I happen to agree with Gerard that the Wikimedia Foundation does indeed support languages through the very act of supporting Wikipedias. I also agree that the foundation should be more proud of this fact and announce it happily. If we are doing something so positive, why not let people know about it, and why not even include it as one of our goals? It may not be our primary goal, but it certainly plays a supporting role towards reaching that goal ("the sum of all knowledge"). It is yet another gem in the WMF's crown.
That being said, I don't really understand what, specifically, Gerard is asking for. In what specific ways does the foundation need to "adapt" in order to meet the needs of smaller languages? "Resources" and "investments" usually imply money, but in what specific ways does Gerard think money can be used to support smaller languages? Please elaborate.
Finally, Mark Williamson's comment:
I imagine it would be easier if those people would be allowed to get Wikipedias instead of having to wait for years.
This is as opposed to Gerard: "we insist on a good user experience so a localised user interface is a must."
Experience proves that Mark is right: Many of today's successful Wikipedia's began without much of a local interface at all. Instead, people just got to work, and the interface got translated bit-by-bit along the way by admins, in exactly the same way a Wikipedia gets built as a whole by editors.
Not allowing people to simply "get to work" on a Wikipedia, and requiring that there be "several hoops to jump through before a new language gets its own project" actually works against smaller languages.
Instead, just let people get started! If the project fails, simply freeze it until speakers of the language come along who want to try to build it again.
To conclude, if there are positive ways that money could be used to support smaller languages, then the foundation should consider doing so. But regardless of monetary issues, people should be allowed to get to work on smaller wikipedias without too many hurdles.
Dovi
__________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com
Hoi, Resources and investments also imply labour. There are several instances where the choice for spending time of our paid for programmers is dedicated to things that are nice to have. Often the choice is made because of what is seen as of a big relevance to the "big" projects ie languages. Even though MediaWiki is exceptional in the amount of languages we support, the fact that a fifth of the languages that have a Wikipedia indicates that contrary what we say MediaWiki does not support 250 languages. People invest in their Wikipedia and we do not cherish this investment by ensuring that the localisation is seen as vital.
It is wrong to suggest that the User Interface is not a vital component of a successful Wikipedia. With some regularity projects are voted to be closed down. They are typically programs that do not have a good localisation, they are typically programs that do not have a community. They are typically Wikipedias that have been started prematurely. Yes, there are Wikipedias that are doing fine, the issue is that many people are upset by Wikipedias failing. We lose support for adding new languages with more projects failing.
What we aim for in new Wikipedias is project that provide good information to the people. We should not expect people to read any other language. A Wikipedia is there to provide people with knowledge and for this an appropriate user interface is essential. As long as the User Interface is not localised it is not yet ready to go life for the people to find information in their language.
Thanks, GerardM
On 10/26/07, Dovi Jacobs dovijacobs@yahoo.com wrote:
I happen to agree with Gerard that the Wikimedia Foundation does indeed support languages through the very act of supporting Wikipedias. I also agree that the foundation should be more proud of this fact and announce it happily. If we are doing something so positive, why not let people know about it, and why not even include it as one of our goals? It may not be our primary goal, but it certainly plays a supporting role towards reaching that goal ("the sum of all knowledge"). It is yet another gem in the WMF's crown.
That being said, I don't really understand what, specifically, Gerard is asking for. In what specific ways does the foundation need to "adapt" in order to meet the needs of smaller languages? "Resources" and "investments" usually imply money, but in what specific ways does Gerard think money can be used to support smaller languages? Please elaborate.
Finally, Mark Williamson's comment:
I imagine it would be easier if those people would be allowed to get
Wikipedias instead of having to wait for years.
This is as opposed to Gerard: "we insist on a good user experience so a localised user interface is a must."
Experience proves that Mark is right: Many of today's successful Wikipedia's began without much of a local interface at all. Instead, people just got to work, and the interface got translated bit-by-bit along the way by admins, in exactly the same way a Wikipedia gets built as a whole by editors.
Not allowing people to simply "get to work" on a Wikipedia, and requiring that there be "several hoops to jump through before a new language gets its own project" actually works against smaller languages.
Instead, just let people get started! If the project fails, simply freeze it until speakers of the language come along who want to try to build it again.
To conclude, if there are positive ways that money could be used to support smaller languages, then the foundation should consider doing so. But regardless of monetary issues, people should be allowed to get to work on smaller wikipedias without too many hurdles.
Dovi
Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: http://lists.wikimedia.org/mailman/listinfo/foundation-l
GerardM ha scritto:
It is wrong to suggest that the User Interface is not a vital component of a successful Wikipedia. With some regularity projects are voted to be closed down. They are typically programs that do not have a good localisation, they are typically programs that do not have a community. They are typically Wikipedias that have been started prematurely. Yes, there are Wikipedias that are doing fine, the issue is that many people are upset by Wikipedias failing. We lose support for adding new languages with more projects failing.
What we aim for in new Wikipedias is project that provide good information to the people. We should not expect people to read any other language. A Wikipedia is there to provide people with knowledge and for this an appropriate user interface is essential. As long as the User Interface is not localised it is not yet ready to go life for the people to find information in their language.
I'm sorry but I cannot fully agree on these points. While I agree that it is important that a user can have the interface translated in his/her own language, this does not mean that such language and the project's language are the same one. For example, Latin or Esperanto have basically no native speaker, and editors may be happier having their own mother tongue for the interface. It becomes even more absurd for those languages which do not have a standard koiné, but only dialects. This is the case for most of the regional languages of Italy, such as Lombard or Emilian-Romagnan. These project have 1 wikipedia each because there is 1 ISO-code for the language, but in many cases articles have more than one version, to cover the different dialects. But they can (actually, must - see above) have only one localisation, which of course for many users will be in a different <strike>language</strike> variant of the language. Ok, dialects are generally similar enough to be mutually comprehensive, but it will still be quite awkward to have an interface in what looks like a funny version of your language (to make an example, it would be like having the interface in English with Scots spelling. And of course, there are only very few native speakers of those languages which are not bilingual and speaking at least another language (not necessarily Italian), and who may prefer to use the interface in the other mother tongue; very few of the non bilingual people would anyway have access to a computer.
Cruccone
On Saturday 27 October 2007 20:01, Marco Chiesa wrote:
mother tongue for the interface. It becomes even more absurd for those languages which do not have a standard koiné, but only dialects. This is the case for most of the regional languages of Italy, such as Lombard or Emilian-Romagnan. These project have 1 wikipedia each because there is 1 ISO-code for the language, but in many cases articles have more than one version, to cover the different dialects. But they can (actually, must - see above) have only one localisation, which of course for many users will be in a different <strike>language</strike> variant of the
They in fact can have more localisations, like Chinese Wikipedia does. And of course logged in users can always change their localisation.
On 10/28/07, GerardM gerard.meijssen@gmail.com wrote:
Resources and investments also imply labour.
...
As long as the User Interface is not localised it is not yet ready to go life for the people to find information in their language.
I don't understand what your argument is. You seem to be talking about priorities in MediaWiki development, but then you talk about interface localisation - which is a matter for the project participants who speak the language and can actually translate the interface, not for Foundation people.
Hoi, When you consider the localisation of messages, the messages can be divided in two. There are messages that are project specific and there are messages that have a global reach. When people localise messages, we do not ensure that the effort on the global messages has a global effect. The best place at this moment to localise messages is in the BetaWiki. A project that should be a Wikimedia Foundation project. Its functionality should exist in the Incubator and any localisation work done on MediaWiki that is global should be happening in this one central place.
To make sure that localisation is efficient, the technology needs to be improved. It needs to have a priority. We cannot maintain the localisation for more than 250 languages and assume that it will be ok. It is not OK. A fifth of the languages that MediaWiki is said to support is not supported in MediaWiki. The percentages of localisation for many languages is dismal. This hurts the usability of MediaWiki. The reason for all this is that the localisation and the maintenance of the localisation does not get the attention that it requires.
In my opinion the maintenance of MediaWiki and its localisations is a responsibility of the Wikimedia Foundation. There needs to be a plan to have the required resources available on a sustained basis. Consequently the WMF needs to acknowledge its responsibility because only then can we find, fund and implement the solutions that work.
Thanks, GerardM
On 10/28/07, Stephen Bain stephen.bain@gmail.com wrote:
On 10/28/07, GerardM gerard.meijssen@gmail.com wrote:
Resources and investments also imply labour.
...
As long as the User Interface is not localised it is not yet ready to go life for the people to find information in their language.
I don't understand what your argument is. You seem to be talking about priorities in MediaWiki development, but then you talk about interface localisation - which is a matter for the project participants who speak the language and can actually translate the interface, not for Foundation people.
-- Stephen Bain stephen.bain@gmail.com
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: http://lists.wikimedia.org/mailman/listinfo/foundation-l
...
On 28/10/2007, GerardM gerard.meijssen@gmail.com wrote:
Hoi, When you consider the localisation of messages, the messages can be divided in two. There are messages that are project specific and there are messages that have a global reach. When people localise messages, we do not ensure that the effort on the global messages has a global effect. The best place at this moment to localise messages is in the BetaWiki. A project that should be a Wikimedia Foundation project. Its functionality should exist in the Incubator and any localisation work done on MediaWiki that is global should be happening in this one central place.
To make sure that localisation is efficient, the technology needs to be improved. It needs to have a priority. We cannot maintain the localisation for more than 250 languages and assume that it will be ok. It is not OK. A fifth of the languages that MediaWiki is said to support is not supported in MediaWiki. The percentages of localisation for many languages is dismal. This hurts the usability of MediaWiki. The reason for all this is that the localisation and the maintenance of the localisation does not get the attention that it requires.
In my opinion the maintenance of MediaWiki and its localisations is a responsibility of the Wikimedia Foundation. There needs to be a plan to have the required resources available on a sustained basis. Consequently the WMF needs to acknowledge its responsibility because only then can we find, fund and implement the solutions that work.
Thanks, GerardM
On 10/28/07, Stephen Bain stephen.bain@gmail.com wrote:
On 10/28/07, GerardM gerard.meijssen@gmail.com wrote:
Resources and investments also imply labour.
...
As long as the User Interface is not localised it is not yet ready to go life for the people to find information in their language.
I don't understand what your argument is. You seem to be talking about priorities in MediaWiki development, but then you talk about interface localisation - which is a matter for the project participants who speak the language and can actually translate the interface, not for Foundation people.
-- Stephen Bain stephen.bain@gmail.com
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: http://lists.wikimedia.org/mailman/listinfo/foundation-l
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: http://lists.wikimedia.org/mailman/listinfo/foundation-l
2007/10/27, GerardM gerard.meijssen@gmail.com:
It is wrong to suggest that the User Interface is not a vital component of a successful Wikipedia. With some regularity projects are voted to be closed down. They are typically programs that do not have a good localisation, they are typically programs that do not have a community. They are typically Wikipedias that have been started prematurely.
So? I think you're having it backward now. Yes, projects that are voted to be closed down typically do not have a good localisation. Why is that? Not because the lack of a localisation causes the lack of a community, but the lack of a community causes the lack of a localisation. Localisations are there on projects where someone has worked on localisation. That implies that there is someone working on something, and projects where someone is working on something are usually not closed down. Projects that have over 50 articles on different plant species are also usually not voted down. Projects that have a bureaucrat are usually not voted down. Projects that have an active village pump are usually not voted down. Why require a localisation, but not 50 plant species, a bureaucrat or an active village pump?
Yes, there are Wikipedias that are doing fine, the issue is that many people are upset by Wikipedias failing. We lose support for adding new languages with more projects failing.
We lose even more support for adding new languages by making the adding of new languages a long, bureaucratic, drawn out process.
What we aim for in new Wikipedias is project that provide good information to the people. We should not expect people to read any other language. A Wikipedia is there to provide people with knowledge and for this an appropriate user interface is essential. As long as the User Interface is not localised it is not yet ready to go life for the people to find information in their language.
I disagree. Should not the people of a Wikipedia make the decision what is important to them, and what is readable? I think a Wikipedia with a beginning of content and a non-localized or partly-localized interface is more useful than one where every little text is translated but there is just one page.
Stop the language prevention committee!!
2007/10/31, Andre Engels andreengels@gmail.com:
Stop the language prevention committee!!
Please disregard this last comment. It was uncalled-for and contraproductive.
On Wednesday 31 October 2007 11:40, Andre Engels wrote:
2007/10/27, GerardM gerard.meijssen@gmail.com:
It is wrong to suggest that the User Interface is not a vital component of a successful Wikipedia. With some regularity projects are voted to be closed down. They are typically programs that do not have a good localisation, they are typically programs that do not have a community. They are typically Wikipedias that have been started prematurely.
So? I think you're having it backward now. Yes, projects that are voted to be closed down typically do not have a good localisation. Why is that? Not because the lack of a localisation causes the lack of a community, but the lack of a community causes the lack of a localisation. Localisations are there on projects where someone has worked on localisation. That implies that there is someone working on something, and projects where someone is working on something are usually not closed down. Projects that have over 50 articles on different plant species are also usually not voted down. Projects that have a bureaucrat are usually not voted down. Projects that have an active village pump are usually not voted down. Why require a localisation, but not 50 plant species, a bureaucrat or an active village pump?
As a mathematician, I believe you will appreciate the metaphor of quantum tunelling. The same way a particle can "tunnel" through an energy barrier it would otherwise not be able to go through, a project could pass a "knowledge barrier", if helped.
Localisation is an excellent example of this. We can all agree that people are less likely to contribute to a Wikipedia if there is no localisation. Localisation, however, requires a technically competent person to do it. If a given community has no such person, or all such persons are too preocupied with other matters to do it in their free time, the localisation will not be done. It might not be done for years, decades, or - ever.
And these years and decades are years and decades during which the project won't be developed, or will be developed at a much slower rate. If the goal of the WMF is producing free knowledge, and the WMF pays for hosting of Wikipedias in order to produce that knowledge, then it may also pay for a localisation (which would otherwise not happen) in order to achieve the same goal. This may enable free knowledge to be produced at a faster rate (for example, a Wikipedia may get 10 new articles per year before localisation, and 50 articles per year after localisation), thus fulfilling the goal of the WMF.
Of course, I am not suggesting that the WMF should immediately start paying for localisations of desolated Wikipedias. But, when it wants to lift a Wikipedia off the ground, it may well consider localisation of interface, documentation or even initial contents, and in some cases such a consideration will have desirable results.
2007/10/31, Nikola Smolenski smolensk@eunet.yu:
As a mathematician, I believe you will appreciate the metaphor of quantum tunelling. The same way a particle can "tunnel" through an energy barrier it would otherwise not be able to go through, a project could pass a "knowledge barrier", if helped.
It can, but I don't see how that means that one way of helping them (giving them a localized interface) would be superior to another way of helping them (giving them a wiki to start with)
Localisation is an excellent example of this. We can all agree that people are less likely to contribute to a Wikipedia if there is no localisation. Localisation, however, requires a technically competent person to do it. If a given community has no such person, or all such persons are too preocupied with other matters to do it in their free time, the localisation will not be done. It might not be done for years, decades, or - ever.
And these years and decades are years and decades during which the project won't be developed, or will be developed at a much slower rate.
So instead we don't allow it to develop at all? We just sit and wait, don't work on other blockades until this one is resolved? That will help a language grow!
I don't disagree that localization is a good thing, although I do think its effect is smaller than you seem to think. In my view, the first and foremost need is people - have 1 contributor, and the project will probably die, have 3 and it's in serious danger, have 5 and it's likely to live, have 10 and it's ready thrive.
Still, having a localization would be a good thing. However, the issue is: how much importance does it get, and how is it achieved. Currently the situation is that it gets foremost importance, and is achieved by withholding most other things a project would need or want (like an own wiki and an official status) until it has been resolved. To me, that's giving much too much importance to just one piece of what makes a successful Wikipedia. Let the people from a Wikipedia language decide for themselves what is important to their project at which moment. Give them ample opportunities to localize their interface, and point them towards the possibility, if necessary multiple times, but if they still decide they'd rather work on a non-localized interface rather than spend their time on localizing it, accept that in the end it is their choice to make.
Hoi, The localisation is of particular importance for the readers and for the editors that do not know the language that is chosen as the secondary language. With the start of the Incubator there is a good place to start a project and build up enough steam to stand on its own.
There are two groups of people in the WMF; there are those that are of the opinion that more language support is a distraction and there are those that are of the opinion that there are too many hoops that new projects have to jump through. By defining minimum requirements we aim to prevent the failure of projects and we aim to provide a good user experience for new languages when the project goes life. In this way we reach out to both groups and both groups are likely not to be happy anyway. Our argument is that in this way we aim to optimise the effectiveness of new projects, not only is a language supported for a WMF project and also MediaWiki supports a new language.
Thanks, GerardM
On 11/1/07, Andre Engels andreengels@gmail.com wrote:
2007/10/31, Nikola Smolenski smolensk@eunet.yu:
As a mathematician, I believe you will appreciate the metaphor of
quantum
tunelling. The same way a particle can "tunnel" through an energy
barrier it
would otherwise not be able to go through, a project could pass a
"knowledge
barrier", if helped.
It can, but I don't see how that means that one way of helping them (giving them a localized interface) would be superior to another way of helping them (giving them a wiki to start with)
Localisation is an excellent example of this. We can all agree that
people are
less likely to contribute to a Wikipedia if there is no localisation. Localisation, however, requires a technically competent person to do it.
If a
given community has no such person, or all such persons are too
preocupied
with other matters to do it in their free time, the localisation will
not be
done. It might not be done for years, decades, or - ever.
And these years and decades are years and decades during which the
project
won't be developed, or will be developed at a much slower rate.
So instead we don't allow it to develop at all? We just sit and wait, don't work on other blockades until this one is resolved? That will help a language grow!
I don't disagree that localization is a good thing, although I do think its effect is smaller than you seem to think. In my view, the first and foremost need is people - have 1 contributor, and the project will probably die, have 3 and it's in serious danger, have 5 and it's likely to live, have 10 and it's ready thrive.
Still, having a localization would be a good thing. However, the issue is: how much importance does it get, and how is it achieved. Currently the situation is that it gets foremost importance, and is achieved by withholding most other things a project would need or want (like an own wiki and an official status) until it has been resolved. To me, that's giving much too much importance to just one piece of what makes a successful Wikipedia. Let the people from a Wikipedia language decide for themselves what is important to their project at which moment. Give them ample opportunities to localize their interface, and point them towards the possibility, if necessary multiple times, but if they still decide they'd rather work on a non-localized interface rather than spend their time on localizing it, accept that in the end it is their choice to make.
-- Andre Engels, andreengels@gmail.com ICQ: 6260644 -- Skype: a_engels
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: http://lists.wikimedia.org/mailman/listinfo/foundation-l
That's all good, all of that stuff about preventing failure of projects, blah blah blah. I'm pretty sure that incubator projects that already have, say, over 200 articles are not going to be deemed failures if they are created now, with or without an interface. (examples include bcl, stq, ext...)
Mark
On 01/11/2007, GerardM gerard.meijssen@gmail.com wrote:
Hoi, The localisation is of particular importance for the readers and for the editors that do not know the language that is chosen as the secondary language. With the start of the Incubator there is a good place to start a project and build up enough steam to stand on its own.
There are two groups of people in the WMF; there are those that are of the opinion that more language support is a distraction and there are those that are of the opinion that there are too many hoops that new projects have to jump through. By defining minimum requirements we aim to prevent the failure of projects and we aim to provide a good user experience for new languages when the project goes life. In this way we reach out to both groups and both groups are likely not to be happy anyway. Our argument is that in this way we aim to optimise the effectiveness of new projects, not only is a language supported for a WMF project and also MediaWiki supports a new language.
Thanks, GerardM
On 11/1/07, Andre Engels andreengels@gmail.com wrote:
2007/10/31, Nikola Smolenski smolensk@eunet.yu:
As a mathematician, I believe you will appreciate the metaphor of
quantum
tunelling. The same way a particle can "tunnel" through an energy
barrier it
would otherwise not be able to go through, a project could pass a
"knowledge
barrier", if helped.
It can, but I don't see how that means that one way of helping them (giving them a localized interface) would be superior to another way of helping them (giving them a wiki to start with)
Localisation is an excellent example of this. We can all agree that
people are
less likely to contribute to a Wikipedia if there is no localisation. Localisation, however, requires a technically competent person to do it.
If a
given community has no such person, or all such persons are too
preocupied
with other matters to do it in their free time, the localisation will
not be
done. It might not be done for years, decades, or - ever.
And these years and decades are years and decades during which the
project
won't be developed, or will be developed at a much slower rate.
So instead we don't allow it to develop at all? We just sit and wait, don't work on other blockades until this one is resolved? That will help a language grow!
I don't disagree that localization is a good thing, although I do think its effect is smaller than you seem to think. In my view, the first and foremost need is people - have 1 contributor, and the project will probably die, have 3 and it's in serious danger, have 5 and it's likely to live, have 10 and it's ready thrive.
Still, having a localization would be a good thing. However, the issue is: how much importance does it get, and how is it achieved. Currently the situation is that it gets foremost importance, and is achieved by withholding most other things a project would need or want (like an own wiki and an official status) until it has been resolved. To me, that's giving much too much importance to just one piece of what makes a successful Wikipedia. Let the people from a Wikipedia language decide for themselves what is important to their project at which moment. Give them ample opportunities to localize their interface, and point them towards the possibility, if necessary multiple times, but if they still decide they'd rather work on a non-localized interface rather than spend their time on localizing it, accept that in the end it is their choice to make.
-- Andre Engels, andreengels@gmail.com ICQ: 6260644 -- Skype: a_engels
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: http://lists.wikimedia.org/mailman/listinfo/foundation-l
foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: http://lists.wikimedia.org/mailman/listinfo/foundation-l
wikimedia-l@lists.wikimedia.org