I have not seen a comprehensive overview of MediaWiki localisation discussed on the lists I am posting this message to, so I thought I might give it a try. All statistics are based on MediaWiki 1.12 alpha, SVN version r29106.
==Introduction== *Localisation or L10n - the process of adapting the software to be as familiar as possible to a specific locale (in scope) *Internationalisation or i18n - the process of ensuring that an application is capable of adapting to local requirements (out of scope)
MediaWiki has a user interface (UI) definition for 319 languages. Of those languages at least 17 language codes are duplicates and/or serve a purpose for usability[1]. Reporting on them, however, is not relevant. So MediaWiki in its current state supports 302 languages. To be able to generate statistics on localisation, a MessagesXx.php file should be present in languages/messages. There currently are 262 such files, of which 16 are redirects from the duplicates/usability group[2]. So MediaWiki has an active in-product localisation for 236 languages. 66 languages have an interface, but simply fall back to English.
The MediaWiki core product recognises several collections of localisable content (three of which are defined in messageTypes.inc): * 'normal' messages that can be localised (1726) * optional messages that can be localised, which usually only happens for languages not using a Latin script (161) * ignored messages that should not be localised (100) * namespace names and namespace aliases (17) * skin names (7) * magic words (120) * special page names (76) * other (directionality, date formats, separators, book store lists, link trail, and others)
Localisation of MediaWiki revolves around all of the above. Reporting is done on the normal messages only.
MediaWiki is more than just the core product. On http://www.mediawiki.org/wiki/Category:All_extensions some 750 extensions have some kind of documentation. This analysis will scope only to the code currently present in svn.wikimedia.org/svnroot/mediawiki/trunk. The source code repository contains give or take 230 extensions. Of those 230 extensions, about 140 contain messages that can be visible in the UI in some use case (debugging excluded). Out of those 140, about 10 extensions have an exotic implementation for localisation localisation support at all (just English text in the code). 10 extensions appear to be outdated. I have seen about 5 different 'standard' implementations of i18n in extensions. Since MediaWiki 1.11 there is wfLoadExtensionMessages. Not that many extensions use this yet for message handling. If you can help add more standard i18n support for extensions (an overview can be found at http://translatewiki.net/wiki/User:Siebrand/tobeadded) or help in standardising L10n for extensions, please do not hesitate.
==MediaWiki localisation in practice== Localisation of MediaWiki is currently done in the following ways I am aware of: * in local wikis: Sysops on local wikis shape and translate messages to fit their needs. This is being done in wikis that are part of Wikimedia, Wikia, Wikitravel, corporate wikis, etc. This type of localisation has the fewest benefits for the core product and extensions because it happens completely out of the scope of svn committers. I have heard Wikia supports languages that are not supported in the svn version. I would like to get some help in identifying and contacting these communities to try and get their localisations in the core product. Together with SPQRobin, I am trying to get what has been localised in local Wikipedias into the core product and recruit users that worked on the localisation to work on a more centralised way of localisation (see Betawiki) * through bugzilla/svn: A user of MediaWiki submits patches for core messages and/or extensions. These users are mostly part of a wiki community that is part of Wikimedia. These are usually taken care of by committers raymond, rotemliss, and sometimes others). Some users maintain a language directly on SVN. At the moment, 10-15 languages are maintained this way: Danish, German, Persian, Hebrew, Indonesian, Kazach (3 scripts), Chinese (3 variants), and some more less frequently. * through Betawiki: Betawiki was founded in mid 2005 by Niklas Laxström. In the years to follow, Betawiki has grown to be a MediaWiki localisation community with over 200 users that has contributed to the localisation of 120 languages each month in the past few months. Users that are only familiar with MediaWiki as a tool can localise almost every aspect of MediaWiki (except for the group 'other' mentioned earlier) in a wiki interface. The work of the translators is regularly committed to svn by nikerabbit, and myself. Betawiki also offers a .po export that enables users to use more advanced translation tools to make their translation. This option was added recently and no translations in this format have been sumitted yet. Betawiki also supports translation of 122 extensions, aiming to support everything that can be supported.
==MediaWiki localisation statistics== MediaWiki localisation statistics have been around since June 2005 at http://www.mediawiki.org/wiki/Localisation_statistics%5B3]. Traditionally reports have focused on the complete set of core messages. Recently a small study was done after usage of messages, which resulted in a set of almost 500 'most often used messages in MediaWiki', based on usage of messages on the cluster of Wikimedia (http://translatewiki.net/wiki/Most_often_used_messages_in_MediaWiki).
Up to recently there were no statistics available on the localisation of extensions. Through groupStatistics.php in the extension Translate, these statistics can now be created. Aside from reporting on 'most often used MediaWiki messages', 'MediaWiki messages', and 'all extension messages supported by extension Translate' (short: extension messages). Additionally, a meta extension group of 34 extensions used in the projects of Wikimedia has been created (short: Wikimedia messages). A regularly updated table of these statistics can be found at http://translatewiki.net/wiki/Translating:Group_statistics.
Some (arbitrary) milestones have been set for the four above mentioned collections of messages. For the usability of MediaWiki in a particular language, the group 'core-mostused' is the most important. A language must qualify for MediaWiki to have minimal support for that language. Reaching the milestones for the first two groups is something the Wikimedia language committee considers to use as a requirement for new Wikimedia wikis: * core-mostused (496 messages): 98% * wikimedia extensions (354 messages): 90% * core (1726 messages): 90% * extensions (1785 messages): 65%
Currently the following numbers of languages have passed the above milestones: * core-mostused: 47 (15,5% of supported languages) * wikimedia extensions: 10 (3,3% of supported languages) * core: 49 (16,2% of supported languages) * extensions: 7 (2,3% of supported languages)
==Conclusion== So... Are we doing well on localisation or do we suck? My personal opinion is that we do something in between. Observing that there are some 250 Wikipedias that all use the Wikimedia Commons media repository, and that only 47 languages have a minimal localisation, we could do better. With Single User Login around the corner (isn't it), we must do better. On the other hand, new language projects within Wikimedia all have excellent localisation of the core product. These languages include Asturian, Bikol Central, Lower Sorbian, Extremaduran, and Galician. But where is Hindi, for example, with currently only 7% of core messages translated?
With the Wikimedia Foundation aiming to put MediaWiki to good use in developing countries and products like NGO-in-a-box that include MediaWiki, the potential of MediaWiki as a tool in creating and preserving knowledge in the languages of the world is huge. We have to tap into that potential and *you* (yes, I am glad you read this far and are now reading my appeal) can help. If you know people that are proficient in a language and like contributing to localisation, please point them in the right direction. If you know of organisations that can help localising MediaWiki: please approach them and ask them to help.
We have all the tools now to successfully localise MediaWiki into any of the 7000 or so languages that have been classified in ISO 639-3. We only need one person per language to make it happen. Reaching the first two milestones (core-mostused and wikimedia extensions) takes about 16 hours of work. Using Betawiki or the .po, little to no technical knowledge is required.
This was the pitch. How about we aim to at least double the numbers by the end of 2008 to: * core-mostused: 120 * wikimedia extensions: 50 * core: 90 * extensions: 20
I would like to wish everyone involved in any aspect of MediaWiki a wonderful 2008.
Cheers!
Siebrand Mazeland
[1] als,crh,iu,kk,kk-cn,kk-kz,kk-tr,ku,sr,sr-jc,sr-jl,zh,zh-cn,zh-sg,zh-hk,zh-min-nan,zh-yue [2] crh,iu,kk,kk-cn,kk-kz,kk-tr,ku,sr,sr-jc,sr-jl,zh,zh-cn,zh-sg,zh-hk,zh-min-nan,zh-yue [3] older locations are http://www.mediawiki.org/wiki/Localisation_statistics/stats and http://meta.wikimedia.org/wiki/Localization_statistics
Very interesting report. Unfortunately I can't help out myself with translation, and don't know anyone who can, but if something is needed in terms of software that I can help out with, tell me and I'll see if I can spare the time.
"Siebrand Mazeland" s.mazeland@xs4all.nl writes:
- through Betawiki: Betawiki was founded in mid 2005 by Niklas
...
the translators is regularly committed to svn by nikerabbit, and myself. Betawiki also offers a .po export that enables users to use
Considering the quality of your commits to MessagesDa.php, I consider Betawiki a poor substitute for a maintainer that actually speaks the language in question.
Anders Wegge Jakobsen schreef:
Considering the quality of your commits to MessagesDa.php, I consider Betawiki a poor substitute for a maintainer that actually speaks the language in question.
Keep in mind that the one who enters these translations at BetaWiki thereby more or less claims to speak Danish. If you think you translate better, go ahead to BetaWiki and show us ;)
Roan Kattouw (Catrope)
Roan Kattouw roan.kattouw@home.nl writes:
Anders Wegge Jakobsen schreef:
Considering the quality of your commits to MessagesDa.php, I consider Betawiki a poor substitute for a maintainer that actually speaks the language in question.
Keep in mind that the one who enters these translations at BetaWiki thereby more or less claims to speak Danish.
Exactly. Sadly many of them don't.
If you think you translate better, go ahead to BetaWiki and show us ;)
If you feel that the betawiki-process produce a better translation than the one I maintain, I'll be happy to stop comitting my work.
Anders Wegge Jakobsen schreef:
If you feel that the betawiki-process produce a better translation than the one I maintain, I'll be happy to stop comitting my work.
Oh, you're actually committing to MessagesDa.php ? Didn't know that. Anyway, there's no real difference between bad Danish getting in through Betawiki and bad Danish being committed directly. I'm sure Siebrand can find out who's responsible for the bad translations.
Roan Kattouw (Catrope)
On 12/31/07, Anders Wegge Jakobsen wegge@wegge.dk wrote:
Roan Kattouw roan.kattouw@home.nl writes:
Keep in mind that the one who enters these translations at BetaWiki thereby more or less claims to speak Danish.
Exactly. Sadly many of them don't.
Perhaps it should be made clearer that you should only enter any translations on BetaWiki if you are either a native or near-native speaker of the language, not just if you took a few classes in the language or spent a year in the country or something. I'm sure many people are happy to help out in whatever way is possible and view it as a nice way to brush up on the languages they know, and don't realize what a poor translation they're adding. (I'm thinking something like Engrish here -- is that about right?)
On 31/12/2007, Anders Wegge Jakobsen wegge@wegge.dk wrote:
Roan Kattouw roan.kattouw@home.nl writes:
Anders Wegge Jakobsen schreef:
Considering the quality of your commits to MessagesDa.php, I consider Betawiki a poor substitute for a maintainer that actually speaks the language in question.
Are you blaming the tool or the translators or our politics? I see no problem with the tool. Assuming that there is translator(s) X. X should be able to produce better translations with the tool than without it. Especially if there is many X.
What comes to translators and politics... I try to not get too involved. See also my reply to similar question in [1]. In short, anyone is allowed to translate. If the quality is poor - this is OS - then fix it.
The problem with your case seems to be that (1) we have someone on betawiki who wishes to translate into Danish (2) you prefer to work without betawiki, and these two combined just doesn't work. I would suggest the obvious solution here, but all I can do is to ask you by using nice words.
Keep in mind that the one who enters these translations at BetaWiki thereby more or less claims to speak Danish.
Exactly. Sadly many of them don't.
See [1] again why we don't (can't) verify this. This is a social problem, not technical.
If you think you translate better, go ahead to BetaWiki and show us ;)
If you feel that the betawiki-process produce a better translation than the one I maintain, I'll be happy to stop comitting my work.
Your work is greatly appreciated. However languages doesn't have owners. Neither does anyone claim to be working on Danish on betawiki but as we don't have restrictions, some may touch the translations anyway.
We could disable Danish translation on betawiki, and force all contributors to go trough your, but I don't find this solution very good, do you? Actually I'm in the same position, with the difference that I use betawiki. The work done by others is not so high quality as I would like, but I go and fix the poor translations, possibly giving tips to them how they can become better. Being a native speaker doesn't mean you are a good translator. This is also opportunity for them to learn the skill I've acquired during the years. Being a small language, we can't be overly selective about the translators, or we have no translations at all.
But I'm on wrong track already... I honestly believe betawiki helps the translators (and thus mediawiki), I wouldn't have done it otherwise. If you know a better way of solving the conflict between betawiki and you, I'm interested to hear it.
[1] http://translatewiki.net/w/?title=Betawiki_talk:Translator&diff=prev&...
"Niklas Laxström" niklas.laxstrom@gmail.com writes:
On 31/12/2007, Anders Wegge Jakobsen wegge@wegge.dk wrote:
Roan Kattouw roan.kattouw@home.nl writes:
Anders Wegge Jakobsen schreef:
Considering the quality of your commits to MessagesDa.php, I consider Betawiki a poor substitute for a maintainer that actually speaks the language in question.
Are you blaming the tool or the translators or our politics? I see no problem with the tool. Assuming that there is translator(s) X. X should be able to produce better translations with the tool than without it. Especially if there is many X.
I'm blaming the result I see. I assume that the tool actually works, so I'm probably blaming the politics, or the people doing the translations. Likely the latter.
What comes to translators and politics... I try to not get too involved. See also my reply to similar question in [1]. In short, anyone is allowed to translate. If the quality is poor - this is OS
- then fix it.
I'm already doing that. But when I see betawiki spamvertised as the biggest thing since sliced bread, I become a little vexed. It may be a question of personal chemistry, me having a bad day, or some third thing. Whatever it is, I don't feel that the results coming out of betawiki for MessagesDa is reflecting your intentions.
The problem with your case seems to be that (1) we have someone on betawiki who wishes to translate into Danish (2) you prefer to work without betawiki, and these two combined just doesn't work. I would suggest the obvious solution here, but all I can do is to ask you by using nice words.
Yes, the obvious solution is that I maintain a private translation; at the same time one or more of the admins at dawiki plays catchup, whenever the inerface suddenly start sprouting english words, and the rest of the world get to see the worst of OSS. Everyone is happy.
Exactly. Sadly many of them don't.
See [1] again why we don't (can't) verify this. This is a social problem, not technical.
I'd call that process broken by design then.
If you feel that the betawiki-process produce a better translation than the one I maintain, I'll be happy to stop comitting my work.
Your work is greatly appreciated. However languages doesn't have owners. Neither does anyone claim to be working on Danish on betawiki but as we don't have restrictions, some may touch the translations anyway.
I'm not trying to own anything. As I stated above, I'll leave the file be, and then we can see what happens.
...
But I'm on wrong track already... I honestly believe betawiki helps the translators (and thus mediawiki), I wouldn't have done it otherwise. If you know a better way of solving the conflict between betawiki and you, I'm interested to hear it.
As I said above, there is no conflict.
On 12/31/07, Niklas Laxström niklas.laxstrom@gmail.com wrote:
What comes to translators and politics... I try to not get too involved. See also my reply to similar question in [1]. In short, anyone is allowed to translate. If the quality is poor - this is OS - then fix it.
Well, is a bad translation better than no translation? If so, people who are non-native speakers should be actively discouraged from contributing. That would, presumably, be one way to solve this particular problem.
On 12/31/07, Anders Wegge Jakobsen wegge@wegge.dk wrote:
I'm blaming the result I see. I assume that the tool actually works, so I'm probably blaming the politics, or the people doing the translations. Likely the latter.
The part I don't understand is, if someone else is doing bad translations, and you're doing good translations regardless of them, why don't you just replace their bad translations with your good ones and move on? Do they actively undo your good translations? If so, you'll have to talk with them, whether they're using BetaWiki or anything else. Maybe if not, you should also talk with them. Non-Danish speakers can't tell the difference between a good translation and a bad one or even a gibberish one, so the Danish speakers (or self-proclaimed Danish speakers) need to work out amongst themselves who should do the translations. This applies, again, regardless of what medium is used to submit the translations.
I'm sure nobody wants you to stop translating because someone else is submitting bad translations -- we would all like it if the matter could be resolved so as to ensure the best possible localization. Unfortunately we can't do much to help that, except ask you and the other translator(s) to work it out.
Simetrical Simetrical+wikilist@gmail.com writes:
On 12/31/07, Anders Wegge Jakobsen wegge@wegge.dk wrote:
I'm blaming the result I see. I assume that the tool actually works, so I'm probably blaming the politics, or the people doing the translations. Likely the latter.
The part I don't understand is, if someone else is doing bad translations, and you're doing good translations regardless of them, why don't you just replace their bad translations with your good ones and move on?
I have, until now. I just happened to be somewhat PO'ed about being told between the lines of a spamvertising for betawiki, that I didn't do a proper job. So now I'm looking forward to see high-quality translations coming from forcing a particular development tool onto people.
Anders Wegge Jakobsen wegge@wegge.dk writes:
Six weeks ago I became involved in an argument about translation of the mediawiki software. I made the comment below.
Yes, the obvious solution is that I maintain a private translation; at the same time one or more of the admins at dawiki plays catchup, whenever the inerface suddenly start sprouting english words, and the rest of the world get to see the worst of OSS. Everyone is happy.
Unfortunately, it have proven to be true. No substantial changes to the danish localization have happened since then. Since I have cooled of a bit since then, and my prediction have proven true, I'll try to summarize the problems with localization on translatewiki as I see them:
* With the current setup, translators will need to access the code, to actually see what cryptical strings like 'You have not specified target revision or revi sions to perform this function on.' actually mean.
* That a web interface exists does not equal that a large horde of skilled translators will be attracted.
* Noone likes to see others credited with their work.
This is not an attempt to renew a heated argument. The idea of providing a relative easy-to-use interface for translation work is better than having no one translating the interface into any particular language. But in my opinion, it is not at present time a substitute for having someone wit at least rudimentary PHP coding skills doing the translation and submitting patches or direct commits to svn.
And yes, the issue of crediting work was what angered me most. It still is, and unless I'm the one individual in the world with the thinnest skin on this matter, this issue will arise again.
Hi Wegge,
Contrary to your belief I think you I are still quite angry at something that happended or something that someone did or wrote. I have problem identifying what it is exactly, which makes it hard to address. I find few of the statements you made below to be true.
Can you let us know what you would need to be satisfied, or in your eyes be properly credited for everything you have done for the Danish translation? Your current choice of debate does not strike me as solution driven, which is something I personally very much prefer.
Kind regards,
Siebrand Mazeland
-----Oorspronkelijk bericht----- Van: wikitech-l-bounces@lists.wikimedia.org [mailto:wikitech-l-bounces@lists.wikimedia.org] Namens Anders Wegge Jakobsen Verzonden: maandag 18 februari 2008 11:25 Aan: Wikimedia developers; Wikimedia Translators; MediaWiki internationalisation Onderwerp: Re: [Wikitech-l] An update on localisation in MediaWiki
Anders Wegge Jakobsen wegge@wegge.dk writes:
Six weeks ago I became involved in an argument about translation of the mediawiki software. I made the comment below.
Yes, the obvious solution is that I maintain a private translation; at the same time one or more of the admins at dawiki plays catchup, whenever the inerface suddenly start sprouting english words, and the rest of the world get to see the worst of OSS. Everyone is happy.
Unfortunately, it have proven to be true. No substantial changes to the danish localization have happened since then. Since I have cooled of a bit since then, and my prediction have proven true, I'll try to summarize the problems with localization on translatewiki as I see them:
* With the current setup, translators will need to access the code, to actually see what cryptical strings like 'You have not specified target revision or revi sions to perform this function on.' actually mean.
* That a web interface exists does not equal that a large horde of skilled translators will be attracted.
* Noone likes to see others credited with their work.
This is not an attempt to renew a heated argument. The idea of providing a relative easy-to-use interface for translation work is better than having no one translating the interface into any particular language. But in my opinion, it is not at present time a substitute for having someone wit at least rudimentary PHP coding skills doing the translation and submitting patches or direct commits to svn.
And yes, the issue of crediting work was what angered me most. It still is, and unless I'm the one individual in the world with the thinnest skin on this matter, this issue will arise again.
-- // Wegge http://geowiki.wegge.dk/wiki/Forside - Alt om geocaching Bruger du den gratis spamfighther ser jeg kun dine indlæg *EN* gang.
-- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
_______________________________________________ Wikitech-l mailing list Wikitech-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-l
"Siebrand Mazeland" s.mazeland@xs4all.nl writes:
Contrary to your belief I think you I are still quite angry at something that happended or something that someone did or wrote. I have problem identifying what it is exactly, which makes it hard to address. I find few of the statements you made below to be true.
Yes, I'm still angry about some of the comments. I will be for eternity. But since that won't change, just forget it. Neither you, nor Niklas is to be blamed for someone else making flippant remarks.
Can you let us know what you would need to be satisfied, or in your eyes be properly credited for everything you have done for the Danish translation? Your current choice of debate does not strike me as solution driven, which is something I personally very much prefer.
The reason I may seem un-constructive, is the simple fact that while I'm able to point out the problems as I see them, I have no idea how to solve them. If we leave my ego out of the equation for the moment, the main issue is that at some of the interface messages will be completely unknown and opaque to the translators. The obvious solution to that is instrumenting the code, so that any message can be seen in its context. That is going to take quite a lot of time, and not something I think will happen right away.
More realistic, would be crafting a set of more or less static pages, that displays all of the messages in the contexts they are used. That will ba a game of constant catchup, but at least it will be easier than to change the entire codebase to include a demo feature of sorts.
Hi Wegge,
Thank you for your clarification. No system is perfect. I am of the opinion though that Translate can give more insight and information which leads to a more effective and efficient translation process than keeping an eye on SVN commits. Our opinions differ, obviously, so no need to elaborate on that. Instead, I chose to inform you. Please read on.
Extension Translate currently offers a way to add translation help in the language 'qqq'. Those hints are displayed in the Betawiki UI and are also exported to the .po files for offline translation. If the translation help is written correctly, translators would have all the context they need. All this is being done 'wiki-style', so improvements must be made and are being made.
An example of a translation hint is http://translatewiki.net/wiki/MediaWiki:Undeletelink/qqq for a message that was added recently.
Currently 588 of 1766 core messages have some form of documentation[1]. For extensions 99 messages have been documented. Making the sets complete is a lot of work. Currently about 10-15 messages are documented every week[2]. I would love more developer types like you to contribute on localisation by adding message documentation (basically it is a part of i18n). Please see this as an invitation. Anyone with the translator role can add such messages in Betawiki.
An example of a translation hint in Translate context can be seen at http://translatewiki.net/wiki/Image:Translation_hint_example.png. Additional fallback languages while translating are a second instrument we use to make life easier for translators[3].
I hope I have given you and others some additional insight in the workings of Betawiki with the above.
Cheers! Siebrand
[1] http://translatewiki.net/w/i.php?title=Special%3ATranslate&task=reviewal... [2] http://translatewiki.net/w/i.php?days=14&limit=250&title=Special%3AR... [3] http://translatewiki.net/wiki/Image:Translation_fallback_example.png
It seems these are complicated issues, so maybe it's better to look at small steps in the right direction, rather than something that will solve everything at once. i.e.
* Have people stop committing other people's translations under their own name, instead only use something like a bot account, that obviously isn't taking credit; maybe even better if the bot could do a commit per translator, and list the translator in the comment.
* Have an easy interface to go from a special page -> system message, such as a hook that would allow rendering the page such that each message is marked, and links back to somewhere useful.
* Try to generate some kind of "what links here" for messages. Maybe use a maintenance script to generate each special page, so the message function can log the page/message; render examples for extensions from their parser tests, etc.
Hi Anders Wegge Jakobsen.
First of all the best wishes for 2008 to you.
Please become a part of the Betawiki community so that we can address your concerns. At the moment you appear to have some reservations that are to vague to address at this point in time.
Cheers! Siebrand
-----Oorspronkelijk bericht----- Van: wikitech-l-bounces@lists.wikimedia.org [mailto:wikitech-l-bounces@lists.wikimedia.org] Namens Anders Wegge Jakobsen Verzonden: maandag 31 december 2007 19:18 Aan: Wikimedia developers CC: 'MediaWiki internationalisation'; 'Wikimedia Translators' Onderwerp: Re: [Wikitech-l] An update on localisation in MediaWiki
"Siebrand Mazeland" s.mazeland@xs4all.nl writes:
- through Betawiki: Betawiki was founded in mid 2005 by Niklas
...
the translators is regularly committed to svn by nikerabbit, and myself. Betawiki also offers a .po export that enables users to use
Considering the quality of your commits to MessagesDa.php, I consider Betawiki a poor substitute for a maintainer that actually speaks the language in question.
-- // Wegge http://geowiki.wegge.dk/wiki/Forside - Alt om geocaching Bruger du den gratis spamfighther ser jeg kun dine indlæg *EN* gang.
-- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
_______________________________________________ Wikitech-l mailing list Wikitech-l@lists.wikimedia.org http://lists.wikimedia.org/mailman/listinfo/wikitech-l
"Siebrand Mazeland" s.mazeland@xs4all.nl writes:
Hi Anders Wegge Jakobsen.
First of all the best wishes for 2008 to you.
Please become a part of the Betawiki community so that we can address your concerns. At the moment you appear to have some reservations that are to vague to address at this point in time.
Plain and simple NO!
I'm a software developer, and I'm not going to confine myself to a web interface. I fully accept that this is where the bees knees are, so I'm not going to stand in the way of progress.
Hi Wegge.
Too bad you appear to be not willing to be more specific (my interpretation) and enter a dialogue. If you would come to *any* platform to address specific your concerns, I am certain that you would find that we are *very* open to any concerns.
As I already explained: i18n is for developers, L10n is for translators. If you choose to be a developer, why do you choose to translate!?
Repeating myself: *Localisation or L10n - the process of adapting the software to be as familiar as possible to a specific locale *Internationalisation or i18n - the process of ensuring that an application is capable of adapting to local requirements
Please acknowledge that I do not wish to depreciate your efforts in any way. On the contrary: I value *and* appreciate *any* contribution to MediaWiki localisation.
Kind regards,
Siebrand
-----Oorspronkelijk bericht----- Van: wikitech-l-bounces@lists.wikimedia.org [mailto:wikitech-l-bounces@lists.wikimedia.org] Namens Anders Wegge Jakobsen Verzonden: dinsdag 1 januari 2008 1:55 Aan: Wikimedia developers CC: 'MediaWiki internationalisation'; 'Wikimedia Translators' Onderwerp: Re: [Wikitech-l] An update on localisation in MediaWiki
<snip>
Plain and simple NO!
I'm a software developer, and I'm not going to confine myself to a web interface. I fully accept that this is where the bees knees are, so I'm not going to stand in the way of progress. <snip>
"Siebrand Mazeland" s.mazeland@xs4all.nl writes:
Hi Wegge.
Too bad you appear to be not willing to be more specific (my interpretation) and enter a dialogue. If you would come to *any* platform to address specific your concerns, I am certain that you would find that we are *very* open to any concerns.
So far my experience have been the opposite. But my specific concerns is *any* commit that you have made, that involves MessagesDa.
As I already explained: i18n is for developers, L10n is for translators. If you choose to be a developer, why do you choose to translate!?
Because noone else does. In addition to that, My personal tak is that people without some sort of devleoper skill is very ill-suited to do neither translation, nor localization. The distinction is in and of itself bogus.
Repeating myself: *Localisation or L10n - the process of adapting the software to be as familiar as possible to a specific locale *Internationalisation or i18n - the process of ensuring that an application is capable of adapting to local requirements
Yes, you already said that once. It's a false premise.
Please acknowledge that I do not wish to depreciate your efforts in any way. On the contrary: I value *and* appreciate *any* contribution to MediaWiki localisation.
Please acknowledge that I strongly urge you to improve your communication skills. You just lost mediawiki it's only (at the time being) translator to Danish.
<snip>
Please acknowledge that I strongly urge you to improve your communication skills.
If there are communication difficulties, It is to be expected, Many of us don't share en as the default language, And we use it as common ground, So being more tolerant is the way to go.
You just lost mediawiki it's only (at the time
being) translator to Danish.
You said there are other translations, So there are other translators. You are saying that other translations aren't as good as it should be. So you you leave and leave bad translations (as you say)? That is a negative attitude if I may say. You should talk with the other translators (or the the people who edit the danish file) and tell them what is wrong or why you think it isn't optimal, That could be done on wiki because it contains talk pages (pages in general) where you can share your thoughts (what cannot be done as good through svn) and reach some sort of agreement..
Another thing, If you want to get something done (make mw localized), You have to promote it.
HNYBTWHYIG!
Siebrand Mazeland wrote:
As I already explained: i18n is for developers, L10n is for translators. If you choose to be a developer, why do you choose to translate!?
"Translator" is not only a BetaWiki user. I (and other people) translate directly in the PHP file(s). One could be both developer and translator, and even handle both translations and i18n problems.
Generally speaking, it is important not to have conflicts between BetaWiki and the SVN translators. Indeed, no one owns a language, but some people don't like to see other people "interfering" their translation work; and if the translation quality is bad, it is even more annoying. Thus, I suggest that if a trusted person says a translation quality is bad, it would not be commited; I also suggest that if a person with SVN rights, who speaks the language, clearly opposes the current translations in BetaWiki, they will not be commited. If one doesn't want to use BetaWiki, he still "edits" the MediaWiki messages in BetaWiki, and BetaWiki translations should not override his translations without asking him about it.
Anders Wegge Jakobsen schreef:
Plain and simple NO! I'm a software developer, and I'm not going to confine myself to a web interface. I fully accept that this is where the bees knees are, so I'm not going to stand in the way of progress.
I don't understand the logic here. You're a software developer, *so* you hate web interfaces? Remember that what you're localizing is exactly that: a web interface. There's nothing you can do with direct SVN access that you can't do through BetaWiki (except for adding comments, maybe? Siebrand?), and the latter provides a wikilinked list of untranslated messages, which makes it lots easier to find that one message you accidentally skipped.
Roan Kattouw (Catrope)
Roan Kattouw wrote:
I don't understand the logic here. You're a software developer, *so* you hate web interfaces? Remember that what you're localizing is exactly that: a web interface. There's nothing you can do with direct SVN access that you can't do through BetaWiki (except for adding comments, maybe? Siebrand?), and the latter provides a wikilinked list of untranslated messages, which makes it lots easier to find that one message you accidentally skipped.
Roan Kattouw (Catrope)
It is his own choice. I, for example, think localizing through a web interface may be slow, and its committing is slower anyway than an immediate "svn commit" (though possibly faster than a patch applying - no idea about that). Commiting is the direct way for the one who can do that, and he may use the scripts "checkLanguage.php" and "rebuildLanguage.php" to simplify the translation work. When a translation approaches a certain level of completeness, the translator usually just has to update or add a few messages per time, and for this, updating the PHP file is better for the ones who have both the commit access and the PHP knowledge. No doubt BetaWiki is needed for people who don't have commit access (and don't want to submit patch) or PHP knowledge, and for people who don't want to directly edit the PHP file; but other people just don't need or want it.
Anyway, the fact someone doesn't use BetaWiki doesn't mean his concerns should be ignored: if he says a specific translation quality is bad, and is known to be a high-level speaker (and translator) of this language, it should not be commited (this is a general statement, and doesn't necessarily refer to the translation in question).
"Rotem Liss" rotemliss_net@fastmail.fm wrote in message news:477A418E.8040206@fastmail.fm...
Anyway, the fact someone doesn't use BetaWiki doesn't mean his concerns
should
be ignored: if he says a specific translation quality is bad, and is known
to be
a high-level speaker (and translator) of this language, it should not be commited (this is a general statement, and doesn't necessarily refer to
the
translation in question).
Why doesn't someone write a post-commit hook that updates the appropriate wiki page on betawiki if a translation is committed via SVN? Then people can use whichever method they like, and they would never be out of sync.
- Mark Clements (HappyDog)
Mark Clements schreef:
Why doesn't someone write a post-commit hook that updates the appropriate wiki page on betawiki if a translation is committed via SVN? Then people can use whichever method they like, and they would never be out of sync.
- Mark Clements (HappyDog)
What happens in case of a conflict? Translations entered at BetaWiki are only committed every so many hours (i.e. whenever Siebrand has time).
Roan Kattouw (Catrope)
"Roan Kattouw" roan.kattouw@home.nl wrote in message news:477BF45B.8050408@home.nl...
Mark Clements schreef:
Why doesn't someone write a post-commit hook that updates the
appropriate
wiki page on betawiki if a translation is committed via SVN? Then
people
can use whichever method they like, and they would never be out of sync.
What happens in case of a conflict? Translations entered at BetaWiki are only committed every so many hours (i.e. whenever Siebrand has time).
If changes via SVN are automatically updated onto the wiki, then any conflicts would have to be picked up by the on-wiki translator(s) and resolved using the standard diff tools. Presumably the page would be watched and a notification about the update sent to the on-wiki translator(s) so they would be made aware of the change. The resolved version would then go back into SVN when the next update is done.
To avoid too many edit conflicts, an extension that automatically commits changes in the other direction (from the wiki into SVN at point of save) would be a good solution (though it would need to avoid getting into a loop!).
In that situation, problems will only occur if there are 'edit wars' between a wiki translator and an SVN translator, but as has already been said, communication is the key here...
- Mark Clements (HappyDog)
On 1/5/08, Mark Clements gmane@kennel17.co.uk wrote:
To avoid too many edit conflicts, an extension that automatically commits changes in the other direction (from the wiki into SVN at point of save) would be a good solution (though it would need to avoid getting into a loop!).
I vote for this. Give BetaWiki its own SVN account and use it automatically. That would make it much easier to filter out the flood of localization-only commits, for those who want to (lately they seem to be dominating the software changes).
Simetrical wrote:
On 1/5/08, Mark Clements wrote:
To avoid too many edit conflicts, an extension that automatically commits changes in the other direction (from the wiki into SVN at point of save) would be a good solution (though it would need to avoid getting into a loop!).
I vote for this. Give BetaWiki its own SVN account and use it automatically. That would make it much easier to filter out the flood of localization-only commits, for those who want to (lately they seem to be dominating the software changes).
Good idea. But maybe delay a bit the commit (about 5 minutes) so if the translator changes again the message it doesn't flood the svn commits?
On 1/7/08, Platonides Platonides@gmail.com wrote:
Simetrical wrote:
I vote for this. Give BetaWiki its own SVN account and use it automatically. That would make it much easier to filter out the flood of localization-only commits, for those who want to (lately they seem to be dominating the software changes).
Good idea. But maybe delay a bit the commit (about 5 minutes) so if the translator changes again the message it doesn't flood the svn commits?
Nikerabbit and I talked about this on IRC. He thought it was a bad idea: there would be too many commits, and they should be reviewed before deployment by a human.
The i18n commits are really drowning out everything else on MediaWiki-CVS-l, though. I can just delete 30 localization e-mails a day, but if I could figure out a way to block them it would be easier. What generates the text of those e-mails? Could it examine the list of changed files and add a keyword if they only change paths matching /.i18n|messages/ or whatever?
Hoi, As it is, messages are entered by translators and with some regularity the messages are bundled and a "sanity check" is applied to the messages before they are committed to SVN. Not all messages are necessarily committed to the SVN and therefore an automatic update to SVN is not a sane thing to do. Of real importance is the communication between translators and developers on #mediawiki-i18n irc://irc.freenode.net/mediawiki-i18n, it is not only first line support that is discussed but also how to implement grammatical support for specific languages or language groups.
It is correct that a lot of localisation is happening at the moment. When you look at the statistics, there is room for a lot more localisation because there is still a lot of "red" to be seen. All these commits are welcome because this is what makes MediaWiki usable for people that do not read English.
When you look at the changes that are a consequence of the localisation, they are in two parts. There are changes to the messages themselves as a consequence of the improving implementation of localisation technology and there are the localisations themselves. When you want to have the localisations filtered, the way commits to SVN are reported has to change. It should not mean that the localisations are no longer reported because some people, myself included, are interested in this.
Thanks, GerardM
http://translatewiki.net/wiki/Translating:Group_statistics
Thanks, GerardM
On Jan 7, 2008 4:49 PM, Platonides Platonides@gmail.com wrote:
Simetrical wrote:
On 1/5/08, Mark Clements wrote:
To avoid too many edit conflicts, an extension that automatically
commits
changes in the other direction (from the wiki into SVN at point of
save)
would be a good solution (though it would need to avoid getting into a loop!).
I vote for this. Give BetaWiki its own SVN account and use it automatically. That would make it much easier to filter out the flood of localization-only commits, for those who want to (lately they seem to be dominating the software changes).
Good idea. But maybe delay a bit the commit (about 5 minutes) so if the translator changes again the message it doesn't flood the svn commits?
Wikitech-l mailing list Wikitech-l@lists.wikimedia.org http://lists.wikimedia.org/mailman/listinfo/wikitech-l
On 1/7/08, Gerard Meijssen gerard.meijssen@gmail.com wrote:
When you want to have the localisations filtered, the way commits to SVN are reported has to change. It should not mean that the localisations are no longer reported because some people, myself included, are interested in this.
Obviously not, which is why I never suggested such a thing. I did, in fact, suggest that "the way commits to SVN are reported has to change". Or I can just archive 30+ commits a day, I guess that works too for me, but I wonder if other devs are discouraged from reading the commit list with so much material that's not relevant to what they do.
Simetrical wrote:
On 1/7/08, Gerard Meijssen gerard.meijssen@gmail.com wrote:
When you want to have the localisations filtered, the way commits to SVN are reported has to change. It should not mean that the localisations are no longer reported because some people, myself included, are interested in this.
Obviously not, which is why I never suggested such a thing. I did, in fact, suggest that "the way commits to SVN are reported has to change". Or I can just archive 30+ commits a day, I guess that works too for me, but I wonder if other devs are discouraged from reading the commit list with so much material that's not relevant to what they do.
Wikitech-l mailing list Wikitech-l@lists.wikimedia.org http://lists.wikimedia.org/mailman/listinfo/wikitech-l
While I fully support any localisation efforts I indeed suffer from this problem - while not a developer I do like to keep a track of code updates in my effort to become more knowledgeable of the core code and keep up with new features, unfortunately so may commits is too much for me to go through with all the other commitments I have. A keyword would be awesome.
MinuteElectron.
Just say the word. Which particular, identifiable string would you like to be present in the commit messages that come from Betawiki and solely contains i18n updates?
How about "Updates from the Ministry of Silly Walks". That should not be seen too often in any other commit ;)
But seriously: if you filter the content on "Localisation updates" you should be in the clear. Both Grondin and I use it in out i18n-only commit messages with updates from Betawiki. If the 'direct to SVN committers' would adapt the same policy, you would also catch them in your filters.
Cheers! Siebrand
-----Oorspronkelijk bericht----- Van: wikitech-l-bounces@lists.wikimedia.org [mailto:wikitech-l-bounces@lists.wikimedia.org] Namens Simetrical Verzonden: maandag 7 januari 2008 17:26 Aan: Wikimedia developers Onderwerp: Re: [Wikitech-l] An update on localisation in MediaWiki
On 1/7/08, Gerard Meijssen gerard.meijssen@gmail.com wrote:
When you want to have the localisations filtered, the way commits to SVN are reported has to change. It should not mean that the localisations are no longer reported because some people, myself included, are interested in this.
Obviously not, which is why I never suggested such a thing. I did, in fact, suggest that "the way commits to SVN are reported has to change". Or I can just archive 30+ commits a day, I guess that works too for me, but I wonder if other devs are discouraged from reading the commit list with so much material that's not relevant to what they do.
Galician and Asturian are hardly new. Also, I'm not sure how Inuktitut or Kurdish serve usability purposes or are duplicates...
Mark
On 31/12/2007, Siebrand Mazeland s.mazeland@xs4all.nl wrote:
I have not seen a comprehensive overview of MediaWiki localisation discussed on the lists I am posting this message to, so I thought I might give it a try. All statistics are based on MediaWiki 1.12 alpha, SVN version r29106.
==Introduction== *Localisation or L10n - the process of adapting the software to be as familiar as possible to a specific locale (in scope) *Internationalisation or i18n - the process of ensuring that an application is capable of adapting to local requirements (out of scope)
MediaWiki has a user interface (UI) definition for 319 languages. Of those languages at least 17 language codes are duplicates and/or serve a purpose for usability[1]. Reporting on them, however, is not relevant. So MediaWiki in its current state supports 302 languages. To be able to generate statistics on localisation, a MessagesXx.php file should be present in languages/messages. There currently are 262 such files, of which 16 are redirects from the duplicates/usability group[2]. So MediaWiki has an active in-product localisation for 236 languages. 66 languages have an interface, but simply fall back to English.
The MediaWiki core product recognises several collections of localisable content (three of which are defined in messageTypes.inc):
- 'normal' messages that can be localised (1726)
- optional messages that can be localised, which usually only happens for languages not using a Latin script (161)
- ignored messages that should not be localised (100)
- namespace names and namespace aliases (17)
- skin names (7)
- magic words (120)
- special page names (76)
- other (directionality, date formats, separators, book store lists, link trail, and others)
Localisation of MediaWiki revolves around all of the above. Reporting is done on the normal messages only.
MediaWiki is more than just the core product. On http://www.mediawiki.org/wiki/Category:All_extensions some 750 extensions have some kind of documentation. This analysis will scope only to the code currently present in svn.wikimedia.org/svnroot/mediawiki/trunk. The source code repository contains give or take 230 extensions. Of those 230 extensions, about 140 contain messages that can be visible in the UI in some use case (debugging excluded). Out of those 140, about 10 extensions have an exotic implementation for localisation localisation support at all (just English text in the code). 10 extensions appear to be outdated. I have seen about 5 different 'standard' implementations of i18n in extensions. Since MediaWiki 1.11 there is wfLoadExtensionMessages. Not that many extensions use this yet for message handling. If you can help add more standard i18n support for extensions (an overview can be found at http://translatewiki.net/wiki/User:Siebrand/tobeadded) or help in standardising L10n for extensions, please do not hesitate.
==MediaWiki localisation in practice== Localisation of MediaWiki is currently done in the following ways I am aware of:
- in local wikis: Sysops on local wikis shape and translate messages to fit their needs. This is being done in wikis that are part of Wikimedia, Wikia, Wikitravel, corporate wikis, etc. This type of localisation has the fewest benefits for the core product and extensions because it happens completely out of the scope of svn committers. I have heard Wikia supports languages that are not supported in the svn version. I would like to get some help in identifying and contacting these communities to try and get their localisations in the core product. Together with SPQRobin, I am trying to get what has been localised in local Wikipedias into the core product and recruit users that worked on the localisation to work on a more centralised way of localisation (see Betawiki)
- through bugzilla/svn: A user of MediaWiki submits patches for core messages and/or extensions. These users are mostly part of a wiki community that is part of Wikimedia. These are usually taken care of by committers raymond, rotemliss, and sometimes others). Some users maintain a language directly on SVN. At the moment, 10-15 languages are maintained this way: Danish, German, Persian, Hebrew, Indonesian, Kazach (3 scripts), Chinese (3 variants), and some more less frequently.
- through Betawiki: Betawiki was founded in mid 2005 by Niklas Laxström. In the years to follow, Betawiki has grown to be a MediaWiki localisation community with over 200 users that has contributed to the localisation of 120 languages each month in the past few months. Users that are only familiar with MediaWiki as a tool can localise almost every aspect of MediaWiki (except for the group 'other' mentioned earlier) in a wiki interface. The work of the translators is regularly committed to svn by nikerabbit, and myself. Betawiki also offers a .po export that enables users to use more advanced translation tools to make their translation. This option was added recently and no translations in this format have been sumitted yet. Betawiki also supports translation of 122 extensions, aiming to support everything that can be supported.
==MediaWiki localisation statistics== MediaWiki localisation statistics have been around since June 2005 at http://www.mediawiki.org/wiki/Localisation_statistics%5B3]. Traditionally reports have focused on the complete set of core messages. Recently a small study was done after usage of messages, which resulted in a set of almost 500 'most often used messages in MediaWiki', based on usage of messages on the cluster of Wikimedia (http://translatewiki.net/wiki/Most_often_used_messages_in_MediaWiki).
Up to recently there were no statistics available on the localisation of extensions. Through groupStatistics.php in the extension Translate, these statistics can now be created. Aside from reporting on 'most often used MediaWiki messages', 'MediaWiki messages', and 'all extension messages supported by extension Translate' (short: extension messages). Additionally, a meta extension group of 34 extensions used in the projects of Wikimedia has been created (short: Wikimedia messages). A regularly updated table of these statistics can be found at http://translatewiki.net/wiki/Translating:Group_statistics.
Some (arbitrary) milestones have been set for the four above mentioned collections of messages. For the usability of MediaWiki in a particular language, the group 'core-mostused' is the most important. A language must qualify for MediaWiki to have minimal support for that language. Reaching the milestones for the first two groups is something the Wikimedia language committee considers to use as a requirement for new Wikimedia wikis:
- core-mostused (496 messages): 98%
- wikimedia extensions (354 messages): 90%
- core (1726 messages): 90%
- extensions (1785 messages): 65%
Currently the following numbers of languages have passed the above milestones:
- core-mostused: 47 (15,5% of supported languages)
- wikimedia extensions: 10 (3,3% of supported languages)
- core: 49 (16,2% of supported languages)
- extensions: 7 (2,3% of supported languages)
==Conclusion== So... Are we doing well on localisation or do we suck? My personal opinion is that we do something in between. Observing that there are some 250 Wikipedias that all use the Wikimedia Commons media repository, and that only 47 languages have a minimal localisation, we could do better. With Single User Login around the corner (isn't it), we must do better. On the other hand, new language projects within Wikimedia all have excellent localisation of the core product. These languages include Asturian, Bikol Central, Lower Sorbian, Extremaduran, and Galician. But where is Hindi, for example, with currently only 7% of core messages translated?
With the Wikimedia Foundation aiming to put MediaWiki to good use in developing countries and products like NGO-in-a-box that include MediaWiki, the potential of MediaWiki as a tool in creating and preserving knowledge in the languages of the world is huge. We have to tap into that potential and *you* (yes, I am glad you read this far and are now reading my appeal) can help. If you know people that are proficient in a language and like contributing to localisation, please point them in the right direction. If you know of organisations that can help localising MediaWiki: please approach them and ask them to help.
We have all the tools now to successfully localise MediaWiki into any of the 7000 or so languages that have been classified in ISO 639-3. We only need one person per language to make it happen. Reaching the first two milestones (core-mostused and wikimedia extensions) takes about 16 hours of work. Using Betawiki or the .po, little to no technical knowledge is required.
This was the pitch. How about we aim to at least double the numbers by the end of 2008 to:
- core-mostused: 120
- wikimedia extensions: 50
- core: 90
- extensions: 20
I would like to wish everyone involved in any aspect of MediaWiki a wonderful 2008.
Cheers!
Siebrand Mazeland
[1] als,crh,iu,kk,kk-cn,kk-kz,kk-tr,ku,sr,sr-jc,sr-jl,zh,zh-cn,zh-sg,zh-hk,zh-min-nan,zh-yue [2] crh,iu,kk,kk-cn,kk-kz,kk-tr,ku,sr,sr-jc,sr-jl,zh,zh-cn,zh-sg,zh-hk,zh-min-nan,zh-yue [3] older locations are http://www.mediawiki.org/wiki/Localisation_statistics/stats and http://meta.wikimedia.org/wiki/Localization_statistics
Wikitech-l mailing list Wikitech-l@lists.wikimedia.org http://lists.wikimedia.org/mailman/listinfo/wikitech-l
Hi Mark,
You are probably missing some background on localisation within MediaWiki.
Iu, which is the code that is being used for Inuktitut in MediaWiki should in my opinion not have been a Wikipedia in a language. Inuktitut is a macro language that consists of 3 classified languages[1]:
* Inuktitut, Eastern Canadian [ike] (Canada) * Inuktitut, Greenlandic [kal] (Greenland) * Inuktitut, Western Canadian [ikt] (Canada)
Of the above three, Eastern Canadian Inuktitut appears to have the most speakers, so the UI 'iu' falls back to 'ike-cans', which is Eastern Canadian Inuktitut in Unified Canadian Aboriginal Syllabics. Eastern Canadian Inuktitut is also written in Latin script (ike-latn). So far we have had no luck in getting any active translators for the above languages, so that we only have what was in iu.wikipedia.org. If you know any, please point them in our direction. It would be very much appreciated.
As for Kurdish, that is currently supported in both Latin (ku-latn) and Arabic (ku-arab) script. 'ku' falls back to 'ku-latn'.
I hope the above makes clear to you that reporting on languages that are only being used as a redirect is useless and pollutes statistics. Hence they have been taken out.
As for Galician and Asturian: full blown localisation of those languages in the MediaWiki core product is definately very recent. I could also have named a few other languages.
Hope this makes things more clear to you.
Cheers!
Siebrand
P.s. I have today imported the MediaWiki messages from na.wikipedia.org to Betawiki. Your named came up as one of the editors to the mediawiki namespace in that wiki. Can you help...?
[1] http://www.ethnologue.com/show_family.asp?subid=91178
-----Oorspronkelijk bericht----- Van: wikitech-l-bounces@lists.wikimedia.org [mailto:wikitech-l-bounces@lists.wikimedia.org] Namens Mark Williamson Verzonden: donderdag 3 januari 2008 19:44 Aan: Wikimedia developers Onderwerp: Re: [Wikitech-l] An update on localisation in MediaWiki
Galician and Asturian are hardly new. Also, I'm not sure how Inuktitut or Kurdish serve usability purposes or are duplicates...
Mark
wikitech-l@lists.wikimedia.org