405 Elements of a strategy for African language web content
Sep 14, 2003
[cross-posted from a12n-forum]
One of the purposes that this group ("a12n-forum")* and its French language counterpart ("a12n-entraide") can serve is facilitating dialogue about African languages and ICT in general, including discussion of strategies and trends. Below is a posting in the latter category - a translation (modified) of a posting to a12n-entraide on the subject of how there can be more web content in African languages.** In brief, it proposes that this task requires thinking beyond the main approaches to developing content, making use of more innovative approaches.
First of all, creation of web content as we usually see it (text-based, sometimes sophisticated graphics) requires much in time and resources. One cannot create significant amounts of content very quickly, especially in the case of African languages, as there is a lack of available human and monetary resources to apply to the task. In any event, the dynamic here is by its nature slow: the proportions of the languages on the Internet change, but gradually. Some expect that these proportions will continue to approach an approximation of the current percentages of speakers of languages in the world. But that assumes relatively comparable written traditions and available resources - conditions that don't hold for much of Africa.
For the case of African language text content on the Internet, therefore, it would seem to be especially important at this point to assure that ground is not "lost" in this process, and that can be done by:
(a) Supporting existing efforts.
(b) Improving coordination and collaboration between the actors (a and b require communication - one of the reasons for the formation of this list, A12n-forum)
(c) Adopting and adapting unicode.
(d) Seeking to put texts already published (e.g. literature in African languages) on the web.
The last point is worth clarifying: since some years ago there have been "classics" published in African languages (generally with parallel text in a European language) and monographs with African language content for limited audiences. Why not put all of them on the web? In short order there would be content of an uncontested quality on the web, and this literature would be made accessible to a new generation of readers. And this would represent a powerful African presence on the Internet to encourage the others.
All that said, however, the most significant progress would be achieved using more innovative approaches such as translation by computer and experimentation with content including audio files and/or interactive images.
The technology of computer translation, although it is still rough, is advancing quickly. It indeed represents a nice tactic for exponentially increasing content accessible by the speakers of languages less represented on the Web - for example of news and works of general reference. This technology should be part of any strategy to increase the quantity of materials available on WWW in African languages. The object would be the creation of software for translation of the principal languages (English, French, etc.) with other languages, as well as of the software for translation among the dialects of the African languages. In any event, the fact of being able to read any text on the Internet in the mother tongue (even if the translation is not perfect) will have immense socio-linguistic implications which one cannot appreciate now (frankly, it surprises me that it is not discussed more).
Beyond text-based content it seems that the improvements in audio technology can allow a vast quantity of spoken content to be quickly added to or communicated on the Internet. Audio files will never replace text on the Internet, but in my opinion, they make it possible to create the new forms of content with various combinations of text, image and audio (based, if so desired, mainly on the audio). In addition, what are the possibilities to link local radio into web content? Text to speech and speech to text technologies could also be important. Could ICT thus foster development that is both "neo-oral" and written?
In addition, there must be ways to better utilize images to communicate with various audiences. This may be done most effectively with audio and supported by text. A few years ago someone in the Mauritius experimented with a computer interface using images and audio files in local languages to communicate with extension agents and farmers in rural areas on crop diseases - even an illiterate person could navigate the presentation by listening to the audio and clicking on the appropriate images. Such presentations could be multiplied in templates for adaptation to the local situations and languages. Already one organization (CAB, if I remember correctly) recently produced a CD-ROM of this kind with information on tropical agriculture. These are but initial steps - it is possible to use this approach to significantly multiply web-based content that can be targeted for many varied audiences in Africa.
One final point: In my opinion, web content authors / webmasters should not hobble themselves by thinking that one is obliged to create a whole site in one language or another, or that one must duplicate all existing content in another language or do nothing in it at all. In the city, village, market, or office almost anywhere in Africa one will hear several languages - but nobody repeats every word in each one of these languages (except in the infrequent situations where a translation is required). The case of web content is a little different, but not too much - it is possible to reflect the multilingual social realities in some aspects of web content with diverse language content on the same site, or even the same page.
These are a few possible lines of a strategy to increase web content in African languages. Comments are most welcome.
* This message will be cross-posted to "african_web_content_owner" - http://groups.yahoo.com/group/africa_web_content_owner/ . The "a12n- forum" webhome is http://lists.kabissa.org/mailman/listinfo/a12n-forum .
** This message began as a brief exploration of African language web content issues in a letter written in January 2000 to Dr. Pascal Baba Couloubaly (one of the principal figures in the Bamako 2000 conference and later Minister of Culture in Mali). This was adapted and expanded for a12n-entraide as part of a discussion there and is again modified here. The topic of local web content in Africa is treated more systematically and broadly (including but not focused specifically on languages) in Peter Ballantyne's excellent study: "Collecting and Propagating Local Development Content: Synthesis and Conclusions." Research Report No. 7. IICD in association with the Tanzania Commission for Science and Technology; funded by DFID. http://www.iconnect-online.org/base/ic_show_news?sc=107&id=1878