Facilitating African Language Wikipedias

From Wikimania

In the runup to the talks by Martin Benjamin and Kasper Souren on African languages and Wikipedia today (6 August) I communicated with them and several others (Ndesanjo Macha, Paa Kewsi Imbeah, John Hutchison) about a possible framework to encourage and help provide an overview (not "oversee"[!], but a perspective of larger parts of the whole) of the development of Wikipedias in African languages. I used the tern "task force" which, although it has a positive use and meaning in the realm of organizing efforts, might seem too hierarchical or centralizing in purpose and ambition.

The term and the concept are of course open to discussion, and indeed need discussion. And for the sake of the discussion I will post here what I see as the role of such a working group / "task force" / association / assemblage. Hope it's okay that I "shout out" some of what folks are doing in the notes.A12n

Informal proposal

My thought is that some sort of informal task force could communicate about African language Wikipedias among ourselves and on Wikipedia itself. Purposes of such a group could be:

  1. Promoting of the concept of developing Wikipedias *in* African languages via individual contacts, talks, blogs, other dissemination.
  2. Supporting the development and refinement of existing Wikipedias in African languages. This would require some individual contacts and publicizing. Since the number of these is defined and finite, we could split them up for our own contact or to find others to work on. It would be a shame to have Wikipedias closed for the languages there. Most of them are significant languages./1
  3. Supporting the formation of groups to initiate new African language Wikipedias. Here too individual contacts and networking could aid the process. There are people who could contribute and gradually more of them have access.
  4. Discussing frameworks to organize Wikipedias in closely related African languages, to minimize duplication of effort, leverage skills and content to widest benefit, harmonize presentations for the benefit of users./2
  5. Developing strategies in communication & collaboration with organizations like ACALAN, national applied linguistics agencies, diverse organizations promoting education and/or development in Africa, ICT4D projects, and African language teachers associations in Africa and abroad./3
  6. Discussing Africa-specific approaches that may be new to Wikipedia(?). These may include consideration of, for instance,
    1. Socioeconomic issues: How in societies with low average income and other pressing concerns to facilitate individuals contributing to Wikipedia in African languages? /4
    2. Educational dimension: Can advanced language classes be involved in content development?/5
    3. Technical factors: Ways of converting among dialects/orthographies/scripts where needed and possible./6
  7. Bringing in experience from other online efforts like Kamusi into the discussions.

Notes (providing more detail and examples):

(1) The representation for some of the major languages of the continent is embarassing. (Swahili is by far the best represented apart from Arabic and Afrikaans.) But many of the others on the current list of Wikipedias are still significant languages. Take Ewe for instance (I used to live in southern Togo): Ewe has about as many speakers as Lithuanian (~3 million, a factoid I looked up when noticing that conferences in Lithuania have Lithuanian as one of the official languages, while a conference *about the Ewe* and neighboring peoples evidently was English & French only), but virtually no Wikipedia presence while Lithuanian has over 20,000 articles with the number growing quickly. There are a lot of reasons for this - income, politics, border, written/oral traditions, and different colonial/occupation histories - but not 4 orders of magnitude worth! Ewe has been written for at least a century. Anyway there are other examples.

(2) This is a sticky issue in some cases. I don't pretend that any of us or an informal group of us and others can solve the issues, which are often sort of specific for particular situations. However it is helpful to keep an eye out and help the discussions as they develop.

(3) This is neither to become a bureaucracy nor turn the Afrophone Wikipedia into one. But rather to recognize that in principle, and hopefully in practice, such organizations can help the movement in one way or another. Conversely, it does not pay to forget them.

(4) One example comes from separate correspondence with Paa Kwesi and Kasper. Paa Kwesi was recently looking at how young people need income and could be paid by Ghanaian expatriates to translate/develop content in Akan specific to the culture and history. As it turns out, Kasper experimented with systems to compensate writers/translators of content in Bambara. Such ideas and experience could be collated and disseminated for for further experiments. I do think that finding strategies to compensate Africans to develop and enhance Wikipedia content could be very positive for a lot of reasons, but it probably needs to be thought through.

(5) There is increasing interest generally in how to best use Wikipedia for academic classes. This includes solutions that involve the classes, and student work, in improving Wikipedia entries. In the area of African language classes in and outside of Africa, this could be a way of increasing quality content in many languages. Ndesanjo has suggested ways to work with Swahili classes to increase the number of Swahili Wikipedia entries. Similarly Kasper mentioned his bringing this up with classes teaching Bambara language. Note that while African languages are not widely taught outside of Africa, there is still a significant and growing interest in at least a number of these languages, and the students who are getting to an advanced level have access to a lot of information as well as the technology - under guidance of instructors and as part of their language learning, they could contribute significantly especially at this time.

(6) For dialects it may be tricky but 10 years ago there were already "computer assisted dialect adjustment" (CADA) in use and I'm sure that with the advances in machine-translation technology this sort of thing is much better now. Toggling between orthographies shouldn't be too much problem - it was over 10 years ago that I first saw such for Chinese simplified/traditional in pre-Windows word processors. For changing between scripts (Latin/N'ko, Latin/Ajami(Arabic), Ethiopic/Latin [for Oromo?], Tifinagh/Arabic/Latin for Berber languages) that's another matter, but we can ride that wave when it comes.

"AfrophoneWikis" Yahoogroup

Martin Benjamin and I set up a new list on 7 August 2006 as a way to facilitate and expand dialogue on this topic. It's at http://groups.yahoo.com/group/afrophonewikis/

This may seem an unusual decision - but we can carry on discussion and work across media with the focus on developing African language Wikipedias. Yahoogroups is a medium that I personally have the impression of being familiar to many African internet users (seen this in English and French Yahoospaces). In any event it, and e-mail lists generally, are certainly much more familiar at this time to the average African internet user than Wikipedia. So we leverage that.

There will be crosslinks of course. A12n