links in Wikipedias from bigger languages (English, French, etc.)
- Creating a bunch of basic articles that would appear in search
results from internet search engines
The existence of these things may bring in people who will become volunteer sustaining editors.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to
live
in peace.” – T. Moore ______________________________
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I like the idea, but have no clue as to how practicable it would be. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of John Erling Blad Sent: Tuesday, March 6, 2018 12:36 AM To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
We do need better tools to curate the existing articles, but that is not a blocker for new ways to create and edit articles.
For example, what if we could simply select a sentence, create a query on some search engine, and then have an ai-bot crawl the result to see if one of the hits can be used as a source? Turned around, the ai-bot could check the sentences in an article and flag those it can't verify, thus guiding the editor to back those sentences with references. That would off-load the bulk of the work on sourcing articles.
Just an idea.
On Tue, Feb 27, 2018 at 6:49 PM, Renée Bagslint reneebagslint@gmail.com wrote:
Does it make sense to have more articles in a language than can be curated by the volunteers who speak that language? This has already happened on the Englisg-language Wikipedia where the five million articles have simply overwhelmed the capability of the few thousand active contributors to self-organise and curate -- for example, there are about one million articles without adequate sources, and thousands of unsourced BLP; there are copyvio cleanups that will not complete, if ever, before 2030. An army of hand-coded bots is just about keeping on top of vandalism. How does that scale to projects where the number of native speaker contributors is in the dozens rather than the thousands?
On Tue, Feb 27, 2018 at 4:17 PM, Vi to vituzzu.wiki@gmail.com wrote:
(This thread is getting terribly interesting)
I generally think Wikipedia should be a strictly non interfering observer for various aspects, language included. I fear if a wiki tries to set a model for a language it may be a model which doesn't represent the
reality
of that language: small wikis are often monopolized by a few users.
That's
not a fault per se but it may introduce a significant bias in linguistic models used.
About one of Amir's emails I think a "small" Wikipedia edition is sign
of a
series of situations, one of the most common of is an endangered
language.
While planning should differentiate between endangered and non endangered language I think most of problems we have to face are related to
languages
endangered at various levels.
On a more practical and less ideological note, I should note that even though I didn't run the numbers, I strongly suspect that translating
10,000
articles to 100 languages is considerably cheaper than teaching 7 billion people English.
I don't why but I tend to second your suspects :p
Vito
2018-02-27 16:53 GMT+01:00 Peter Southwood <peter.southwood@telkomsa.net :
If the people creating the basic encyclopaedic terminology and style in the language are native speakers, then it would not be a thing imposed
from
outside. It would be a development within the language, just like it
was
with the languages that already have encyclopaedias. The basic encyclopaedic terminology and style in languages that have then also
had
to
be created before it existed, it just happened earlier. Living
languages
evolve to deal with the realities of the present. Those which don’t,
tend
to die out as they become less useful. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of Vi to Sent: Tuesday, February 27, 2018 1:43 PM To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I see Amir's points, which are pretty reasonable, but I fear this would suit languages with a significant presence on the web.
Among them I agree with points 1, 3 and 4 while I'm not sure about #2 "creating basic encyclopedic terminology and style in that language",
if
we
want to preserve a language we shouldn't create a thing.
By the way I was wondering my concerns about cultural colonization may
be
addressed -for wikis which has some contents (let's say at least 1000 articles)- by starting expanding existing articles instead of
translating
new ones. This would solve the problem of choosing what to translate
though
would leave problems about the perspective contents are created.
Vito
2018-02-27 12:31 GMT+01:00 Amir E. Aharoni <
amir.aharoni@mail.huji.ac.il
:
2018-02-27 13:00 GMT+02:00 mathieu stumpf guntz < psychoslave@culture-libre.org>:
Le 24/02/2018 à 18:08, Vi to a écrit :
*finally I think paid translators would hardly turn into stable Wikipedians.
I think this misses an important point that is, we don't need the
initial
translator to turn into a sustaining editor, we need the article to
evolve
with call to action incentives. And articles which don't exist at all – even as a stub – or don't meet an audience of potential contributors will never catch such an evolving cycle.
This is one of the issues with what I alluded to in my earlier email in this thread: the privilege that the "big" languages have. It's the privilege of already having other encyclopedias, textbooks, public education, etc., in this language. A lot of languages don't have
these
things. When you speak a language that has had these things before Wikipedia came along, it's hard to perceive the world like a person who speaks a language that doesn't perceives it.
If you define the purpose of paying somebody to translate as "turning the paid translator" into a sustaining editor, then this is indeed likely to fail.
But if you define the purpose differently, it may succeed. For example, you may define the purpose as one or more of the following:
- Demonstrating that it's possible to write an encyclopedia in
that language
- Creating basic encyclopedic terminology and style in that
language
- Creating a bunch of basic articles that would appear in
interlanguage links in Wikipedias from bigger languages (English, French, etc.)
- Creating a bunch of basic articles that would appear in search
results from internet search engines
The existence of these things may bring in people who will become volunteer sustaining editors.
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to
live
in peace.” – T. Moore ______________________________
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscr ibe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscrib e
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscrib e
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
_______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Those who pay get to select what is translated. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of Jean-Philippe Béland Sent: 24 February 2018 16:55 To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my simple opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com wrote:
Should have added that the remaining points are somewhat less interesting in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should also be pretty broad, not very narrow technical or medical, ie vertical articles, as the number of editors that can handle those will be pretty small.
In particular: Do not believe you can turn a teanslator into a new editor! You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles
are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the project.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would
require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles
are extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP.
Thus we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts
more efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would
require significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple
of years. The translators at TWB did not want to become Wikipedians or learn how
to
use our systems. The coordinator created account like TransSW001 (one
for
each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job
of
teaching high school students.
- The project continues but at a slower pace than before. The
Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken
by
40 million people in Eastern India. The amazing thing is that for many
of
these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into
Chinese.
There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad
to
make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles
from
"List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word
(about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that
lacks
good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a
community at
all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be avoided.
Perhaps
we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does
not
have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they need
a
lot
of adjustments. Perhaps such a project would identify our inherit
bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscri be
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
_______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
--- This email has been checked for viruses by AVG. http://www.avg.com
Yes, and then there's always the question. If he's getting paid, why aren't I? Why is he getting paid per word of article translated? Why am I not getting paid per spamvertisement deleted or vandal blocked? Why am I not getting paid for closing discussions that it takes hours of reading input and considering all sides and getting rocks thrown at me no matter what I do? Is that not valuable to the project as well?
If you want to pay anyone, you better start paying me. I'm okay with the idea of being a volunteer as long as everyone is a volunteer. But if you start paying some people and not me, we're going to have a problem.
Todd
On Sat, Feb 24, 2018 at 12:47 PM, Peter Southwood < peter.southwood@telkomsa.net> wrote:
Those who pay get to select what is translated. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of Jean-Philippe Béland Sent: 24 February 2018 16:55 To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my simple opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com wrote:
Should have added that the remaining points are somewhat less interesting in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should also be pretty broad, not very narrow technical or medical, ie vertical articles, as the number of editors that can handle those will
be pretty small.
In particular: Do not believe you can turn a teanslator into a new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles
are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the project.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would
require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com
wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles
are extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP.
Thus we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts
more efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would
require significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple
of years. The translators at TWB did not want to become Wikipedians or learn how
to
use our systems. The coordinator created account like TransSW001 (one
for
each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job
of
teaching high school students.
- The project continues but at a slower pace than before. The
Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken
by
40 million people in Eastern India. The amazing thing is that for many
of
these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into
Chinese.
There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad
to
make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles
from
"List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word
(about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that
lacks
good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a
community at
all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be avoided.
Perhaps
we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does
not
have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they need
a
lot
of adjustments. Perhaps such a project would identify our inherit
bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscri be
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
This email has been checked for viruses by AVG. http://www.avg.com
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I actually agree with Todd, and I though this is actually a reason why WMF staff may not edit articles (at least not from WMF accounts). I am afraid disadvantages due to the broken symmetry will be bigger than advantages due to actual content translated.
Cheers Yaroslav
On Sat, Feb 24, 2018 at 9:40 PM, Todd Allen toddmallen@gmail.com wrote:
Yes, and then there's always the question. If he's getting paid, why aren't I? Why is he getting paid per word of article translated? Why am I not getting paid per spamvertisement deleted or vandal blocked? Why am I not getting paid for closing discussions that it takes hours of reading input and considering all sides and getting rocks thrown at me no matter what I do? Is that not valuable to the project as well?
If you want to pay anyone, you better start paying me. I'm okay with the idea of being a volunteer as long as everyone is a volunteer. But if you start paying some people and not me, we're going to have a problem.
Todd
On Sat, Feb 24, 2018 at 12:47 PM, Peter Southwood < peter.southwood@telkomsa.net> wrote:
Those who pay get to select what is translated. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of Jean-Philippe Béland Sent: 24 February 2018 16:55 To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my simple opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com
wrote:
Should have added that the remaining points are somewhat less interesting in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should also be pretty broad, not very narrow technical or medical, ie vertical articles, as the number of editors that can handle those will
be pretty small.
In particular: Do not believe you can turn a teanslator into a new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles
are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the project.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would
require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com
wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles
are extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP.
Thus we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts
more efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would
require significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple
of years. The translators at TWB did not want to become Wikipedians or learn how
to
use our systems. The coordinator created account like TransSW001 (one
for
each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job
of
teaching high school students.
- The project continues but at a slower pace than before. The
Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken
by
40 million people in Eastern India. The amazing thing is that for many
of
these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into
Chinese.
There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad
to
make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles
from
"List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word
(about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that
lacks
good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a
community at
all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be
avoided.
Perhaps
we should also identify good source articles, that would be a
help.
Translated articles should be above some minimum size, but they does
not
have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they need
a
lot
of adjustments. Perhaps such a project would identify our inherit
bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscri be
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
This email has been checked for viruses by AVG. http://www.avg.com
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
It is a long time since everyone on these projects were solely volunteers. :)
On Sat, Feb 24, 2018 at 9:40 PM, Todd Allen toddmallen@gmail.com wrote:
Yes, and then there's always the question. If he's getting paid, why aren't I? Why is he getting paid per word of article translated? Why am I not getting paid per spamvertisement deleted or vandal blocked? Why am I not getting paid for closing discussions that it takes hours of reading input and considering all sides and getting rocks thrown at me no matter what I do? Is that not valuable to the project as well?
If you want to pay anyone, you better start paying me. I'm okay with the idea of being a volunteer as long as everyone is a volunteer. But if you start paying some people and not me, we're going to have a problem.
Todd
On Sat, Feb 24, 2018 at 12:47 PM, Peter Southwood < peter.southwood@telkomsa.net> wrote:
Those who pay get to select what is translated. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of Jean-Philippe Béland Sent: 24 February 2018 16:55 To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my simple opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com
wrote:
Should have added that the remaining points are somewhat less interesting in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should also be pretty broad, not very narrow technical or medical, ie vertical articles, as the number of editors that can handle those will
be pretty small.
In particular: Do not believe you can turn a teanslator into a new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles
are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the project.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would
require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com
wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles
are extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP.
Thus we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts
more efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would
require significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple
of years. The translators at TWB did not want to become Wikipedians or learn how
to
use our systems. The coordinator created account like TransSW001 (one
for
each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job
of
teaching high school students.
- The project continues but at a slower pace than before. The
Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken
by
40 million people in Eastern India. The amazing thing is that for many
of
these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into
Chinese.
There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad
to
make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles
from
"List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word
(about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that
lacks
good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a
community at
all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be
avoided.
Perhaps
we should also identify good source articles, that would be a
help.
Translated articles should be above some minimum size, but they does
not
have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they need
a
lot
of adjustments. Perhaps such a project would identify our inherit
bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscri be
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
This email has been checked for viruses by AVG. http://www.avg.com
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
John and All,
As a possible complement to this discussion, CC-4 MIT OpenCourseWare-centric World University and School seeks to matriculate students in all ~200 countries' official/main languages ( https://wiki.worlduniversityandschool.org/wiki/Languages), and may compensate them for work in a number of ways, including translation and developing machine translation (and in all 7,099 living languages eventually).
World Univ. and Sch. donated ourselves to Wikidata in 2015 for co-development, and got a new WUaS Miraheze Mediawiki last year in these regards too.
Cheers, Scott - https://wiki.worlduniversityandschool.org/wiki/Nation_States (each to become a major online University for free CC-4 OCW degrees)
On Sat, Feb 24, 2018 at 1:49 PM, John Erling Blad jeblad@gmail.com wrote:
It is a long time since everyone on these projects were solely volunteers. :)
On Sat, Feb 24, 2018 at 9:40 PM, Todd Allen toddmallen@gmail.com wrote:
Yes, and then there's always the question. If he's getting paid, why
aren't
I? Why is he getting paid per word of article translated? Why am I not getting paid per spamvertisement deleted or vandal blocked? Why am I not getting paid for closing discussions that it takes hours of reading input and considering all sides and getting rocks thrown at me no matter what I do? Is that not valuable to the project as well?
If you want to pay anyone, you better start paying me. I'm okay with the idea of being a volunteer as long as everyone is a volunteer. But if you start paying some people and not me, we're going to have a problem.
Todd
On Sat, Feb 24, 2018 at 12:47 PM, Peter Southwood < peter.southwood@telkomsa.net> wrote:
Those who pay get to select what is translated. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of Jean-Philippe Béland Sent: 24 February 2018 16:55 To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my simple opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com
wrote:
Should have added that the remaining points are somewhat less interesting in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves.
Articles
should also be pretty broad, not very narrow technical or medical, ie vertical articles, as the number of editors that can handle those
will
be pretty small.
In particular: Do not believe you can turn a teanslator into a new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles
are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the
project.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also
love
the tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish,
and
Italian there is often already at least some content on many of
the
topics in question.
The
issue than becomes integration which needs an expert Wikipedia.
And
for languages in which we have little content there are often few avaliable volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would
require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the
project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com
wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles
are extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN
WP.
Thus we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made
efforts
more efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also
love
the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our
partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish,
and
Italian there is often already at least some content on many of
the
topics in question.
The
issue than becomes integration which needs an expert Wikipedia.
And
for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would
require significant checks and balances to make sure people are taking the work seriously and not simple using Google translate
for
the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple
of years. The translators at TWB did not want to become Wikipedians or learn how
to
use our systems. The coordinator created account like TransSW001 (one
for
each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads
of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job
of
teaching high school students.
- The project continues but at a slower pace than before. The
Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken
by
40 million people in Eastern India. The amazing thing is that for many
of
these topics this is the first and only information online about
it.
Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into
Chinese.
There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad
to
make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
> This discussion is going to be fun! =D > > A little more than seventy Wikipedia-projects has more than 65k articles, > the remaining two hundred or so are pretty small. > > What if a base set of articles were opened for paid translators? > There are > several lists of such base sets. We have both the thousand > articles
from
> "List of articles every Wikipedia should have"[1] and and the
ten
thousand > articles from the expanded list[2]. > > Lets say verified good translators was paid about $0.01 per word
(about
$1 > for a 1k-article) for translating one of those articles into > another language, with perhaps a higher pay for contributors in > high-cost countries. The pay would also have to be higher for > languages that
lacks
> good translation tools. > > I believe this would be an _enabling_ activity for the > communities, as without a base set of articles it won't be > possible to build a community at > all. By not paying for new articles, and only translating well-referenced > articles, some of the disputes in the communities could be
avoided.
Perhaps > we should also identify good source articles, that would be a
help.
> Translated articles should be above some minimum size, but they > does
not
> have to be full translations of the source article. > > A real problem is that our existing lists of good articles other projects > should have is pretty much biased towards Western World, so they > need
a
lot > of adjustments. Perhaps such a project would identify our
inherit
bias?
> > [1] > https://meta.wikimedia.org/wiki/List_of_articles_every_ > Wikipedia_should_have > [2] > https://meta.wikimedia.org/wiki/List_of_articles_every_ > Wikipedia_should_have/Expanded > _______________________________________________ > Wikimedia-l mailing list, guidelines at: > https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and > https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: > Wikimedia-l@lists.wikimedia.org > Unsubscribe: > https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
,
> <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscri
> be>
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe
>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
This email has been checked for viruses by AVG. http://www.avg.com
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I like this! +1000!!
On Sat, Feb 24, 2018 at 11:07 PM, Info WorldUniversity < info@worlduniversityandschool.org> wrote:
John and All,
As a possible complement to this discussion, CC-4 MIT OpenCourseWare-centric World University and School seeks to matriculate students in all ~200 countries' official/main languages ( https://wiki.worlduniversityandschool.org/wiki/Languages), and may compensate them for work in a number of ways, including translation and developing machine translation (and in all 7,099 living languages eventually).
World Univ. and Sch. donated ourselves to Wikidata in 2015 for co-development, and got a new WUaS Miraheze Mediawiki last year in these regards too.
Cheers, Scott
(each to become a major online University for free CC-4 OCW degrees)
On Sat, Feb 24, 2018 at 1:49 PM, John Erling Blad jeblad@gmail.com wrote:
It is a long time since everyone on these projects were solely
volunteers.
:)
On Sat, Feb 24, 2018 at 9:40 PM, Todd Allen toddmallen@gmail.com
wrote:
Yes, and then there's always the question. If he's getting paid, why
aren't
I? Why is he getting paid per word of article translated? Why am I not getting paid per spamvertisement deleted or vandal blocked? Why am I
not
getting paid for closing discussions that it takes hours of reading
input
and considering all sides and getting rocks thrown at me no matter
what I
do? Is that not valuable to the project as well?
If you want to pay anyone, you better start paying me. I'm okay with
the
idea of being a volunteer as long as everyone is a volunteer. But if
you
start paying some people and not me, we're going to have a problem.
Todd
On Sat, Feb 24, 2018 at 12:47 PM, Peter Southwood < peter.southwood@telkomsa.net> wrote:
Those who pay get to select what is translated. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org]
On
Behalf Of Jean-Philippe Béland Sent: 24 February 2018 16:55 To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my
simple
opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com
wrote:
Should have added that the remaining points are somewhat less interesting in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves.
Articles
should also be pretty broad, not very narrow technical or medical,
ie
vertical articles, as the number of editors that can handle those
will
be pretty small.
In particular: Do not believe you can turn a teanslator into a new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad <
jeblad@gmail.com>
wrote:
- You must start with high quality content and thus all articles
are > extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the
project.
- The "Content Translation" tool developed by the WMF made
efforts
more > efficient than handing around word documents. Would love to see > that
tool
> improved further such as having it support specific lists of > articles
that
> are deemed ready for translation by certain groups. Would also
love
> the tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty
obvious.
- We used volunteer translators mostly associated with our
partner
> Translators Without Borders. One issue we found was that
languages
> in which their are lots of translators such as French, Spanish,
and
> Italian there is often already at least some content on many of
the
> topics in question.
The
> issue than becomes integration which needs an expert Wikipedia.
And
> for languages in which we have little content there are often
few
> avaliable volunteers.
I used projects below 65k articles as an example, as the chance
of
competing articles are pretty low.
- With respect to "paying per word" the problem is this would
require > significant checks and balances to make sure people are taking
the
> work seriously and not simple using Google translate for the 70
or
> so
languages
> in which it claims to work. We often had translations undergo a > second review and the volunteers at TWB have to pass certain
tests
> to be accepted.
I'n my original email I wrote "verified good translators". It is
as
simple as "Has the editor contributed other articles at the
project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman <jmh649@gmail.com
wrote:
> We learned a few things during the medical translation project > which started back in 2011: > > 1) You must start with high quality content and thus all
articles
> are extensively improved before being proposed for translation. > > 2) A lot of languages want "less" content than is present on EN
WP.
> Thus we moved to just improving and suggesting for translation
the
> leads of the English articles. > > 3) The "Content Translation" tool developed by the WMF made
efforts
> more efficient than handing around word documents. Would love to > see that
tool
> improved further such as having it support specific lists of > articles
that
> are deemed ready for translation by certain groups. Would also
love
> the tool to have tracking metrics for these types of projects. > > 4) We used volunteer translators mostly associated with our
partner
> Translators Without Borders. One issue we found was that
languages
> in which their are lots of translators such as French, Spanish,
and
> Italian there is often already at least some content on many of
the
> topics in question.
The
> issue than becomes integration which needs an expert Wikipedia.
And
> for languages in which we have little content there are often
few
> avaliable volunteers. > > 5) With respect to "paying per word" the problem is this would > require significant checks and balances to make sure people are > taking the work seriously and not simple using Google translate
for
> the 70 or so
languages
> in which it claims to work. We often had translations undergo a > second review and the volunteers at TWB have to pass certain
tests
> to be accepted. > > 6) I hired a coordinator for the translation project for a
couple
> of years. > The translators at TWB did not want to become Wikipedians or
learn
> how
to
> use our systems. The coordinator created account like TransSW001 > (one
for
> each volunteer) and preloaded the article to be translated into > Content Translation. They than gave the volunteer translator the > user name and password to the account. > > 7) Were are we at now? There are currently just over 1,000 leads
of
> articles that have been improved and are ready for translation. > This includes articles on the 440 medications that are on the
WHO
> Essential List. We have worked a bit in some 100 languages. The > efforts have resulted in more than 5 million works translated
and
> integrated into different Wikipedias. The coordinator has > unfortunately moved on to his real job
of
> teaching high school students. > > 8) The project continues but at a slower pace than before. The
Wikipedian
> and retired orthopedic surgeon Subas Chandra Rout has basically > single handedly translated nearly all 1,000 leads into Odia a > language spoken
by
> 40 million people in Eastern India. The amazing thing is that
for
> many
of
> these topics this is the first and only information online about
it.
> Google > translate does not even claim to work in this language. Our > partnerships with WMTW and medical school in Taipai continue to > translate into
Chinese.
> There the students translate and than their translations are > reviewed by their profs before being posted. They translate in > groups using hackpad
to
> make it more social. > > I am currently working to re invigorate the project :-) James > > On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad > jeblad@gmail.com > wrote: > > > This discussion is going to be fun! =D > > > > A little more than seventy Wikipedia-projects has more than
65k
> articles, > > the remaining two hundred or so are pretty small. > > > > What if a base set of articles were opened for paid
translators?
> > There > are > > several lists of such base sets. We have both the thousand > > articles
from
> > "List of articles every Wikipedia should have"[1] and and the
ten
> thousand > > articles from the expanded list[2]. > > > > Lets say verified good translators was paid about $0.01 per
word
(about
> $1 > > for a 1k-article) for translating one of those articles into > > another language, with perhaps a higher pay for contributors
in
> > high-cost countries. The pay would also have to be higher for > > languages that
lacks
> > good translation tools. > > > > I believe this would be an _enabling_ activity for the > > communities, as without a base set of articles it won't be > > possible to build a > community at > > all. By not paying for new articles, and only translating > well-referenced > > articles, some of the disputes in the communities could be
avoided.
> Perhaps > > we should also identify good source articles, that would be a
help.
> > Translated articles should be above some minimum size, but
they
> > does
not
> > have to be full translations of the source article. > > > > A real problem is that our existing lists of good articles
other
> projects > > should have is pretty much biased towards Western World, so
they
> > need
a
> lot > > of adjustments. Perhaps such a project would identify our
inherit
bias?
> > > > [1] > > https://meta.wikimedia.org/wiki/List_of_articles_every_ > > Wikipedia_should_have > > [2] > > https://meta.wikimedia.org/wiki/List_of_articles_every_ > > Wikipedia_should_have/Expanded > > _______________________________________________ > > Wikimedia-l mailing list, guidelines at: > > https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and > > https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: > > Wikimedia-l@lists.wikimedia.org > > Unsubscribe: > > https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
,
> > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscri
> > be> > > > > > -- > James Heilman > MD, CCFP-EM, Wikipedian > _______________________________________________ > Wikimedia-l mailing list, guidelines at: > https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and > https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: > Wikimedia-l@lists.wikimedia.org > Unsubscribe: > https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, > <mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe
> > >
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
<mailto:wikimedia-l-request@lists.wikimedia.org?subject=
unsubscribe>
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
This email has been checked for viruses by AVG. http://www.avg.com
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
--
--
Scott MacLeod - Founder & President
World University and School
CC World University and School - like CC Wikipedia with best STEM-centric
CC OpenCourseWare - incorporated as a nonprofit university and school in California, and is a U.S. 501 (c) (3) tax-exempt educational organization. _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Well, what I wrote about was translating the articles on the lists at meta. In addition the translators themselves chose which one they want to translate.
On Sat, Feb 24, 2018 at 8:47 PM, Peter Southwood < peter.southwood@telkomsa.net> wrote:
Those who pay get to select what is translated. Cheers, Peter
-----Original Message----- From: Wikimedia-l [mailto:wikimedia-l-bounces@lists.wikimedia.org] On Behalf Of Jean-Philippe Béland Sent: 24 February 2018 16:55 To: Wikimedia Mailing List Subject: Re: [Wikimedia-l] Paid translation
I think the request for such projects should come from the concerned language projects, same for the list of articles. If not, in my simple opinion, it is a form of coloniasm again.
Jean-Philippe Béland Vice President, Wikimedia Canada
On Sat, Feb 24, 2018 at 9:40 AM John Erling Blad jeblad@gmail.com wrote:
Should have added that the remaining points are somewhat less interesting in this context. Preloading a set of articles is a bad idea, the translators should be able to chose for themselves. Articles should also be pretty broad, not very narrow technical or medical, ie vertical articles, as the number of editors that can handle those will
be pretty small.
In particular: Do not believe you can turn a teanslator into a new
editor!
You can although turn an existing editor into a translator.
On Sat, Feb 24, 2018 at 3:34 PM, John Erling Blad jeblad@gmail.com wrote:
- You must start with high quality content and thus all articles
are
extensively improved before being proposed for translation.
Note that to much pressure on "quality" can easily kill the project.
- The "Content Translation" tool developed by the WMF made efforts
more
efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
Didn't mention ContentTranslation, but it should be pretty obvious.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
I used projects below 65k articles as an example, as the chance of competing articles are pretty low.
- With respect to "paying per word" the problem is this would
require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
I'n my original email I wrote "verified good translators". It is as simple as "Has the editor contributed other articles at the project?"
On Sat, Feb 24, 2018 at 2:26 PM, James Heilman jmh649@gmail.com
wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles
are extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP.
Thus we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts
more efficient than handing around word documents. Would love to see that
tool
improved further such as having it support specific lists of articles
that
are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question.
The
issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would
require significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so
languages
in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple
of years. The translators at TWB did not want to become Wikipedians or learn how
to
use our systems. The coordinator created account like TransSW001 (one
for
each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job
of
teaching high school students.
- The project continues but at a slower pace than before. The
Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken
by
40 million people in Eastern India. The amazing thing is that for many
of
these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into
Chinese.
There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad
to
make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles
from
"List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word
(about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that
lacks
good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a
community at
all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be avoided.
Perhaps
we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does
not
have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they need
a
lot
of adjustments. Perhaps such a project would identify our inherit
bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscri be
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
This email has been checked for viruses by AVG. http://www.avg.com
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Thank you James for this detailed feedback. It is very interesting.
JP
On Sat, Feb 24, 2018 at 8:27 AM James Heilman jmh649@gmail.com wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP. Thus we
moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that tool improved further such as having it support specific lists of articles that are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question. The issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so languages in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple of years.
The translators at TWB did not want to become Wikipedians or learn how to use our systems. The coordinator created account like TransSW001 (one for each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job of teaching high school students.
- The project continues but at a slower pace than before. The Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken by 40 million people in Eastern India. The amazing thing is that for many of these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into Chinese. There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad to make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community
at
all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided.
Perhaps
we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a
lot
of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- James Heilman MD, CCFP-EM, Wikipedian _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I think the experience I've had with translating matches up well with the conclusions James has outlined. Even though I'm more likely to translate content into English rather than out of English, the principles still hold.
Trying to produce a translation without quality content in the original article is a frustrating and pointless exercise for the translator. Unless the original meets certain standards, it would be better and easier to write the article from scratch in the "destination" language and translate it back to the "source" language.
Assuming we have a good article in the original language, I definitely encourage translators to use editorial judgment in what they carry over. Focusing on the lead section is one possible approach. In general, because we are trying to translate information and not literature, we should have different priorities. It is more important that the translation maintain fidelity to the facts than to the language and structure of the article. Sometimes it makes sense to pass over certain details, even a beginning-to-end translation might come out a bit condensed. As one reason for this, making some details accessible to the cultural audience in the new language can at times require a fair amount of elaboration, more than may be ideal for the context under discussion. The best approach to use is one of adaptation as much as translation.
I don't have strong feelings about whether a paid model will work, or work better than purely volunteer activity, but I would be open to seeing a trial. The essential thing is that we find translators who can understand and apply standards of quality in their work, much like we would expect if they were editors writing entirely new articles.
--Michael Snow
On 2/24/2018 5:26 AM, James Heilman wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP. Thus we
moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that tool improved further such as having it support specific lists of articles that are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question. The issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so languages in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple of years.
The translators at TWB did not want to become Wikipedians or learn how to use our systems. The coordinator created account like TransSW001 (one for each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job of teaching high school students.
- The project continues but at a slower pace than before. The Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken by 40 million people in Eastern India. The amazing thing is that for many of these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into Chinese. There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad to make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
The source article should meet certain standards, but do not fall in the trap where the translated articles must themselves be better than some imagined standard. That would lead to a defunc process.
On Sat, Feb 24, 2018 at 8:41 PM, Michael Snow wikipedia@frontier.com wrote:
I think the experience I've had with translating matches up well with the conclusions James has outlined. Even though I'm more likely to translate content into English rather than out of English, the principles still hold.
Trying to produce a translation without quality content in the original article is a frustrating and pointless exercise for the translator. Unless the original meets certain standards, it would be better and easier to write the article from scratch in the "destination" language and translate it back to the "source" language.
Assuming we have a good article in the original language, I definitely encourage translators to use editorial judgment in what they carry over. Focusing on the lead section is one possible approach. In general, because we are trying to translate information and not literature, we should have different priorities. It is more important that the translation maintain fidelity to the facts than to the language and structure of the article. Sometimes it makes sense to pass over certain details, even a beginning-to-end translation might come out a bit condensed. As one reason for this, making some details accessible to the cultural audience in the new language can at times require a fair amount of elaboration, more than may be ideal for the context under discussion. The best approach to use is one of adaptation as much as translation.
I don't have strong feelings about whether a paid model will work, or work better than purely volunteer activity, but I would be open to seeing a trial. The essential thing is that we find translators who can understand and apply standards of quality in their work, much like we would expect if they were editors writing entirely new articles.
--Michael Snow
On 2/24/2018 5:26 AM, James Heilman wrote:
We learned a few things during the medical translation project which started back in 2011:
- You must start with high quality content and thus all articles are
extensively improved before being proposed for translation.
- A lot of languages want "less" content than is present on EN WP. Thus
we moved to just improving and suggesting for translation the leads of the English articles.
- The "Content Translation" tool developed by the WMF made efforts more
efficient than handing around word documents. Would love to see that tool improved further such as having it support specific lists of articles that are deemed ready for translation by certain groups. Would also love the tool to have tracking metrics for these types of projects.
- We used volunteer translators mostly associated with our partner
Translators Without Borders. One issue we found was that languages in which their are lots of translators such as French, Spanish, and Italian there is often already at least some content on many of the topics in question. The issue than becomes integration which needs an expert Wikipedia. And for languages in which we have little content there are often few avaliable volunteers.
- With respect to "paying per word" the problem is this would require
significant checks and balances to make sure people are taking the work seriously and not simple using Google translate for the 70 or so languages in which it claims to work. We often had translations undergo a second review and the volunteers at TWB have to pass certain tests to be accepted.
- I hired a coordinator for the translation project for a couple of
years. The translators at TWB did not want to become Wikipedians or learn how to use our systems. The coordinator created account like TransSW001 (one for each volunteer) and preloaded the article to be translated into Content Translation. They than gave the volunteer translator the user name and password to the account.
- Were are we at now? There are currently just over 1,000 leads of
articles that have been improved and are ready for translation. This includes articles on the 440 medications that are on the WHO Essential List. We have worked a bit in some 100 languages. The efforts have resulted in more than 5 million works translated and integrated into different Wikipedias. The coordinator has unfortunately moved on to his real job of teaching high school students.
- The project continues but at a slower pace than before. The Wikipedian
and retired orthopedic surgeon Subas Chandra Rout has basically single handedly translated nearly all 1,000 leads into Odia a language spoken by 40 million people in Eastern India. The amazing thing is that for many of these topics this is the first and only information online about it. Google translate does not even claim to work in this language. Our partnerships with WMTW and medical school in Taipai continue to translate into Chinese. There the students translate and than their translations are reviewed by their profs before being posted. They translate in groups using hackpad to make it more social.
I am currently working to re invigorate the project :-) James
On Sat, Feb 24, 2018 at 5:51 AM, John Erling Blad jeblad@gmail.com wrote:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
On 2/24/2018 1:53 PM, John Erling Blad wrote:
The source article should meet certain standards, but do not fall in the trap where the translated articles must themselves be better than some imagined standard. That would lead to a defunc process.
I'm not saying a translated article must be flawless. But certainly if we were paying for translations, it would be appropriate to have some level of expectations for the quality of the result. With volunteers, any honest effort is encouraged, although if the quality is low enough to be worse than the alternative, they can be encouraged to redirect that effort more productively.
--Michael Snow
I'll start by saying that I'm one of the developers of Content Translation, so I'm obviously biased about this topic.
A lot of good points were raised here, but there's one that is not really mentioned. If it sounds obvious to you, it's great, but it's not obvious to everyone. Here it is:
More successful Wikipedia projects tend to be in languages in which there is an established history and tradition of: * elementary and higher education where teachers and professors speak to students in that language, and in which students write papers in that language * publishing textbooks * publishing encyclopedias * publishing dictionaries * translating works from (any) other languages, both fiction and reference
People who can read in these developed languages should remember this privilege that they have: English, French, Russian, Spanish, German, Polish, Italian, Dutch, Czech, Japanese, Norwegian, Hebrew and a few other well-developed Wikipedias are written in languages in which good encyclopedias had already existed before Wikipedia came along. A Wikipedia in these languages didn't make encyclopedic knowledge available in these languages; it made encyclopedic knowledge *more easily* available in them.
There are many other things that (probably) affect the development of a Wikipedia, such as web connectivity; speakers' population; speakers' attitude to the language; work week length (and the remaining free time); volunteering culture (or lack thereof); support of common operating systems for the language; economic indicators like GDP and HDI in the countries where the language is spoken; etc. I'm not aware of research that checks the correlation between these aspects and the development of a Wikipedia project in a language, but I strongly suspect that it exists for at least some of the above. (If anybody reading this is aware of such research, I'll be very happy to read it.)
But it's important to go back to the first point here: The existence of previous encyclopedias makes it easier for writers in these languages to simply start writing. "An encyclopedia" is not a new concept for them. The culture around these languages already had well-developed scientific terminology and a language style.
When I speak to people who write in Wikipedia in languages of India, Philippines, and other developing countries, they complain about different things from people that write in European languages. For example, they very often complain about the difficulty of writing in an encyclopedic style and bridging the colloquial language that common people can read and the standardized versions of the respective languages. This makes me think that they were standardized in a way that is problematic for *actually* writing an encyclopedia that would be useful to the general public.
A *massive* project for writing in a language, would create a critical mass of people who would either make the general public accustomed to reading in this standard language, or create a new de facto standard. But I guess that none of the current Wikipedia projects in these languages have this critical mass of writers.
A translation project, such as what Jon Erling Blad and Lane Rasberry are suggesting in this thread *may* create such a critical mass. It also needs bold leaders, who will take it upon themselves Languages that are developed today went through periods of directed development in the past; Lomonosov did it for Russian, Diderot did it for French, and so on. This can happen today as well. (English went through this, too, although I'm not sure which person should be tied to it: Isaac Newton? Samuel Johnson? John Harris (Q562265)? Alfred the Great? Probably all of them to some degree.)
I'd even go further and say that I don't agree with Lane when he says that the WMF cannot and will never pay for content. It sounds like a given thing to some people, but it isn't. Quite the contrary; it's imaginable that a careful and thoughtful project of this nature can be carried out by the WMF itself. "WMF never does this" is not a rule, and it must not be a mental blocker. I increasingly feel that the WMF is gradually, increasingly understanding that different languages need different kinds of resources and support, and this may include paid content creation. (Before you jump to conclusions: I'm a WMF staff member, but please don't understand from this that I know about some internal project to do such a thing, or that I am suggesting to do this. Neither thing is true. I'm just writing a sincere stream of consciousness about my opinions and feelings, and I might be wrong about it all.)
That said, it does make more sense to me that organizations other than the WMF should lead such work, perhaps with some WMF funding, for the sake of thought diversity if for nothing else. But whether it's paid for by the WMF directly, by Wikimedia chapters, by thematic interest groups, or by somebody else is not the main issue. What is important, is that *local* people and native speakers are as involved as possible in the content creation, and that the list of topics to be translated is not too strongly dictated.
(I also like the suggestion of translating from different languages. For practical reasons, English is the most common translation source [1], but translating from French, Russian, Chinese, or other languages, is awesome for diversity—not just politically, but philosophically as well.)
[1] https://en.wikipedia.org/wiki/Special:CXStats
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore
2018-02-24 14:51 GMT+02:00 John Erling Blad jeblad@gmail.com:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Some years ago I tried to figure out whether there was some kind of mechanism that kept the community sizes at a fixed level. Taking the population in countries that spoke a specific language, adjusting for access to internet, and family sizes, made me realize that most stable projects have 0.2–0.4 ‰ contributors within a normalized language group. If you then say a stable community consists of 10-20 users, then this creates a pretty hard limit on the language size. A community of 10-20 users would imply a language size of 250–1000 000 people. That makes a Wikipedia-project out of reach for a lot of languages.
(If you somehow limit this group, for example by demanding that only trained medics should write medical articles, then you draw those medics from the already limited set, in effect demanding an even larger language group to make a working community.)
To create the initial interest to make a core encyclopedia is even more difficult. To bridge that gap, and initiate building of a sustainable community, a core set of articles are necessary. Perhaps that core set will be short-lived, and will be replaced with articles that somehow better reflects the user basis at the local language, but a core set that reflects some common ground is nonetheless necessary. The capitol of Sweden doesn't magically disappears in Bengali. The moon doesn't magically turns into cheese unless in a fairy tale. There are some universal constants that all languages must adhere to, even in Sicilian Wikipedia a mafioso is a mobster [1] (someone have messed up the interlinking)
In Norwegian Bokmål we have a few users that has this kind of weird idea that if something lacks an explicit name, either a word or phrase, then it should not be described. In my opinion that is nonsense. Some kind of entity can be described, in any language, no matter if it has a name. We describe the World as we know it, using words or phrases from the language to do so. If what we describe has a name, then we use that name. In some languages that means describing a specific entity is difficult because the local language has many words and phrases for the same thing. In some other language it might be difficult because there are no word or phrases to describe the entity. Neither of those problems arise because the entity is non-existing in our world, it is just difficult to describe in the specific language.
Give people knowledge! If they need to somehow clarify that knowledge to make it more accessible to them, then let them do that! That is why Wikipedia is editable for everyone!
[1] https://scn.wikipedia.org/wiki/Mafiusu
On Sun, Feb 25, 2018 at 2:07 PM, Amir E. Aharoni < amir.aharoni@mail.huji.ac.il> wrote:
I'll start by saying that I'm one of the developers of Content Translation, so I'm obviously biased about this topic.
A lot of good points were raised here, but there's one that is not really mentioned. If it sounds obvious to you, it's great, but it's not obvious to everyone. Here it is:
More successful Wikipedia projects tend to be in languages in which there is an established history and tradition of:
- elementary and higher education where teachers and professors speak to
students in that language, and in which students write papers in that language
- publishing textbooks
- publishing encyclopedias
- publishing dictionaries
- translating works from (any) other languages, both fiction and reference
People who can read in these developed languages should remember this privilege that they have: English, French, Russian, Spanish, German, Polish, Italian, Dutch, Czech, Japanese, Norwegian, Hebrew and a few other well-developed Wikipedias are written in languages in which good encyclopedias had already existed before Wikipedia came along. A Wikipedia in these languages didn't make encyclopedic knowledge available in these languages; it made encyclopedic knowledge *more easily* available in them.
There are many other things that (probably) affect the development of a Wikipedia, such as web connectivity; speakers' population; speakers' attitude to the language; work week length (and the remaining free time); volunteering culture (or lack thereof); support of common operating systems for the language; economic indicators like GDP and HDI in the countries where the language is spoken; etc. I'm not aware of research that checks the correlation between these aspects and the development of a Wikipedia project in a language, but I strongly suspect that it exists for at least some of the above. (If anybody reading this is aware of such research, I'll be very happy to read it.)
But it's important to go back to the first point here: The existence of previous encyclopedias makes it easier for writers in these languages to simply start writing. "An encyclopedia" is not a new concept for them. The culture around these languages already had well-developed scientific terminology and a language style.
When I speak to people who write in Wikipedia in languages of India, Philippines, and other developing countries, they complain about different things from people that write in European languages. For example, they very often complain about the difficulty of writing in an encyclopedic style and bridging the colloquial language that common people can read and the standardized versions of the respective languages. This makes me think that they were standardized in a way that is problematic for *actually* writing an encyclopedia that would be useful to the general public.
A *massive* project for writing in a language, would create a critical mass of people who would either make the general public accustomed to reading in this standard language, or create a new de facto standard. But I guess that none of the current Wikipedia projects in these languages have this critical mass of writers.
A translation project, such as what Jon Erling Blad and Lane Rasberry are suggesting in this thread *may* create such a critical mass. It also needs bold leaders, who will take it upon themselves Languages that are developed today went through periods of directed development in the past; Lomonosov did it for Russian, Diderot did it for French, and so on. This can happen today as well. (English went through this, too, although I'm not sure which person should be tied to it: Isaac Newton? Samuel Johnson? John Harris (Q562265)? Alfred the Great? Probably all of them to some degree.)
I'd even go further and say that I don't agree with Lane when he says that the WMF cannot and will never pay for content. It sounds like a given thing to some people, but it isn't. Quite the contrary; it's imaginable that a careful and thoughtful project of this nature can be carried out by the WMF itself. "WMF never does this" is not a rule, and it must not be a mental blocker. I increasingly feel that the WMF is gradually, increasingly understanding that different languages need different kinds of resources and support, and this may include paid content creation. (Before you jump to conclusions: I'm a WMF staff member, but please don't understand from this that I know about some internal project to do such a thing, or that I am suggesting to do this. Neither thing is true. I'm just writing a sincere stream of consciousness about my opinions and feelings, and I might be wrong about it all.)
That said, it does make more sense to me that organizations other than the WMF should lead such work, perhaps with some WMF funding, for the sake of thought diversity if for nothing else. But whether it's paid for by the WMF directly, by Wikimedia chapters, by thematic interest groups, or by somebody else is not the main issue. What is important, is that *local* people and native speakers are as involved as possible in the content creation, and that the list of topics to be translated is not too strongly dictated.
(I also like the suggestion of translating from different languages. For practical reasons, English is the most common translation source [1], but translating from French, Russian, Chinese, or other languages, is awesome for diversity—not just politically, but philosophically as well.)
[1] https://en.wikipedia.org/wiki/Special:CXStats
-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com “We're living in pieces, I want to live in peace.” – T. Moore
2018-02-24 14:51 GMT+02:00 John Erling Blad jeblad@gmail.com:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community
at
all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided.
Perhaps
we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a
lot
of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_ Wikipedia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I am very happy to follow this thread as I believe it is addressing a very relevant issue.
In my mind we can divide up the different language version into 5 categories:
1.Enwp,
2.the next 6-7 (de,fr, es,jp,pt,ru..)
3.the next 20 or so, where the basic workprocesses are applied
4.the next 40-50 which are struggling to generate more input then what is vandalised
5.the rest which in reality is no viable online encyclopedias
And for me no 1 priority is to accept that there are these categories, and that what is applicable for cat 1 and 2 is not so for 4 and 5.
I believe the grant model could easily make room for subsiding good initiatives addressing the problem for cat 4 and 5 (and perhaps 3).
And I think it is very presumptuous to start talking of what technique to use and things like translation. If we open up for creative brainstorming (among the ones having the need) I think very many other ways can turn up. Myself I am deeply impressed what you can create using Wikidata as a base source of info, and being from a version of type 3 I see how much my homeversion improve content with wikidata created infoboxes
Anders
Den 2018-02-24 kl. 13:51, skrev John Erling Blad:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikipedia_should_have... _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I should have joined in this discussion a little earlier. I work a lot with the French Wikipedia, and we do not just translate articles from English (6 million articles) to French (only 2 million articles). The French community is large and active, and provide a unique local perspective on the different articles that are written. And when I say local, I mean that things are seen differently in France than in the French speaking part of Switzerland or Belgium.
I think that we are ignoring something very important here: putting it simply, Wikipedia contributors do two things. They add information to the encyclopedia by improving articles or writing new ones, and they curate or check the existing articles. All this talk about machine translation does not address the second aspect of what the volunteer contributors do. This means that we could have hundreds of thousands of articles in a language with very few active contributors. Will that small community be able to oversee so many articles ?
For example, have a look at the list of Wikipedias ordered by number of articles: 1. English - 5,578,081 articles - 138,479 active users - 1,230 admins 2. Cebuano - 5,383,108 articles - 162 active users - 5 admins 3. Swedish - 3,784,331 articles - 2,929 active users - 65 admins 4. German - 2,157,495 articles - 20, 085 active users - 194 admins
When I have some time, I will look into different ratios like number or articles/active users or number of articles/number of native language speakers... Now I am not saying that our Swedish friends have abused machine translation of articles, but I definetly that something is not quite right about the Cebuano wiki... Gabe
On Sun, Feb 25, 2018 at 4:06 PM, Anders Wennersten <mail@anderswennersten.se
wrote:
I am very happy to follow this thread as I believe it is addressing a very relevant issue.
In my mind we can divide up the different language version into 5 categories:
1.Enwp,
2.the next 6-7 (de,fr, es,jp,pt,ru..)
3.the next 20 or so, where the basic workprocesses are applied
4.the next 40-50 which are struggling to generate more input then what is vandalised
5.the rest which in reality is no viable online encyclopedias
And for me no 1 priority is to accept that there are these categories, and that what is applicable for cat 1 and 2 is not so for 4 and 5.
I believe the grant model could easily make room for subsiding good initiatives addressing the problem for cat 4 and 5 (and perhaps 3).
And I think it is very presumptuous to start talking of what technique to use and things like translation. If we open up for creative brainstorming (among the ones having the need) I think very many other ways can turn up. Myself I am deeply impressed what you can create using Wikidata as a base source of info, and being from a version of type 3 I see how much my homeversion improve content with wikidata created infoboxes
Anders
Den 2018-02-24 kl. 13:51, skrev John Erling Blad:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I wonder if creating dynamic articles from Wikidata is better than creating static articles. Because we lack tools for this, it is easier to do this offline, and as a consequence we get the static bot-articles.
Den søn. 25. feb. 2018, 16.26 skrev Gabriel Thullen gabriel@thullen.com:
I should have joined in this discussion a little earlier. I work a lot with the French Wikipedia, and we do not just translate articles from English (6 million articles) to French (only 2 million articles). The French community is large and active, and provide a unique local perspective on the different articles that are written. And when I say local, I mean that things are seen differently in France than in the French speaking part of Switzerland or Belgium.
I think that we are ignoring something very important here: putting it simply, Wikipedia contributors do two things. They add information to the encyclopedia by improving articles or writing new ones, and they curate or check the existing articles. All this talk about machine translation does not address the second aspect of what the volunteer contributors do. This means that we could have hundreds of thousands of articles in a language with very few active contributors. Will that small community be able to oversee so many articles ?
For example, have a look at the list of Wikipedias ordered by number of articles:
- English - 5,578,081 articles - 138,479 active users - 1,230 admins
- Cebuano - 5,383,108 articles - 162 active users - 5 admins
- Swedish - 3,784,331 articles - 2,929 active users - 65 admins
- German - 2,157,495 articles - 20, 085 active users - 194 admins
When I have some time, I will look into different ratios like number or articles/active users or number of articles/number of native language speakers... Now I am not saying that our Swedish friends have abused machine translation of articles, but I definetly that something is not quite right about the Cebuano wiki... Gabe
On Sun, Feb 25, 2018 at 4:06 PM, Anders Wennersten < mail@anderswennersten.se
wrote:
I am very happy to follow this thread as I believe it is addressing a
very
relevant issue.
In my mind we can divide up the different language version into 5 categories:
1.Enwp,
2.the next 6-7 (de,fr, es,jp,pt,ru..)
3.the next 20 or so, where the basic workprocesses are applied
4.the next 40-50 which are struggling to generate more input then what is vandalised
5.the rest which in reality is no viable online encyclopedias
And for me no 1 priority is to accept that there are these categories,
and
that what is applicable for cat 1 and 2 is not so for 4 and 5.
I believe the grant model could easily make room for subsiding good initiatives addressing the problem for cat 4 and 5 (and perhaps 3).
And I think it is very presumptuous to start talking of what technique to use and things like translation. If we open up for creative brainstorming (among the ones having the need) I think very many other ways can turn
up.
Myself I am deeply impressed what you can create using Wikidata as a base source of info, and being from a version of type 3 I see how much my homeversion improve content with wikidata created infoboxes
Anders
Den 2018-02-24 kl. 13:51, skrev John Erling Blad:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
wonder if creating dynamic articles from Wikidata is better than creating static articles
Not for years to decades.
https://twitter.com/AustenAllred/status/967842020151603200
On Mon, Feb 26, 2018 at 3:02 AM, John Erling Blad jeblad@gmail.com wrote:
I wonder if creating dynamic articles from Wikidata is better than creating static articles. Because we lack tools for this, it is easier to do this offline, and as a consequence we get the static bot-articles.
Den søn. 25. feb. 2018, 16.26 skrev Gabriel Thullen gabriel@thullen.com:
I should have joined in this discussion a little earlier. I work a lot with the French Wikipedia, and we do not just translate articles from English (6 million articles) to French (only 2 million articles). The French community is large and active, and provide a unique local perspective on the different articles that are written. And when I say local, I mean that things are seen differently in France than in the French speaking part of Switzerland or Belgium.
I think that we are ignoring something very important here: putting it simply, Wikipedia contributors do two things. They add information to the encyclopedia by improving articles or writing new ones, and they curate or check the existing articles. All this talk about machine translation does not address the second aspect of what the volunteer contributors do. This means that we could have hundreds of thousands of articles in a language with very few active contributors. Will that small community be able to oversee so many articles ?
For example, have a look at the list of Wikipedias ordered by number of articles:
- English - 5,578,081 articles - 138,479 active users - 1,230 admins
- Cebuano - 5,383,108 articles - 162 active users - 5 admins
- Swedish - 3,784,331 articles - 2,929 active users - 65 admins
- German - 2,157,495 articles - 20, 085 active users - 194 admins
When I have some time, I will look into different ratios like number or articles/active users or number of articles/number of native language speakers... Now I am not saying that our Swedish friends have abused machine translation of articles, but I definetly that something is not quite right about the Cebuano wiki... Gabe
On Sun, Feb 25, 2018 at 4:06 PM, Anders Wennersten < mail@anderswennersten.se
wrote:
I am very happy to follow this thread as I believe it is addressing a
very
relevant issue.
In my mind we can divide up the different language version into 5 categories:
1.Enwp,
2.the next 6-7 (de,fr, es,jp,pt,ru..)
3.the next 20 or so, where the basic workprocesses are applied
4.the next 40-50 which are struggling to generate more input then what is vandalised
5.the rest which in reality is no viable online encyclopedias
And for me no 1 priority is to accept that there are these categories,
and
that what is applicable for cat 1 and 2 is not so for 4 and 5.
I believe the grant model could easily make room for subsiding good initiatives addressing the problem for cat 4 and 5 (and perhaps 3).
And I think it is very presumptuous to start talking of what technique to use and things like translation. If we open up for creative brainstorming (among the ones having the need) I think very many other ways can turn
up.
Myself I am deeply impressed what you can create using Wikidata as a base source of info, and being from a version of type 3 I see how much my homeversion improve content with wikidata created infoboxes
Anders
Den 2018-02-24 kl. 13:51, skrev John Erling Blad:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There
are
several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Hoi, The Cebuano Wikipedia articles were created based on information available in databases. So creating static articles is a known quantity. In Reasonator there is functionality that creates text for humans. This has been available for years as well and when data changes, the text changes.
Consequently both static and dynamic texts based on data has been with us for years. It is only in the opposition by some that we have not served the data that is available to us as information for those who seek knowledge. Technically there is nothing that stops us. Thanks, GerardM
On 26 February 2018 at 12:50, James Salsman jsalsman@gmail.com wrote:
wonder if creating dynamic articles from Wikidata is better than creating static articles
Not for years to decades.
https://twitter.com/AustenAllred/status/967842020151603200
On Mon, Feb 26, 2018 at 3:02 AM, John Erling Blad jeblad@gmail.com wrote:
I wonder if creating dynamic articles from Wikidata is better than
creating
static articles. Because we lack tools for this, it is easier to do this offline, and as a consequence we get the static bot-articles.
Den søn. 25. feb. 2018, 16.26 skrev Gabriel Thullen <gabriel@thullen.com :
I should have joined in this discussion a little earlier. I work a lot
with
the French Wikipedia, and we do not just translate articles from
English (6
million articles) to French (only 2 million articles). The French
community
is large and active, and provide a unique local perspective on the different articles that are written. And when I say local, I mean that things are seen differently in France than in the French speaking part
of
Switzerland or Belgium.
I think that we are ignoring something very important here: putting it simply, Wikipedia contributors do two things. They add information to
the
encyclopedia by improving articles or writing new ones, and they curate
or
check the existing articles. All this talk about machine translation
does
not address the second aspect of what the volunteer contributors do. This means that we could have hundreds of thousands of articles in a language with very few active contributors. Will that small community
be
able to oversee so many articles ?
For example, have a look at the list of Wikipedias ordered by number of articles:
- English - 5,578,081 articles - 138,479 active users - 1,230 admins
- Cebuano - 5,383,108 articles - 162 active users - 5 admins
- Swedish - 3,784,331 articles - 2,929 active users - 65 admins
- German - 2,157,495 articles - 20, 085 active users - 194 admins
When I have some time, I will look into different ratios like number or articles/active users or number of articles/number of native language speakers... Now I am not saying that our Swedish friends have abused machine translation of articles, but I definetly that something is not quite right about the Cebuano wiki... Gabe
On Sun, Feb 25, 2018 at 4:06 PM, Anders Wennersten < mail@anderswennersten.se
wrote:
I am very happy to follow this thread as I believe it is addressing a
very
relevant issue.
In my mind we can divide up the different language version into 5 categories:
1.Enwp,
2.the next 6-7 (de,fr, es,jp,pt,ru..)
3.the next 20 or so, where the basic workprocesses are applied
4.the next 40-50 which are struggling to generate more input then
what is
vandalised
5.the rest which in reality is no viable online encyclopedias
And for me no 1 priority is to accept that there are these categories,
and
that what is applicable for cat 1 and 2 is not so for 4 and 5.
I believe the grant model could easily make room for subsiding good initiatives addressing the problem for cat 4 and 5 (and perhaps 3).
And I think it is very presumptuous to start talking of what
technique to
use and things like translation. If we open up for creative
brainstorming
(among the ones having the need) I think very many other ways can turn
up.
Myself I am deeply impressed what you can create using Wikidata as a
base
source of info, and being from a version of type 3 I see how much my homeversion improve content with wikidata created infoboxes
Anders
Den 2018-02-24 kl. 13:51, skrev John Erling Blad:
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k
articles,
the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators?
There
are
several lists of such base sets. We have both the thousand articles
from
"List of articles every Wikipedia should have"[1] and and the ten
thousand
articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word
(about
$1
for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that
lacks
good translation tools.
I believe this would be an _enabling_ activity for the communities,
as
without a base set of articles it won't be possible to build a
community
at all. By not paying for new articles, and only translating
well-referenced
articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does
not
have to be full translations of the source article.
A real problem is that our existing lists of good articles other
projects
should have is pretty much biased towards Western World, so they
need a
lot of adjustments. Perhaps such a project would identify our inherit
bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wik
i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/
mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at:
https://meta.wikimedia.org/wik
i/Mailing_lists/Guidelines and
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l
,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/ wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/ wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I'm not against the idea of paid translation /per se/, but it shouldn't be managed by the WMF, should it be only to ensure that it doesn't cross too far the line of non-intervention regarding editorial decisions.
Debate can go on to which level it stands with this line, but to my mind WMF always have been mainly about hosting works, not about what will be published by who under which (non-)remunerated conditions. I think that it is important that it stay so for example due to legal reasons regarding responsibility of what is stated in this works.
From this perspective, it would be probably better to have locale collective initiatives which decide what seems the more important to be translated and means to achieve them, should it be through paid editing with money coming from the said collective itself. Directly financing that kind of initiative would blur the line of the hosting position I think. But giving visibility to this kind of locale fund raising initiatives could be a donation in kind that would be maybe less problematic, wouldn't it?
Le 24/02/2018 à 13:51, John Erling Blad a écrit :
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikipedia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikipedia_should_have... _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Agree with mathieo it needs to be something driven by the receiving language community with WMF support rather than something being pushed in to communities from the WMF or other projects. Such if the Swahili community thought that having say medical articles translated was something it felt was needed then the WMF could support by assisting with tools, and facilities to make it happen.
On 27 February 2018 at 18:42, mathieu stumpf guntz < psychoslave@culture-libre.org> wrote:
I'm not against the idea of paid translation /per se/, but it shouldn't be managed by the WMF, should it be only to ensure that it doesn't cross too far the line of non-intervention regarding editorial decisions.
Debate can go on to which level it stands with this line, but to my mind WMF always have been mainly about hosting works, not about what will be published by who under which (non-)remunerated conditions. I think that it is important that it stay so for example due to legal reasons regarding responsibility of what is stated in this works.
From this perspective, it would be probably better to have locale collective initiatives which decide what seems the more important to be translated and means to achieve them, should it be through paid editing with money coming from the said collective itself. Directly financing that kind of initiative would blur the line of the hosting position I think. But giving visibility to this kind of locale fund raising initiatives could be a donation in kind that would be maybe less problematic, wouldn't it?
Le 24/02/2018 à 13:51, John Erling Blad a écrit :
This discussion is going to be fun! =D
A little more than seventy Wikipedia-projects has more than 65k articles, the remaining two hundred or so are pretty small.
What if a base set of articles were opened for paid translators? There are several lists of such base sets. We have both the thousand articles from "List of articles every Wikipedia should have"[1] and and the ten thousand articles from the expanded list[2].
Lets say verified good translators was paid about $0.01 per word (about $1 for a 1k-article) for translating one of those articles into another language, with perhaps a higher pay for contributors in high-cost countries. The pay would also have to be higher for languages that lacks good translation tools.
I believe this would be an _enabling_ activity for the communities, as without a base set of articles it won't be possible to build a community at all. By not paying for new articles, and only translating well-referenced articles, some of the disputes in the communities could be avoided. Perhaps we should also identify good source articles, that would be a help. Translated articles should be above some minimum size, but they does not have to be full translations of the source article.
A real problem is that our existing lists of good articles other projects should have is pretty much biased towards Western World, so they need a lot of adjustments. Perhaps such a project would identify our inherit bias?
[1] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have [2] https://meta.wikimedia.org/wiki/List_of_articles_every_Wikip edia_should_have/Expanded _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wik i/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wik i/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
wikimedia-l@lists.wikimedia.org