Hi everybody,
While it's sort of obvious, given the digital divide, that the number of articles in Wikipedias is not proportional to the number of speakers, for example Hindi has a much smaller number of articles compared to speakers than most active Wikipedias; German has more.
However, something that people may not notice as much is the incongruency between popularity of a particular language version and the number of articles in that version.
The most visited Wikipedias, in order, are:
1 English (65%) 2 German (10%) 3 Japanese (6%) 4 Spanish (3%) 5 French (2%) 6 Polish (2%) 7 Chinese (2%) 8 Arabic (2%) 9 Italian (1%) 10 Hebrew (1%) 11 Turkish (1%) 12 Dutch (1%) 13 Portuguese (1%) (all others combined total 1% of visits)
On the other hand, the list of Wikipedias ranked by number of articles is: 1 English (1048.7K) 2 German (376.9K) 3 French (261.1K) 4 Polish (223.8K) 5 Japanese (196.3K) 6 Dutch (156.9K) ... 8 Italian (146.8K) 9 Portuguese (123.8K) 10 Spanish (105.0K) ... 12 Chinese (61.48K) ... 17 Hebrew (34.35K) ... 29 Turkish (19.94K) ... 37 Arabic (12.03K)
What this says to me is that these Wikipedias are not attracting new pages proportional to views when compared with other Wikipedias. This may be because people don't want to write new pages, but it seems to me more likely that people simply don't know they can.
How can this be fixed? Perhaps a site notice inviting people to write quality pages or register, or a drive to recruit new Wikipedians from the academic community.
Mark
-- "Take away their language, destroy their souls." -- Joseph Stalin
On 3/29/06, Mark Williamson node.ue@gmail.com wrote:
How can this be fixed? Perhaps a site notice inviting people to write quality pages or register, or a drive to recruit new Wikipedians from the academic community.
The site notice can indeed be a useful instrument, especially when used for specific campaigns, I think. On the English Wikinews I proposed the idea of a community-approved site notice of the week (in this case, it would only be shown to registered users): http://en.wikinews.org/wiki/Wikinews:Site_notice_of_the_week
Make sure there are enough entry points. The hard distinction between "community" and "content" is not as important for a young project -- I think it's fine to have red links on a frontpage of a Wikipedia that is still in the growth stage, for example. Also stuff like "collaboration of the week". Stub notices might also help, though some people hate them.
What also makes a huge difference is press coverage. People who only discover Wikipedia through their regular Internet searches may never understand how it works. The German Wikipedia has had _gigantic_ media coverage in the last few years -- more hype than even the English one, in my opinion. Even events like Wikimania were covered in national newspapers. Arguably, in the case of de, it was even a little too much for the community to deal with.
The new Communications Committee might be able to help with raising Wikipedia-awareness in specific countries: http://meta.wikimedia.org/wiki/Communications_committee
By now, a general "Wikipedia is great, everyone can edit it" story doesn't work so well anymore. ;-) But finding an interesting regional hook should make things interesting -- if there's one thing media everywhere seem to have in common, it's that they will happily accept any excuse to write about Wikipedia. One story angle many media are choosing now is the "Meet the Wikipedians" type, with lots of individual portraits (kudos to Wired for pioneering this form). The community can help in obvious ways in laying the groundwork for such stories.
But, importantly, don't worry too much about these issues. The key factor are likely cultural reasons which we cannot change. Wikimedia is like a forest, and some trees just grow faster than others -- but they all grow. Worry more about the issue of access (both because of censorship and lack of technology). Wikimedia can do a lot more in this area.
Erik
How one can keep in touch with the work of the committee?
Another question: can somebody take the effort to summarize all the great ideas about promoting a smaller wp that have sounded in this discussion (and probably the others)? Enthusiasts of new and smaller wikipedias might be glad to find such a text.
ru/os:User:Amikeco
2006/3/29, Erik Moeller eloquence@gmail.com:
The new Communications Committee might be able to help with raising Wikipedia-awareness in specific countries: http://meta.wikimedia.org/wiki/Communications_committee
-- Esperu cxiam!
One way is posting flyers at universities and other intellectual "hotspots". The Icelandic Wikipedia had printable fliers a while back for people to post around your school, they were neat, and could probably very easily be adapted to any other language. (sort of like, "Come help us build the first free online encyclopedia in Icelandic" or something like that).
Other ways, especially for minority languages, are to contact groups which promote the language, if they exist. If there is any sort of periodical in that language, it might be good if you can get them to write something about your Wiki, and if you can't, you can possibly place an ad if you're willing to spend money. Even if there isn't a periodical in the language, though, you may be able to get a local newspaper (in the LWC) to write an article.
You can also post flyers around town (on lampposts, public telephones, abandoned buildings, etc., as long as it's legal where you live).
Now, this generally assumes people have internet -- so they can visit Wikipedia and edit it. However, if internet access is low, you can have "If you want to help us write the encyclopaedia, call for more information" with your phone number. If phone access is low, you can give your address, or some other way they can reach you.
Mark
On 30/03/06, V. Ivanov amikeco@gmail.com wrote:
How one can keep in touch with the work of the committee?
Another question: can somebody take the effort to summarize all the great ideas about promoting a smaller wp that have sounded in this discussion (and probably the others)? Enthusiasts of new and smaller wikipedias might be glad to find such a text.
ru/os:User:Amikeco
2006/3/29, Erik Moeller eloquence@gmail.com:
The new Communications Committee might be able to help with raising Wikipedia-awareness in specific countries: http://meta.wikimedia.org/wiki/Communications_committee
-- Esperu cxiam! _______________________________________________ Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
-- "Take away their language, destroy their souls." -- Joseph Stalin
On 3/30/06, Mark Williamson node.ue@gmail.com wrote:
One way is posting flyers at universities and other intellectual "hotspots". The Icelandic Wikipedia had printable fliers a while back for people to post around your school, they were neat, and could probably very easily be adapted to any other language. (sort of like, "Come help us build the first free online encyclopedia in Icelandic" or something like that).
Is this flyer still available online somewhere? I think that's a great idea, I'd like to take a look at it.
I've been thinking about doing something similar in my area, which is pretty multilingual. I know several people offhand who could probably be persuaded to translate the flier, assuming it's a short document.
Plenty of places to post them, as well -- most coffee places have community bulletin boards, so do libraries.
Pat -- ˙6uısnɯɐ sı əpoɔıun
Hi!
At PMS we receive periodic vandalism from anon user 69.232.157.175, located at adsl-69-232-157-175.dsl.irvnca.pacbell.net. Is there anyway we can block this IP to access pages? and can we please configure the instance in such a way, that anon users won't be able to modify pages? I got problems enough in my life even without this kind of "contributors".
Thanks Bèrto
2006/3/31, Berto albertoserra@ukr.net:
At PMS we receive periodic vandalism from anon user 69.232.157.175, located at adsl-69-232-157-175.dsl.irvnca.pacbell.net. Is there anyway we can block this IP to access pages?
You will have to ask a Steward to make one (or more) of you a sysop. They then have the ability to block an IP address, which means that no edits can be done from that address.
-- Andre Engels, andreengels@gmail.com ICQ: 6260644 -- Skype: a_engels
Hi!
Probably there's just a single connection on this IP address (one house, school or small company). Usually in this case the various users have different but similar IP addresses.
It actually does not seem to be a floating IP, neither a political thing. He simply drops in meaningless chars instead of the original page, or redirects the page to the english edition (which does not work), his style seems to be pretty stable. So you are likely to be right. It's a unique IP. Let's block it.
You will have to ask a Steward to make one (or more) of you a sysop. They then have the ability to block an IP address, which means that no edits can be done from that address.
Yeap! I've asked for it some 10 days ago :) Still waiting for an answer :) I hoped that maybe there could be someone with that access here, and that they could terminate the guy for us, while we wait for operative indipendence :)
Bèrto
Berto wrote:
Hi!
Probably there's just a single connection on this IP address (one house, school or small company). Usually in this case the various users have different but similar IP addresses.
It actually does not seem to be a floating IP, neither a political thing. He simply drops in meaningless chars instead of the original page, or redirects the page to the english edition (which does not work), his style seems to be pretty stable. So you are likely to be right. It's a unique IP. Let's block it.
You will have to ask a Steward to make one (or more) of you a sysop. They then have the ability to block an IP address, which means that no edits can be done from that address.
Yeap! I've asked for it some 10 days ago :) Still waiting for an answer :) I hoped that maybe there could be someone with that access here, and that they could terminate the guy for us, while we wait for operative indipendence :)
Bèrto
Err, not really. Your first request 10 days ago was incomplete. It lacked the necessary links (which was understandable since the language was not yet created... no project ---> no pages ---> no sysop). Jon explained that to you.
Then, when the page was created (2 days ago), Jon explained to you we usually wait 1 week before giving the status.
So, no, you are not waiting for an answer from steward... stewards are waiting for any editor on your language answer on the matter :-) Posting here will not make a difference ;-)
Now... let me tell you about all the tricks which will allow you to survive .... in case of vandalism, ask temporary access here : http://meta.wikimedia.org/wiki/Requests_for_permissions#Current_temporary_pe...
Cheers
Ant
Hi!
Err, not really. Your first request 10 days ago was incomplete. It lacked the necessary links (which was understandable since the language was not yet created... no project ---> no pages ---> no sysop). Jon explained that to you.
Then, when the page was created (2 days ago), Jon explained to you we usually wait 1 week before giving the status.
So, no, you are not waiting for an answer from steward... stewards are waiting for any editor on your language answer on the matter :-) Posting here will not make a difference ;-)
LOL yes... I wasn't actually trying to push the matters :) It's just that in the last days I spent so much time on the wiki that my circadians rythms are beginning to be shaky :) Normally I can count up to 10 :)))) Sometimes I even reach eleven without mistakes :)))
Now... let me tell you about all the tricks which will allow you to survive .... in case of vandalism, ask temporary access here : http://meta.wikimedia.org/wiki/Requests_for_permissions#Current_temporary_p
ermissions_for_emergency_or_technical_purposes
Thanks! It's what I needed :))
Bèrto
Hi!
I get a constant error message when trying to access the fulltext function search on PMS wiki.
Try this http://pms.wikipedia.org/wiki/Speciale:Search?search=colonnetti&fulltext... ca
Is it me, or is it a problem with settings? It's been going for 6 hours in a row, now.
Thanks Bèrto
Berto wrote:
I get a constant error message when trying to access the fulltext function search on PMS wiki.
Try this http://pms.wikipedia.org/wiki/Speciale:Search?search=colonnetti&fulltext... ca
Is it me, or is it a problem with settings? It's been going for 6 hours in a row, now.
The new wikis don't appear in the search index yet.
-- brion vibber (brion @ pobox.com)
Hi!
At PMS we receive periodic vandalism from anon user 69.232.157.175,
located
at adsl-69-232-157-175.dsl.irvnca.pacbell.net. Is there anyway we can
block
this IP to access pages?
On a second thought, we obviously dont want to cut off an entire ADSL network... So probably the only possible form of filter is to have him at least spend the time to create a new user.
Bèrto
2006/3/31, Berto albertoserra@ukr.net:
On a second thought, we obviously dont want to cut off an entire ADSL network... So probably the only possible form of filter is to have him at least spend the time to create a new user.
Probably there's just a single connection on this IP address (one house, school or small company). Usually in this case the various users have different but similar IP addresses.
-- Andre Engels, andreengels@gmail.com ICQ: 6260644 -- Skype: a_engels
I'd definately like to see a standard flier. I'm sure people could arrange translations. Perhaps even just a paragraph or so and then have many many translations on one flier :)
Fran
On Thu, 2006-03-30 at 18:20 -0500, Patrick Hall wrote:
On 3/30/06, Mark Williamson node.ue@gmail.com wrote:
One way is posting flyers at universities and other intellectual "hotspots". The Icelandic Wikipedia had printable fliers a while back for people to post around your school, they were neat, and could probably very easily be adapted to any other language. (sort of like, "Come help us build the first free online encyclopedia in Icelandic" or something like that).
Is this flyer still available online somewhere? I think that's a great idea, I'd like to take a look at it.
I've been thinking about doing something similar in my area, which is pretty multilingual. I know several people offhand who could probably be persuaded to translate the flier, assuming it's a short document.
Plenty of places to post them, as well -- most coffee places have community bulletin boards, so do libraries.
Pat
˙6uısnɯɐ sı əpoɔıun _______________________________________________ Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
Francis Tyers schrieb:
I'd definately like to see a standard flier. I'm sure people could arrange translations. Perhaps even just a paragraph or so and then have many many translations on one flier :)
The evil version of me would now say (especially in regard to the current fight on foundation-l): it's on meta, try to find it.
The more gentle version of me says: http://meta.wikimedia.org/wiki/Leaflet
greetings, elian
Haha!
Sorry for that, but my brain obviously wasn't turned on and I'm not subscribed to foundation-l :)
Thanks for the pointer,
Fran
On Sat, 2006-04-01 at 06:04 +0200, Elisabeth Bauer wrote:
Francis Tyers schrieb:
I'd definately like to see a standard flier. I'm sure people could arrange translations. Perhaps even just a paragraph or so and then have many many translations on one flier :)
The evil version of me would now say (especially in regard to the current fight on foundation-l): it's on meta, try to find it.
The more gentle version of me says: http://meta.wikimedia.org/wiki/Leaflet
greetings, elian
Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
Mark, I don't agree with all of your conclusions here....
What this says to me is that these Wikipedias are not attracting new pages proportional to views when compared with other Wikipedias. This may be because people don't want to write new pages, but it seems to me more likely that people simply don't know they can.
There are all kinds of possible reasons for this. It isn't necessarily because non-Europeans (and the wikipedias you seem to be referring to are mostly non-European wikipedias) are less likely to know that they can write new articles. Perhaps people are less interested in writing new articles, perhaps they're content with the articles that are there, or perhaps they have other ways to spend their time. Perhaps they _are_ increasingly writing new articles - some of the wikipedias you listed might still be small, but they are growing rapidly.
How can this be fixed? Perhaps a site notice inviting people to write quality pages or register, or a drive to recruit new Wikipedians from the academic community.
I think here you're confusing the issues of quantity and quality. Writing 'quality pages', or inviting academics to do so, isn't the way to boost page count quickly. And I'm sure many academics would be less interested in the number of articles a wikipedia has than the quality of those articles anyway. Perhaps it would be better to forget about page count and focus on improving the overall quality of wikipedia.
Hyunsung
Mark
-- "Take away their language, destroy their souls." -- Joseph Stalin _______________________________________________ Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
------------------------------------------------- Your Life on the Net DreamWiz Free Mail @ http://www.dreamwiz.com/
Hi!
I think here you're confusing the issues of quantity and quality. Writing 'quality pages', or inviting academics to do so, isn't the way to boost page count quickly. And I'm sure many academics would be less interested in the number of articles a wikipedia has than the quality of those articles anyway. Perhaps it would be better to forget about page count and focus on improving the overall quality of wikipedia.
Yes, and high quality often acts as a barrier against massive partecipation. I myself would think twice before changing a word on the Britannica... Some balance must be reached, though, because quality is what makes reading a wiki interesting for the public. It's also very difficult to judge how the quality vs quantity balance is achieved in a single wiki, since the first term of the equation is mainly a subjective parameter.
I guess the main thing a wiki needs is a solid group of managers, that are capable of attracting quality from academical level environments (we are contacting university teachers who have retired, for example) and at the same time can put up a good "village pump", where people may socialize and become community members. Not an easy thing, though. Many people who love culture and may be wonderful researchers might lack the socializing capability that is needed to build a constructive environment.
A wiki is a world in itself, a book that writes itself. But directors are needed, unless you want the story to go nowhere. Bulding a good PR team means more than achieving a quick rise in published page numbers by running bots to make millions of empty stubs. Nonetheless, empty stubs are necessary to shape content growth... It's a HUGE job, as I am realizing.
Bèrto
Mark, I don't agree with all of your conclusions here....
What this says to me is that these Wikipedias are not attracting new pages proportional to views when compared with other Wikipedias. This may be because people don't want to write new pages, but it seems to me more likely that people simply don't know they can.
There are all kinds of possible reasons for this. It isn't necessarily because non-Europeans (and the wikipedias you seem to be referring to are mostly non-European wikipedias) are less likely to know that they can write new articles. Perhaps people are less interested in writing new articles, perhaps they're content with the articles that are there, or perhaps they have other ways to spend their time. Perhaps they _are_ increasingly writing new articles - some of the wikipedias you listed might still be small, but they are growing rapidly.
How can this be fixed? Perhaps a site notice inviting people to write quality pages or register, or a drive to recruit new Wikipedians from the academic community.
I think here you're confusing the issues of quantity and quality. Writing 'quality pages', or inviting academics to do so, isn't the way to boost page count quickly. And I'm sure many academics would be less interested in the number of articles a wikipedia has than the quality of those articles anyway. Perhaps it would be better to forget about page count and focus on improving the overall quality of wikipedia.
Hyunsung
Mark
-- "Take away their language, destroy their souls." -- Joseph Stalin _______________________________________________ Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
------------------------------------------------- Your Life on the Net DreamWiz Free Mail @ http://www.dreamwiz.com/
Alright, if it's quality you're after (which I certainly agree with), there's another statistic that can be used, total database size.
Of course, you could fill a database with 3 gigabytes of stub pages, but it would be many, many, more stub pages and a much more "complete" Stub-Wikipedia than if you were to fill it with 3 gigabytes of HQ non-stub articles. It also doesn't take linguistic considerations into effect (texts in one language may be longer than equivalent texts in another)
It goes (stats from December)
1 English - ~2500mb 2 German - ~1000mb 3 Japanese - ~550mb 4 French - ~550mb 5 Polish - ~300mb 6 Italian - ~300mb 7 Dutch - ~250mb 8 Spanish - ~225mb 9 Portuguese - ~175mb ... 12 Chinese - ~125mb 13 Hebrew - ~125mb ... 26 Turkish - ~33mb ... 32 Arabic - ~26mb
The biggest divorce seems to be with the Spanish, Chinese, Arabic, and Turkish Wikipedias.
This may or may not have something to do with it, but take a look at the Spanish vs French mainpages -- the French one mentions very, very prominently that anyone can edit, the Spanish one does not. The Chinese WP does mention it, I'm not 100% sure about the Arabic or Turkish.
Erik Moeller also has some good points.
Mark
On 28/03/06, 오현성 chamdalae@dreamwiz.com wrote:
Mark, I don't agree with all of your conclusions here....
What this says to me is that these Wikipedias are not attracting new pages proportional to views when compared with other Wikipedias. This may be because people don't want to write new pages, but it seems to me more likely that people simply don't know they can.
There are all kinds of possible reasons for this. It isn't necessarily because non-Europeans (and the wikipedias you seem to be referring to are mostly non-European wikipedias) are less likely to know that they can write new articles. Perhaps people are less interested in writing new articles, perhaps they're content with the articles that are there, or perhaps they have other ways to spend their time. Perhaps they _are_ increasingly writing new articles - some of the wikipedias you listed might still be small, but they are growing rapidly.
How can this be fixed? Perhaps a site notice inviting people to write quality pages or register, or a drive to recruit new Wikipedians from the academic community.
I think here you're confusing the issues of quantity and quality. Writing 'quality pages', or inviting academics to do so, isn't the way to boost page count quickly. And I'm sure many academics would be less interested in the number of articles a wikipedia has than the quality of those articles anyway. Perhaps it would be better to forget about page count and focus on improving the overall quality of wikipedia.
Hyunsung
Mark
-- "Take away their language, destroy their souls." -- Joseph Stalin _______________________________________________ Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
Your Life on the Net DreamWiz Free Mail @ http://www.dreamwiz.com/
Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
-- "Take away their language, destroy their souls." -- Joseph Stalin
The biggest divorce seems to be with the Spanish, Chinese, Arabic, and Turkish Wikipedias.
This may or may not have something to do with it, but take a look at the Spanish vs French mainpages -- the French one mentions very, very prominently that anyone can edit, the Spanish one does not. The Chinese WP does mention it, I'm not 100% sure about the Arabic or Turkish.
Turkish Wikipedia clearly mentions on its main page (using large font size) that everyone can edit.
There's so little Turkish language content on the Internet in general and most Turkish speaking people don't speak English or any other language. This may be one of the reasons of the inconsistancy between article number and visit count.
I suppose there is no such thing as a single explanation. Some editions may have an approach problem, that is, they may not be good enough in marketing themselves. Other may stumble against a local tradition in which a written word is a law, so people read and believes, rather than editing and adding/discussing content. Cultures are very different. Besides, most people who can properly use a wiki will also look for an english version in the first place, and only eventually access a local variant. Too bad these are also the guys who can mostly use an editor.
IMHO, edition lacking partecipation should better focus on their work with children, who are quicker into adopting technologies. The wiki generation had time enough to grow up in many language spaces, but is only being born in many others. Instead of looking for a quick rise in rating, I'd rather advice people to build solid grounds for the future.
Bèrto
----- Original Message ----- From: "Mark Williamson" node.ue@gmail.com To: wikipedia-l@wikimedia.org Sent: Wednesday, March 29, 2006 5:18 AM Subject: [Wikipedia-l] Language versions' popularity vs. number of articles(vs. number of speakers)
Hi everybody,
While it's sort of obvious, given the digital divide, that the number of articles in Wikipedias is not proportional to the number of speakers, for example Hindi has a much smaller number of articles compared to speakers than most active Wikipedias; German has more.
However, something that people may not notice as much is the incongruency between popularity of a particular language version and the number of articles in that version.
The most visited Wikipedias, in order, are:
1 English (65%) 2 German (10%) 3 Japanese (6%) 4 Spanish (3%) 5 French (2%) 6 Polish (2%) 7 Chinese (2%) 8 Arabic (2%) 9 Italian (1%) 10 Hebrew (1%) 11 Turkish (1%) 12 Dutch (1%) 13 Portuguese (1%) (all others combined total 1% of visits)
On the other hand, the list of Wikipedias ranked by number of articles is: 1 English (1048.7K) 2 German (376.9K) 3 French (261.1K) 4 Polish (223.8K) 5 Japanese (196.3K) 6 Dutch (156.9K) ... 8 Italian (146.8K) 9 Portuguese (123.8K) 10 Spanish (105.0K) ... 12 Chinese (61.48K) ... 17 Hebrew (34.35K) ... 29 Turkish (19.94K) ... 37 Arabic (12.03K)
What this says to me is that these Wikipedias are not attracting new pages proportional to views when compared with other Wikipedias. This may be because people don't want to write new pages, but it seems to me more likely that people simply don't know they can.
How can this be fixed? Perhaps a site notice inviting people to write quality pages or register, or a drive to recruit new Wikipedians from the academic community.
Mark
-- "Take away their language, destroy their souls." -- Joseph Stalin _______________________________________________ Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
Mark Williamson wrote:
Hi everybody,
While it's sort of obvious, given the digital divide, that the number of articles in Wikipedias is not proportional to the number of speakers, for example Hindi has a much smaller number of articles compared to speakers than most active Wikipedias; German has more.
However, something that people may not notice as much is the incongruency between popularity of a particular language version and the number of articles in that version.
The most visited Wikipedias, in order, are:
1 English (65%) 2 German (10%) 3 Japanese (6%) 4 Spanish (3%) 5 French (2%) 6 Polish (2%) 7 Chinese (2%) 8 Arabic (2%) 9 Italian (1%) 10 Hebrew (1%) 11 Turkish (1%) 12 Dutch (1%) 13 Portuguese (1%) (all others combined total 1% of visits)
On the other hand, the list of Wikipedias ranked by number of articles is: 1 English (1048.7K) 2 German (376.9K) 3 French (261.1K) 4 Polish (223.8K) 5 Japanese (196.3K) 6 Dutch (156.9K) ... 8 Italian (146.8K) 9 Portuguese (123.8K) 10 Spanish (105.0K) ... 12 Chinese (61.48K) ... 17 Hebrew (34.35K) ... 29 Turkish (19.94K) ... 37 Arabic (12.03K)
Hi
By the way, I have wondered for a long time how it could be that the french wikipedia be the third largest (which is a fact)... and only supposingly 2% of the visits. I also remember that meta had very significantly more visits than the french wikipedia. Which.. to me... is not something I *believe*.
Answer : these values are collected strictly thanks to the use of the Alexa bar...; which is apparently very little used at least in France. Hence possibly messing up the statistics. It might well be the same for other languages, in particular those for which the Alexa bar... is simply not working/translated. So... be careful with the use of the Alexa data. They are a very interesting feedback... but also to take with a pinch of salt.
By the way, there was recently a study by Mediametry (a statistical agency, making polls. In France, probably the largest and most famous one).
The study showed this
déc. 2004 avr. 2005 août 2005 déc. 2005 USA 4 544 000 6 753 000 11 874 000 17 498 000 Germany 1 707 000 2 956 000 4 502 000 5 529 000 France 681 000 1 169 000 1 429 000 2 421 000 UK 540 000 766 000 1 525 000 2 298 000
Please see that : http://www.google.com/search?hl=en&q=mediametrie+wikipedia&btnG=Goog...
anthere
Anthere wrote:
Answer : these values are collected strictly thanks to the use of the Alexa bar...; which is apparently very little used at least in France. Hence possibly messing up the statistics. It might well be the same for other languages, in particular those for which the Alexa bar... is simply not working/translated. So... be careful with the use of the Alexa data. They are a very interesting feedback... but also to take with a pinch of salt.
It's the same in Poland - I think very few Poles use Alexa bar.
Hi, I wonder if wikipedia only relies on this sort of external statistics (like Alexa) to gather information about visits to the sites. Aren't there statistcs collected on wikipedia servers itself? This would be more useful and reliable. BTW, if we want to know the popularity of an specific article (not a specific wikipedia), is there a tool for that?
Regards, Miguel.
On 3/29/06, Pawe³ Dembowski fallout@lexx.eu.org wrote:
Anthere wrote:
Answer : these values are collected strictly thanks to the use of the Alexa bar...; which is apparently very little used at least in France. Hence possibly messing up the statistics. It might well be the same for other languages, in particular those for which the Alexa bar... is simply not working/translated. So... be careful with the use of the Alexa data. They are a very interesting feedback... but also to take with a pinch of salt.
It's the same in Poland - I think very few Poles use Alexa bar.
-- Ausir Wikipedia, wolna encyklopedia http://pl.wikipedia.org
Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
Miguel Chaves wrote:
Hi, I wonder if wikipedia only relies on this sort of external statistics (like Alexa) to gather information about visits to the sites. Aren't there statistcs collected on wikipedia servers itself? This would be more useful and reliable.
Not at this time. At our traffic level, web server logs are too large to handle comfortably without a dedicated infrastructure, and we've been forced to simply disable them until something easier to handle gets set up.
(If we were an ad-supported site, such statistics would be much much more important and we'd have put in the time and money for it a lot sooner.)
BTW, if we want to know the popularity of an specific article (not a specific wikipedia), is there a tool for that?
Not really, sorry.
-- brion vibber (brion @ pobox.com)
On 29/03/06, Brion Vibber brion@pobox.com wrote:
Miguel Chaves wrote:
Hi, I wonder if wikipedia only relies on this sort of external statistics (like Alexa) to gather information about visits to the sites. Aren't there statistcs collected on wikipedia servers itself? This would be more useful and reliable.
Not at this time. At our traffic level, web server logs are too large to handle comfortably without a dedicated infrastructure, and we've been forced to simply disable them until something easier to handle gets set up.
Is that "keep recording but ignore them", or disable in the sense of turn off logging totally? Just curious...
-- - Andrew Gray andrew.gray@dunelm.org.uk
Andrew Gray wrote:
Is that "keep recording but ignore them", or disable in the sense of turn off logging totally? Just curious...
After a few months of having logs that you're not reading fill up the servers' hard disks every few days, you turn them off. :)
-- brion vibber (brion @ pobox.com)
Brion Vibber wrote:
Andrew Gray wrote:
Is that "keep recording but ignore them", or disable in the sense of turn off logging totally? Just curious...
After a few months of having logs that you're not reading fill up the servers' hard disks every few days, you turn them off. :)
-- brion vibber (brion @ pobox.com)
How about a cron job that turns logging on, then off, intermittently? Eg.
For example, on each server, have a cron job that does this:
Every 5 mins: Is logging on? Then: turn it off Else: generate a random number If it's == 0 mod 1000: Then: turn logging on Else: do nothing
This way, you get representative short blocks of 5 minutes of traffic, kicking in once every three days or so on each of the 100 or so servers at random times of the day or night. This would also suffice for gross statistical analysis, and wouldn't require any modification of the squid code, just a short external shell script.
Log-rotation should handle the rest and prevent the disks filling up, since the average sampling rate would then be low enough to cope with.
-- Neil
Loi,
I think, it's really good for the community that we have no reliable statistics on "language popularity" and article popularity. Otherwise there would be inevitable talks like "let's close this or that wiki, it's a place visited by a couple of same people" (I've already heard similar ideas about the wp-s edited by one to few editors).
In the recent environment smaller language wikipedias can feel like really equal parts of the community -- and that's great. The same about the larger language wikipedias, when the language users have not come into the Internet massively yet.
Sl./ru:User:Amikeco
2006/3/29, Brion Vibber brion@pobox.com:
Not at this time. At our traffic level, web server logs are too large to handle comfortably without a dedicated infrastructure, and we've been forced to simply disable them until something easier to handle gets set up.
(If we were an ad-supported site, such statistics would be much much more important and we'd have put in the time and money for it a lot sooner.)
BTW, if we want to know the popularity of an specific article (not a specific wikipedia), is there a tool for that?
Not really, sorry.
-- brion vibber (brion @ pobox.com)
-- Esperu cxiam!
Brion Vibber wrote:
Miguel Chaves wrote:
Hi, I wonder if wikipedia only relies on this sort of external statistics (like Alexa) to gather information about visits to the sites. Aren't there statistcs collected on wikipedia servers itself? This would be more useful and reliable.
Not at this time. At our traffic level, web server logs are too large to handle comfortably without a dedicated infrastructure, and we've been forced to simply disable them until something easier to handle gets set up.
(If we were an ad-supported site, such statistics would be much much more important and we'd have put in the time and money for it a lot sooner.)
BTW, if we want to know the popularity of an specific article (not a specific wikipedia), is there a tool for that?
Not really, sorry.
-- brion vibber (brion @ pobox.com)
Since the traffic is so vast, why not use random sampling? At each page hit, call a random-number generator (eg read four bytes from /dev/urandom, or call a seeded pseudo-random number routine), and make a log entry only if its result == 0 mod 1000. That way, the logs will be statistically representative, but only require a relatively tiny amount of disk I/O, compute time, and disk space.
Alternatively, you could log using UDP syslog, and have a listener that threw away 999 out of 1000 packets.
-- Neil
Neil Harris wrote:
Since the traffic is so vast, why not use random sampling?
That's been suggested, and would indeed be useful. If someone's interested in writing patches for squid to do this, speak up. :)
-- brion vibber (brion @ pobox.com)
Miguel Chaves wrote:
Hi, I wonder if wikipedia only relies on this sort of external statistics (like Alexa) to gather information about visits to the sites. Aren't there statistcs collected on wikipedia servers itself? This would be more useful and reliable. BTW, if we want to know the popularity of an specific article (not a specific wikipedia), is there a tool for that?
There used to be a long time ago. It was switched off for various reasons .... unfortunately :(
Waerth/Walter
That's certainly a good point.
Unfortunately, the most recent statistics we have for site visits are from October 2004 but they are as follows:
1 English: 42.75% 2 Japanese: 14.72% 3 German: 13.74% 4 Spanish: 5.34% 5 French: 4.14% 6 Polish: 3.93% 7 Dutch: 2.94% 8 Swedish: 1.74% 9 Chinese: 1.42% 10 Italian: 1.20% 11 Portuguese: 1.20% 12 Hebrew: 0.94% 13 Danish: 0.53% 14 Finnish: 0.52% 15 Esperanto: 0.32% 16 Norwegian: 0.28% 17 Russian: 0.25% 18 Arabic: 0.25% 19 Slovene: 0.23% 20 Catalan: 0.21% 21 Korean: 0.16% 22 Malaysian: 0.16% 23 Bulgarian: 0.15% 24 Romanian: 0.15% 25 Simple: 0.14% 26 Czech: 0.13% 27 Hungarian: 0.13% 28 Indonesian: 0.13% 29 Estonian: 0.12% 30 Interlingua: 0.12% 31 Croatian: 0.11% 32 Turkish: 0.10% 33 Farsi: 0.10%
This still raises some questions; for example the Catalan WP has over 20k articles while the Arabic WP has just over 10k; numbers for Chinese, Spanish, and even Japanese (!) are similarly incongruent.
Mark
On 29/03/06, Anthere Anthere9@yahoo.com wrote:
Mark Williamson wrote:
Hi everybody,
While it's sort of obvious, given the digital divide, that the number of articles in Wikipedias is not proportional to the number of speakers, for example Hindi has a much smaller number of articles compared to speakers than most active Wikipedias; German has more.
However, something that people may not notice as much is the incongruency between popularity of a particular language version and the number of articles in that version.
The most visited Wikipedias, in order, are:
1 English (65%) 2 German (10%) 3 Japanese (6%) 4 Spanish (3%) 5 French (2%) 6 Polish (2%) 7 Chinese (2%) 8 Arabic (2%) 9 Italian (1%) 10 Hebrew (1%) 11 Turkish (1%) 12 Dutch (1%) 13 Portuguese (1%) (all others combined total 1% of visits)
On the other hand, the list of Wikipedias ranked by number of articles is: 1 English (1048.7K) 2 German (376.9K) 3 French (261.1K) 4 Polish (223.8K) 5 Japanese (196.3K) 6 Dutch (156.9K) ... 8 Italian (146.8K) 9 Portuguese (123.8K) 10 Spanish (105.0K) ... 12 Chinese (61.48K) ... 17 Hebrew (34.35K) ... 29 Turkish (19.94K) ... 37 Arabic (12.03K)
Hi
By the way, I have wondered for a long time how it could be that the french wikipedia be the third largest (which is a fact)... and only supposingly 2% of the visits. I also remember that meta had very significantly more visits than the french wikipedia. Which.. to me... is not something I *believe*.
Answer : these values are collected strictly thanks to the use of the Alexa bar...; which is apparently very little used at least in France. Hence possibly messing up the statistics. It might well be the same for other languages, in particular those for which the Alexa bar... is simply not working/translated. So... be careful with the use of the Alexa data. They are a very interesting feedback... but also to take with a pinch of salt.
By the way, there was recently a study by Mediametry (a statistical agency, making polls. In France, probably the largest and most famous one).
The study showed this
déc. 2004 avr. 2005 août 2005 déc. 2005USA 4 544 000 6 753 000 11 874 000 17 498 000 Germany 1 707 000 2 956 000 4 502 000 5 529 000 France 681 000 1 169 000 1 429 000 2 421 000 UK 540 000 766 000 1 525 000 2 298 000
Please see that : http://www.google.com/search?hl=en&q=mediametrie+wikipedia&btnG=Goog...
anthere
Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
-- "Take away their language, destroy their souls." -- Joseph Stalin
Surely the quality of the articles matters more than the quantity (although I appreciate that it does not look how there is such a disparity in numbers).
Oliver
On 3/29/06, Mark Williamson node.ue@gmail.com wrote:
That's certainly a good point.
Unfortunately, the most recent statistics we have for site visits are from October 2004 but they are as follows:
1 English: 42.75% 2 Japanese: 14.72% 3 German: 13.74% 4 Spanish: 5.34% 5 French: 4.14% 6 Polish: 3.93% 7 Dutch: 2.94% 8 Swedish: 1.74% 9 Chinese: 1.42% 10 Italian: 1.20% 11 Portuguese: 1.20% 12 Hebrew: 0.94% 13 Danish: 0.53% 14 Finnish: 0.52% 15 Esperanto: 0.32% 16 Norwegian: 0.28% 17 Russian: 0.25% 18 Arabic: 0.25% 19 Slovene: 0.23% 20 Catalan: 0.21% 21 Korean: 0.16% 22 Malaysian: 0.16% 23 Bulgarian: 0.15% 24 Romanian: 0.15% 25 Simple: 0.14% 26 Czech: 0.13% 27 Hungarian: 0.13% 28 Indonesian: 0.13% 29 Estonian: 0.12% 30 Interlingua: 0.12% 31 Croatian: 0.11% 32 Turkish: 0.10% 33 Farsi: 0.10%
This still raises some questions; for example the Catalan WP has over 20k articles while the Arabic WP has just over 10k; numbers for Chinese, Spanish, and even Japanese (!) are similarly incongruent.
Mark
On 29/03/06, Anthere Anthere9@yahoo.com wrote:
Mark Williamson wrote:
Hi everybody,
While it's sort of obvious, given the digital divide, that the number of articles in Wikipedias is not proportional to the number of speakers, for example Hindi has a much smaller number of articles compared to speakers than most active Wikipedias; German has more.
However, something that people may not notice as much is the incongruency between popularity of a particular language version and the number of articles in that version.
The most visited Wikipedias, in order, are:
1 English (65%) 2 German (10%) 3 Japanese (6%) 4 Spanish (3%) 5 French (2%) 6 Polish (2%) 7 Chinese (2%) 8 Arabic (2%) 9 Italian (1%) 10 Hebrew (1%) 11 Turkish (1%) 12 Dutch (1%) 13 Portuguese (1%) (all others combined total 1% of visits)
On the other hand, the list of Wikipedias ranked by number of articles is: 1 English (1048.7K) 2 German (376.9K) 3 French (261.1K) 4 Polish (223.8K) 5 Japanese (196.3K) 6 Dutch (156.9K) ... 8 Italian (146.8K) 9 Portuguese (123.8K) 10 Spanish (105.0K) ... 12 Chinese (61.48K) ... 17 Hebrew (34.35K) ... 29 Turkish (19.94K) ... 37 Arabic (12.03K)
Hi
By the way, I have wondered for a long time how it could be that the french wikipedia be the third largest (which is a fact)... and only supposingly 2% of the visits. I also remember that meta had very significantly more visits than the french wikipedia. Which.. to me... is not something I *believe*.
Answer : these values are collected strictly thanks to the use of the Alexa bar...; which is apparently very little used at least in France. Hence possibly messing up the statistics. It might well be the same for other languages, in particular those for which the Alexa bar... is simply not working/translated. So... be careful with the use of the Alexa data. They are a very interesting feedback... but also to take with a pinch of salt.
By the way, there was recently a study by Mediametry (a statistical agency, making polls. In France, probably the largest and most famous one).
The study showed this
déc. 2004 avr. 2005 août 2005 déc. 2005USA 4 544 000 6 753 000 11 874 000 17 498 000 Germany 1 707 000 2 956 000 4 502 000 5 529 000 France 681 000 1 169 000 1 429 000 2 421 000 UK 540 000 766 000 1 525 000 2 298 000
Please see that : http://www.google.com/search?hl=en&q=mediametrie+wikipedia&btnG=Goog...
anthere
Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
-- "Take away their language, destroy their souls." -- Joseph Stalin _______________________________________________ Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
Mark Williamson wrote:
This still raises some questions; for example the Catalan WP has over 20k articles while the Arabic WP has just over 10k; numbers for Chinese, Spanish, and even Japanese (!) are similarly incongruent.
To write a wiki with 100K articles, you don't need the 480M speakers of Hindi, the 46M speakers of Polish or the 4.6M speakers of Norwegian. All you need is 20 or 50 very determined people, who really want to write a wiki with 100K articles. This determination is a mindset, which can be very hard to reach. Have you tried to convince your parents to contribute actively to Wikipedia? I suppose that pyramid schemes, religious cults, and even suicide bombings need a similar mindset. Most people just wouldn't get it into their head, that this is a good idea. But once you are determined, the actual job is not so hard to do.
Perhaps we have an understanding of ideas, ideologies, and mindsets today that is similar to the medical understanding of viruses and bacteria 200 years ago.
To write a wiki with 100K articles, you don't need the 480M speakers of Hindi, the 46M speakers of Polish or the 4.6M speakers of Norwegian. All you need is 20 or 50 very determined people, who really want to write a wiki with 100K articles. This determination is a mindset, which can be very hard to reach. Have you tried to convince your parents to contribute actively to Wikipedia? I suppose that pyramid schemes, religious cults, and even suicide bombings need a similar mindset. Most people just wouldn't get it into their head, that this is a good idea. But once you are determined, the actual job is not so hard to do.
Perhaps we have an understanding of ideas, ideologies, and mindsets today that is similar to the medical understanding of viruses and bacteria 200 years ago.
LOL Count this as signed by me, too :) It's an absolutely EXACT picture :) Bèrto
Hoi, I have read the thread as it was published so far and I am amazed that nobody mentioned one simple reason why people do not edit or add content to the Arab, the Farsi, the Hebrew and Assamese projects .. It is too bloody hard. When you say "everybody can edit", it is as if it is the same effort is involved. I read somewhere where an African president said; "we do not have scripts yet for all of our indigenous languages. When the yi.wikipedia celebrated its 1000th article Gangleri was thanked for his hard work to make this technically possible. When I created some Farsi training material on Wikibooks, I needed two browsers to complete certain tasks; both Internet Explorer and Firefox were not up to the task.
Gangleri does a great job, he is imho one of the most valuable Wikimedians because he tries to make it possible to have information in all languages. To take things to the next level, we need more developers; people of all the language families and make sure with them that MediaWiki is up to the task. So far we have been self congratulatory about how well we do. We profess that we want to do better in Africa Asia and South America. We can if we make it a priority.
For me improving these issues /is/ a priority. http://WiktionaryZ.org requires good support for all languages. I am happy that we initiated the "Multilingual Mediawiki" project as it will further improve the multilingual capabilities of MediaWiki. It will still not do all the things that are necessary to make MediaWiki as easy to edit as it is for us. For that I need people that speak Hindi Assamese Twi Farsi Arab Hebrew and help us define what /their /problem with our software is and when we are lucky help us fix these issues.
Thanks, GerardM
Mark Williamson wrote:
Hi everybody,
While it's sort of obvious, given the digital divide, that the number of articles in Wikipedias is not proportional to the number of speakers, for example Hindi has a much smaller number of articles compared to speakers than most active Wikipedias; German has more.
However, something that people may not notice as much is the incongruency between popularity of a particular language version and the number of articles in that version.
The most visited Wikipedias, in order, are:
1 English (65%) 2 German (10%) 3 Japanese (6%) 4 Spanish (3%) 5 French (2%) 6 Polish (2%) 7 Chinese (2%) 8 Arabic (2%) 9 Italian (1%) 10 Hebrew (1%) 11 Turkish (1%) 12 Dutch (1%) 13 Portuguese (1%) (all others combined total 1% of visits)
On the other hand, the list of Wikipedias ranked by number of articles is: 1 English (1048.7K) 2 German (376.9K) 3 French (261.1K) 4 Polish (223.8K) 5 Japanese (196.3K) 6 Dutch (156.9K) ... 8 Italian (146.8K) 9 Portuguese (123.8K) 10 Spanish (105.0K) ... 12 Chinese (61.48K) ... 17 Hebrew (34.35K) ... 29 Turkish (19.94K) ... 37 Arabic (12.03K)
What this says to me is that these Wikipedias are not attracting new pages proportional to views when compared with other Wikipedias. This may be because people don't want to write new pages, but it seems to me more likely that people simply don't know they can.
How can this be fixed? Perhaps a site notice inviting people to write quality pages or register, or a drive to recruit new Wikipedians from the academic community.
Mark
Assuming you're speaking about complex script rendering, part of this is dependent on software -- There is nothing that will make Internet Explorer properly render traditional Mohawk hieroglyphic narrative script, for instance, unless it is represented as an image. Even if a font existed, the rendering is simply so complex that it would really need some programmers to concentrate on it, and there is really no economic motivation for this.
Hebrew is generally well-supported nowadays in software. Arabic, Farsi, Hindi, etc. are a little bit farther behind.
But I think in many of these cases, it's not our issue so much as it is the issue of developers of browers and operating systems. Granted, we do have problems with BiDi, which Gangleri is always working on. But even with that, only a tiny fraction of the problems with BiDi have ever been on our end, many of them simply aren't fixable by us.
That aside, none of the languages of South America that I know of, a minority of the languages of Africa, and really mostly only the languages of South and Southeast Asia (ie, excluding languages of Central Asia, North Asia) are written in complex scripts. This means that population-wise, if we concentrate on complex scripts, we are more specifically concentrating on Indic languages, Southeast Asian languages (Thai, Lao, Khmer, Burmese), minority languages in China (Tibetan, Mongolian), and Arabic-script languages (Arabic, Farsi, Urdu, Uyghur, Balochi, S. Azeri). Certainly these are very population-heavy languages. But I think the root of the issue is internet access. With many of these languages, it seems that people are trying to resolve script problems before access is really widespread, which is certainly lamentable, but at the same time makes it clear that problems with rendering are certainly not one of the major reasons for current numbers for most of these languages.
Mark
On 05/04/06, Gerard Meijssen gerard.meijssen@gmail.com wrote:
Hoi, I have read the thread as it was published so far and I am amazed that nobody mentioned one simple reason why people do not edit or add content to the Arab, the Farsi, the Hebrew and Assamese projects .. It is too bloody hard. When you say "everybody can edit", it is as if it is the same effort is involved. I read somewhere where an African president said; "we do not have scripts yet for all of our indigenous languages. When the yi.wikipedia celebrated its 1000th article Gangleri was thanked for his hard work to make this technically possible. When I created some Farsi training material on Wikibooks, I needed two browsers to complete certain tasks; both Internet Explorer and Firefox were not up to the task.
Gangleri does a great job, he is imho one of the most valuable Wikimedians because he tries to make it possible to have information in all languages. To take things to the next level, we need more developers; people of all the language families and make sure with them that MediaWiki is up to the task. So far we have been self congratulatory about how well we do. We profess that we want to do better in Africa Asia and South America. We can if we make it a priority.
For me improving these issues /is/ a priority. http://WiktionaryZ.org requires good support for all languages. I am happy that we initiated the "Multilingual Mediawiki" project as it will further improve the multilingual capabilities of MediaWiki. It will still not do all the things that are necessary to make MediaWiki as easy to edit as it is for us. For that I need people that speak Hindi Assamese Twi Farsi Arab Hebrew and help us define what /their /problem with our software is and when we are lucky help us fix these issues.
Thanks, GerardM
Mark Williamson wrote:
Hi everybody,
While it's sort of obvious, given the digital divide, that the number of articles in Wikipedias is not proportional to the number of speakers, for example Hindi has a much smaller number of articles compared to speakers than most active Wikipedias; German has more.
However, something that people may not notice as much is the incongruency between popularity of a particular language version and the number of articles in that version.
The most visited Wikipedias, in order, are:
1 English (65%) 2 German (10%) 3 Japanese (6%) 4 Spanish (3%) 5 French (2%) 6 Polish (2%) 7 Chinese (2%) 8 Arabic (2%) 9 Italian (1%) 10 Hebrew (1%) 11 Turkish (1%) 12 Dutch (1%) 13 Portuguese (1%) (all others combined total 1% of visits)
On the other hand, the list of Wikipedias ranked by number of articles is: 1 English (1048.7K) 2 German (376.9K) 3 French (261.1K) 4 Polish (223.8K) 5 Japanese (196.3K) 6 Dutch (156.9K) ... 8 Italian (146.8K) 9 Portuguese (123.8K) 10 Spanish (105.0K) ... 12 Chinese (61.48K) ... 17 Hebrew (34.35K) ... 29 Turkish (19.94K) ... 37 Arabic (12.03K)
What this says to me is that these Wikipedias are not attracting new pages proportional to views when compared with other Wikipedias. This may be because people don't want to write new pages, but it seems to me more likely that people simply don't know they can.
How can this be fixed? Perhaps a site notice inviting people to write quality pages or register, or a drive to recruit new Wikipedians from the academic community.
Mark
Wikipedia-l mailing list Wikipedia-l@Wikimedia.org http://mail.wikipedia.org/mailman/listinfo/wikipedia-l
-- "Take away their language, destroy their souls." -- Joseph Stalin
Mark, I do not care for Internet Explorer until I can get the ear of a developer at Microsoft, this should however be doable. I am convinced that we can get attention at the Mozilla Foundation. But most of all I want MediaWiki to just work and work well for all languages. You know how difficult the first editing efforts are. When it becomes really screwy only a few will preservere. As to economic motivation, when we can create a cool project we can find money for things like fonts and scripts. It then becomes a question if someone does it for free or for a fee. The other cost is that some languages are just not economic and therefore without merit; this is an argument Microsoft loved to not have gotten into in many instances...
What we can do as a Wikimedia Foundation is limited. We cannot bring a computer to every house of this world but we can make our software ready for the moment when the people from these houses find their way to a computer.
Thanks, GerardM
Mark Williamson wrote:
Assuming you're speaking about complex script rendering, part of this is dependent on software -- There is nothing that will make Internet Explorer properly render traditional Mohawk hieroglyphic narrative script, for instance, unless it is represented as an image. Even if a font existed, the rendering is simply so complex that it would really need some programmers to concentrate on it, and there is really no economic motivation for this.
Hebrew is generally well-supported nowadays in software. Arabic, Farsi, Hindi, etc. are a little bit farther behind.
But I think in many of these cases, it's not our issue so much as it is the issue of developers of browers and operating systems. Granted, we do have problems with BiDi, which Gangleri is always working on. But even with that, only a tiny fraction of the problems with BiDi have ever been on our end, many of them simply aren't fixable by us.
That aside, none of the languages of South America that I know of, a minority of the languages of Africa, and really mostly only the languages of South and Southeast Asia (ie, excluding languages of Central Asia, North Asia) are written in complex scripts. This means that population-wise, if we concentrate on complex scripts, we are more specifically concentrating on Indic languages, Southeast Asian languages (Thai, Lao, Khmer, Burmese), minority languages in China (Tibetan, Mongolian), and Arabic-script languages (Arabic, Farsi, Urdu, Uyghur, Balochi, S. Azeri). Certainly these are very population-heavy languages. But I think the root of the issue is internet access. With many of these languages, it seems that people are trying to resolve script problems before access is really widespread, which is certainly lamentable, but at the same time makes it clear that problems with rendering are certainly not one of the major reasons for current numbers for most of these languages.
Mark
On 05/04/06, Gerard Meijssen gerard.meijssen@gmail.com wrote:
Hoi, I have read the thread as it was published so far and I am amazed that nobody mentioned one simple reason why people do not edit or add content to the Arab, the Farsi, the Hebrew and Assamese projects .. It is too bloody hard. When you say "everybody can edit", it is as if it is the same effort is involved. I read somewhere where an African president said; "we do not have scripts yet for all of our indigenous languages. When the yi.wikipedia celebrated its 1000th article Gangleri was thanked for his hard work to make this technically possible. When I created some Farsi training material on Wikibooks, I needed two browsers to complete certain tasks; both Internet Explorer and Firefox were not up to the task.
Gangleri does a great job, he is imho one of the most valuable Wikimedians because he tries to make it possible to have information in all languages. To take things to the next level, we need more developers; people of all the language families and make sure with them that MediaWiki is up to the task. So far we have been self congratulatory about how well we do. We profess that we want to do better in Africa Asia and South America. We can if we make it a priority.
For me improving these issues /is/ a priority. http://WiktionaryZ.org requires good support for all languages. I am happy that we initiated the "Multilingual Mediawiki" project as it will further improve the multilingual capabilities of MediaWiki. It will still not do all the things that are necessary to make MediaWiki as easy to edit as it is for us. For that I need people that speak Hindi Assamese Twi Farsi Arab Hebrew and help us define what /their /problem with our software is and when we are lucky help us fix these issues.
Thanks, GerardM
Mark Williamson wrote:
Hi everybody,
While it's sort of obvious, given the digital divide, that the number of articles in Wikipedias is not proportional to the number of speakers, for example Hindi has a much smaller number of articles compared to speakers than most active Wikipedias; German has more.
However, something that people may not notice as much is the incongruency between popularity of a particular language version and the number of articles in that version.
The most visited Wikipedias, in order, are:
1 English (65%) 2 German (10%) 3 Japanese (6%) 4 Spanish (3%) 5 French (2%) 6 Polish (2%) 7 Chinese (2%) 8 Arabic (2%) 9 Italian (1%) 10 Hebrew (1%) 11 Turkish (1%) 12 Dutch (1%) 13 Portuguese (1%) (all others combined total 1% of visits)
On the other hand, the list of Wikipedias ranked by number of articles is: 1 English (1048.7K) 2 German (376.9K) 3 French (261.1K) 4 Polish (223.8K) 5 Japanese (196.3K) 6 Dutch (156.9K) ... 8 Italian (146.8K) 9 Portuguese (123.8K) 10 Spanish (105.0K) ... 12 Chinese (61.48K) ... 17 Hebrew (34.35K) ... 29 Turkish (19.94K) ... 37 Arabic (12.03K)
What this says to me is that these Wikipedias are not attracting new pages proportional to views when compared with other Wikipedias. This may be because people don't want to write new pages, but it seems to me more likely that people simply don't know they can.
How can this be fixed? Perhaps a site notice inviting people to write quality pages or register, or a drive to recruit new Wikipedians from the academic community.
Mark
Hi!
What we can do as a Wikimedia Foundation is limited. We cannot bring a computer to every house of this world but we can make our software ready for the moment when the people from these houses find their way to a computer.
A perfect phrase to describe what wikimedia is doing. May we quote it the pms edition? We are getting a page ready, to explain to the people what's this all about.
Consider it a request for you to declare your phrase a PD statement :))
Bèrto
Hoi, Obviously yes. Nice that you like my "flowery language" :) GerardM
Berto wrote:
Hi!
What we can do as a Wikimedia Foundation is limited. We cannot bring a computer to every house of this world but we can make our software ready for the moment when the people from these houses find their way to a computer.
A perfect phrase to describe what wikimedia is doing. May we quote it the pms edition? We are getting a page ready, to explain to the people what's this all about.
Consider it a request for you to declare your phrase a PD statement :))
Bèrto
On 3/29/06, Mark Williamson node.ue@gmail.com wrote:
The most visited Wikipedias, in order, are:
1 English (65%) 2 German (10%) 3 Japanese (6%) 4 Spanish (3%) 5 French (2%) 6 Polish (2%) 7 Chinese (2%) 8 Arabic (2%) 9 Italian (1%) 10 Hebrew (1%) 11 Turkish (1%) 12 Dutch (1%) 13 Portuguese (1%) (all others combined total 1% of visits)
I'm interested where you found those numbers, because I thought they were not available anywhere. :-)
Delphine -- ~notafish
Delphine Ménard wrote:
On 3/29/06, Mark Williamson node.ue@gmail.com wrote:
The most visited Wikipedias, in order, are:
1 English (65%) 2 German (10%) 3 Japanese (6%) 4 Spanish (3%) 5 French (2%) 6 Polish (2%) 7 Chinese (2%) 8 Arabic (2%) 9 Italian (1%) 10 Hebrew (1%) 11 Turkish (1%) 12 Dutch (1%) 13 Portuguese (1%) (all others combined total 1% of visits)
I'm interested where you found those numbers, because I thought they were not available anywhere. :-)
Delphine
~notafish
Alexa...
http://www.alexa.com/data/details/traffic_details?q=&url=www.wikipedia.o...
And as I already said, results are questionable :-)
ant
On 4/6/06, Anthere Anthere9@yahoo.com wrote:
Alexa...
http://www.alexa.com/data/details/traffic_details?q=&url=www.wikipedia.o...
And as I already said, results are questionable :-)
If that is the case, basing a "study" on those figures is probably indeed questionnable.
Alexa only provides their toolbar, (with which they collect data) [1] in English. And only for Internet Explorer.
Which probably gives a good overview of the English part of Wikipedia/Wikimedia projects, but definitely gives only a partial, or even totally biased view of other languages and websites. And leaves out a pretty good section of the population using other browsers than Internet Explorer.
This said, I find the idea behind the "study" definitely interesting, and it *does* call for *real* statistics... ;-)
Delphine
[1]About the Alexa Toolbar Alexa could not exist without the participation of the Alexa Toolbar community. Each member of the community, in addition to getting a useful tool, is giving back. Simply by using the toolbar each member contributes valuable information about the web, how it is used, what is important and what is not. This information is returned to the community as Related Links, Traffic Rankings and more. from http://pages.alexa.com/company/index.html -- ~notafish
wikipedia-l@lists.wikimedia.org