User:Thuvack/links/Bantu Community Fellow
Comments on this draft are welcome on the talk page
South Africa has 11 official languages. Of these, 9 are South Bantu languages. Most of these languages are represented in Wikipedia. Unfotunately these wikipedias has almost no native speaking contributors. These wikipedias have not grown in the past 3 years following trends of other small wikipedias. The following table is a summary of the South African Bantoid Wikipedias:
| Language | Views/month | # Articles |
|---|---|---|
| Zulu | 105139 | 496 |
| Xhosa | 44766 | 143 |
| Sesotho | 56169 | 155 |
| Northern Sotho | 54292 | 610 |
| Tswana | 49935 | 498 |
| Tsonga | 66804 | 212 |
| Venda | 43,134 | 207 |
| Swati | 92712 | 372 |
From the above table it is clear that there is a missing language from the Sout African equation. The language is Ndebele, this is the missing language wikipedia to complete the Bantoid wikipeia picture.
Bantoid languages have what i would call a fledging or non-existant community of editors. What are barriers to attracting editors? what are the barriers to retaining the few that come? Why are female contributors desirable? What should be the main goal for bantoid wikis, translation or unique content creation? What about language localization? what have I learned from Xitsonga wikipedia?
Background[edit]
Bantoid language wikis, have not grown as expected. This is a proporsal to Wikimedia Foundation Fellowship program. The report details percieve and real barriers for new users. The project goal of Bantoid language wiki projects is discussed. The role and context of language localization is explored. The concerpt of Content creation through collaboration and cross wiki-editting is suggested and the role of wiki meet-ups, delved. Fellowship project targets and deliverables are outlined. Project logistics and items that will be necessary for the successful completion of the followship, are detailed.
Why the "Start and they will come" model is not working for Bantoid Wikis[edit]
The "Start and they will come" concerpt in building African Wikis (I will concentrate on Bantu languages of the SADEC region of Africa), has not worked well over the years. If success is to be realied in this part of the world, editor recruitment and retention methods can nolonger be Arbitrary, they must be deliberate and systematic. I proporse that a "catalyst" method need to be used o set up a community of editors for these language wikis to grow as a movement and culture. The role of the catalysts must be well defined and a clear path must be charted.
Barriers to community growth[edit]
There are "real" and "percieved" barriers to new and old editors of small wikis. These range from cultural backgrounds to socio-economic issue which need to be clearly identified and a solutioin for each, formulated. Below I list some of these:
- Account creation and user interphase (Perceived): Wiki environment is too strange.
- Internet accessibility (Real): Users often access internet at their work place.
- Cultural barriers and steriotypes (Real): Users often feel inadequate to contribute original work and research, women are seldom expected to participate.
- Lack of/Unclear Project Goals (Perceived): new users are often unsure what the project is trying to achieve and hence not sure what to contribute.
- Lack of collaborator network (Real): new users prefer to be guided through projects. Where there are no collaborators and support network, users tend to be discouraged to continue editing.
A network of experienced native users must be established in order to help solve and over come some of the above barrier for new users.
Main goal for bantoid wikis: translation or unique content creation?[edit]
There must be clear project goals that a small wiki sets out to achieve. This must be articulated clearly and concisely so that a new user can be persuated to evaluate if it is worth their while to contribute in the small wiki. Below are some of the projects that a community of a small wiki may be interested in persuing:
- Native botanical names: Bantoid languages have very rich content when it comes to botanical names, often linked to medicinal nature of a specific plant.
- Native Second names: Bantoid languages have unique family names and an interesting link between these expressed through family praises. Often it is a vehicle for preservation of family history too.
- Native animal kingdom names: Bantoid languages have rich content when it comes to naming animals. This aspect is very important in story telling.
The above does not have to be just a simple writting of a name. The background information usually passed from oral cultural practice needs to be documented for future generations. This can give rise to articles unique to the language. Keeping things simple is one way of reducing the percieved daunting task of contributing to a small wiki.
Language localization[edit]
The user interface for language localization has now been improved to an intuitive and easy to use platform. New users should be encouraged and guided in language localization and the review method of localized content. A new user will often have no idea of language localization, thus recruitment for this should not be happening in talk pages and village pumps, but rather it should be in main pages and clear enough for new users to understand and signup to.
Content creation through collaboration and cross wiki-editting[edit]
Bantoid wikis will most often than not include users who are profficient in more than one language or include languages that are almost dialects or derivatives of each other. These users need to be harnessed and encouraged to contribute across wikis. Users can research and write articles in their native languages, then translate it to the languages that they are profficient with.
The quality of translation and content can always be improved with the influx of new users with affinity to these articles. It then becomes important for user information boxes (Babel) to be implemented at user pages.
User Recruitment Mechanics[edit]
As described above, the recruitment process for South Bantu Wikis needs a direct approach. Here are the mechanics to be tested/applied.
Who: Head-Hunting[edit]
People to be recruited with a survival/retention rate that is sustainable are in my experience:
- Previous Editors (Who may have left for whatever reasons)
- Students in Universities
- Teachers of these Languages (Preferably Secondary school teachers), who can also include their students.
- People in Work places with access to internet e.g Librarians. (This will be harder to do).
How: Recruitment drives[edit]
Suggested recruitment method for each of the above identified groups.
- Track down previous users and compile a database of contact details or FB pages. Even create a FB group page?
- Suggest a weekend seminar where wikipedia edting can be demonstrated for students.
- Select a Pilot school and give live demonstrations to teachers on editing wikipedia.
- Include in a GLAM initiative?
The idea is to try and create small editing circles which can be monitored and supported from time to time.
Wiki-Imbizo (meet-ups) and editor retention[edit]
Wiki - Imbizo (Meet ups) should be encouraged and held often in order to build and maintain editting circles and support structures. Wikimedia-ZA will become very essential in organizing and encouraging these meet-ups.
Project Logistics[edit]
This community fellowship project requires a lot of planning and co-ordination to succeed due to my work and family commitments. I proporse the following project terms and parameters.
Targets and Deliverables[edit]
- Increase core Active permanent (language catalysts) users to 5 per Bantoid language wiki, 2 of which should be female (Total of 40 users).
- Aim for 1000 articles of acceptable quality per South Bantu wiki.
- Atleast 2 Cross editors mongst bantoid language per wiki. (Minimum of 16).
- Increase new users with more than 50 edits per month to 50 per South Bantu language wiki.
Possible collaborators[edit]
- Wiki Africa
Time frame and project plan[edit]
- 1 year fellowship.
Non-Financial requirements[edit]
- Access to Wikimedia shop merchandise during the fellowship period.
- Access to Wikimedia Foundation mentor.
Countries to be served[edit]
South Africa; Lesotho; Swaziland; Mozambique; Zimbabwe; Botswana
My Contact details[edit]
| Full Names: | Dumisani Ndubane |
| User Name: | Thuvack |
| e-mail: | thuvack@gmail.com |
| Country: | South Africa |
| Affiliation: | Wikimedia-ZA |