Task force/Local language projects/Summary of QNA
Note that the last time this document was updated with new information was the 22nd of November, it is a good place to start to quick dive into the discussions that has been done. But for further discussions, please visit the talk page for the local language project Task Force.
I will here try to summarize the answers given to the QNA questions so far, for more information consult the local language, questions that need answers, section and the Local Language Projects talk page.
- 1 Fostering the open content movement
- 2 Relevant material for given region
- 3 Translation
- 4 Overall strategies and localization
- 5 Large Wikipedias supporting small
- 6 Importance of a strong local community
- 7 Cooperation at a meta
- 8 Providing technical support
- 9 Local Arabic chapter
- 10 Quality indication
Fostering the open content movement
The open content movement is young even in countries where Wikipedias has succeeded, even more so in less developed countries. This together with the "expert knowledge cannot be produced by non-experts" view hinders development of local language Wikipedias. Fostering a positive view toward this would help. The open content movement seems to agree with the philosophy of Islam (and probably most other religions too) which could help such fostering.
Relevant material for given region
Writing articles the users want to read, and developing a healthy community is important in increasing the use and participation of local language Wikipedias. Well written articles about what local people want to read would attract local users. By consulting the traffic logs and asking Google (and others) what people in a certain region want to read about can help in getting to know which articles are of importance. Establishing local Chapters is only useful when there is a large enough contributor community.
Translation of wiki content could help kick-start a Wikipedia. Concentrating such translation on educational/technological/computer related articles could help attract native users that are most likely to themself contribute. Word-by-word translation is however a painful process and concentrating on "retelling" the article content might be a more effective way of translation. There is a translation project for Arabic Wikipedia that uses googles translation toolkit to simplify translation, but almost every sentence has to be corrected. Translation of articles made by bots or translators that ain't from the area of the local language themself are prone to concentrate translation on material that doesn't agree with interest of the local people, which makes the Wikipedia grow, but in an area that will be of no use. At http://www.translatewiki.net translation is concentrated to articles that are seen and thought to be of importance for local users.
Translation could be simplified by automating some steps of the translation process. Add a translation option where by choosing language you are taken directly from the article into the edit page of a new language. The original article source should be copied into the Edit box of the new language article with links automatically replaced with links to the corresponing articles of the new language Wikipedia. So that translators only have to concentrate on actuall translation. Not looking up the right links, knowing the markup language and so on. Some care has to be taken about how to handle cases where the links link to articles that doesn't exsit in the new language though, which is very probable to happen for small language Wikipedias.
What languages might be most important to translate is hard to tell. There has been proposals that the official languages of the United Nations (Arabic, Chinese, English, French, Russian, and Spanish) are important languages. A wider language scope of around 40 languages could include lingua francas because they in turn could be a base for further translation. How do we judge which languages are most probalbe to make most impact if translated into?
Overall strategies and localization
Strategies that help small Wikipedias to grow might be of larger importance than the actual translation. Improving the overall coverage of subjects might be one such strategy. Even if the English Wikipedia were fully translated into an African language it might not contain much information of value for the local African community. Covering more subjects of local interest, even if the articles are written in the language of a large Wikipedia, can stimulate the growth of smaller Wikipedias. Another problem is the lack of localization of Wikimedia tools for many languages. Such localization would improve the ability of local contributors to contribute.
Large Wikipedias supporting small
Successful Wikipedias could help smaller Wikipedias by providing templates, technical information, coordination of local language projects. Some positive experience from such cooperation has been found in the Russian language project ru:Википедия:Проект:Малые разделы Википедии на языках России. However, a problem that can arise when large language Wikipedia editors try to help small language Wikipedias is that moral and value standards from the former is imposed on the later which might hinder developement. One such value can be quality, which might be among the most important for a large language Wikipedia. To much emphasis on it on a small Wikipedia can hinder its developement because the most important thing there is to get past the infliction point where the project doesn't depend on individual editors. It is important that the large language Wikipedia contributors realizes the importance of the smaller language Wikipedias for cooperation to be ansuccessful. Because when it is consider that the newbie experience is nasty for the "mature" Wikipedias and that the growth of editors is the key to the health of smaller projects, the help from larger Wikipedias could even be counter productive if these problems ain't realized and dealt with in an approperiate way.
Importance of a strong local community
The strength of Wikimedia projects is their independence. When a strong local community is formed, the local wiki will develop itself. Therefore, it is first important to help the development of local communities and chapters. A local chapter will know best how to promote the local project in the local society/culture. A local project shouldn't have to invent the wheel again, so valuable advice about general project and community development should be available when asked for. This can be done by monitoring their development offering help when problems are encountered in an active (but not authoritarian) way. A short yearly report (at meta) could raise awareness of the problems and advantages of a certain project, and after every report experienced users from other projects could have the opportunity to post advice and comments.
Cooperation at a meta
Contributors with specialist knowledge might be hindered from contributing to large language Wikipedias because the large language ain't his first language and he feels to insecure to use it. The contribution will then only benefit the small language Wikipedia and will remain in isolation because of the language barrier. Cooperation on global wikiprojects over the language borders, on a meta where experts in one subject but with many languages can join, could help pull such contributors out of isolation. Also multiple versions of the same topic in different languages are now created by each project inventing their own wheel. To bring people from different projects together can be done by giving meta a larger role in sharing experience and knowledge about community and project building. The page at meta can be the place where users from small projects can raise questions about content that can be discussed/answered by the experts of larger projects. Meta could host wiki-embassies and wikiprojects shared by multiple projects, corners where users from different projects can share information. Most contributors to meta are at the moment admins and wiki-politicians. The user type that adds content should be encouraged to join too. If a wikiproject exists at multiple projects (example: en|it|fr|pl|sv|pt|de|etc) an umbrella project page could be made at meta, where all discussions that don't exclusively have local importance should go. Meta-wikiprojects will also encourage participation in meta by the type of user that only adds content and isn't interested in wikipolitics. What here is meant by a wiki-embassy is not the thing found at many smaller projects, but a project page at meta where a local wiki presents itself (differences with other projects, what the community finds important, how they work, etc). Many wikis have different guidelines/rules, which can be confusing for a new user accustomed with another project. There should be an 'ambassador' for every project too, one (or more) user(s) to man the embassy. Their task would be to notice the local community when there are developments at meta and vice versa, and to write the yearly report.
The usefullness of coordination of such projects at a meta is however questioned, and what kind of material that would be intresting to present there is unclear. For projects undertaken by a single or a few editors the extra effort put into writting repports hardly will seem worthful to them, especially because the information has to be presented in another language than their own to be useful for others. And all projects not in the top 50 are likely to be projects where individuals may have a strong influence on a project. Anyway, some examples of data that could be of use is statistics (content/users/pages/vandalism) with some comments, local decisions/solutions about community problems.
Providing technical support
Experienced Wikipedia editors could provide technical support to translators with knowledge in a small language that are willing to translate articles but finds it to difficult for technical reasons. Translation groups consisting of experienced Wikipedia editors, people with knowledge in a small language but from the same country as the Wikipedia editor and native speakers could provide a good ground for translation of articles. Such a group could for example consist of a Wikipedia editor, people he knows that has done voluntary work in a developing country and contacts the volunteers has in the actuall country. Such a collaboration could eventually lead to actuall knowledge production from within the area of the local language. Maybee organisations such as Engineers Without Borders (http://www.ewb-international.org/) would be willing to arange such groups as they are technology oriented and easily can learn to edit a Wikipedia, the aim of translation agrees with their knowledge spreading philosophy and they might have good contacts in developing countries.
Local Arabic chapter
A local chapter in MENA, most likely in Egypt because it has most Arabic contributors could help develop Arabic Wikipedia. Such a chapter could help organizing events and cooperation with universities as well as improving the press coverage of Wikipedia.
One problem that arises when articles from large Wikipedias are translated into smaller language Wikipedias is that errors get copied. In large language Wikipedias the errors are quickly corrected but on the smaller ones it stays for a long time, because there are less people able to correct it. More effort into indicating the quality of large language articles could help translators judge the value of the article they translate.