Steve, Thanks for the note. Replies below...
----- Original Message -----
From: "Steven G. Huter" <[log in to unmask]>
To: "Don Osborn" <[log in to unmask]>
Cc: <[log in to unmask]>
Sent: Friday, November 14, 2003 1:08 AM
Subject: Re: IUC25 & African ICT - opportunity in DC
> > The 25th Internationalization and Unicode Conference (IUC25) that will
> > held in Washington, DC in March 2004 may be of interest to members of
> > list who are not already aware of it. Its theme of "Unicode in
> > Building a Multilingual Infrastructure" would seem to harmonize with the
> > emphasis in the recent African WSIS preparatory meeting in Maputo on
> > bringing "e-strategies" for mobilizing ICT for development into the
> > mainstream of government planning.
> Are all African scripts and characters in the Unicode set ?
Re African scripts, Ethiopic/Ge'ez is there, and Tifinagh and N'ko are in
the "pipeline." The Tifinagh proposal unfortunately languished for about 5
years but is getting urgent attention since Morocco decided to use this
script in Berber language instruction. Vai and Bamum and lesser known
scripts in Africa (and for that matter the world) are the focus of the
Script Encoding Initiative at U.C. Berkeley.
Re characters in Latin-based scripts, I would say yes, so long as one
doesn't hear about this or that character not being there. It's a dynamic
process, but the important thing is that all characters for all major
Latin-based orthographies are there, and anything not there can be proposed.
One of the reasons I set up the "a12n-collaboration" working group was as a
way to discuss such issues.
That said, there was concern raised in a workshop on African languages and
the internet at the Bamako 2002 prepcon for WSIS about the lack of
"precomposed" diacritic characters to represent combinations used in a
number of languages. This discussion did not apparently get into the
question of dynamic composition which despite imperfections is significantly
improved, arguably obviating the need for addition of lots of precomposed
combinations. And in any event, as far as I can tell, the workshops
recommendation re proposal to the Unicode consortium re adding precomposeds
has not been acted on. Basically, this may turn out to be a non-issue and as
such does not hinder one from saying that all extended Latin characters used
in African languages are in Unicode.
Re characters in Arabic-based scripts or Ajami, the repertoire is quite
complete, but I think a more thorough study may be necessary. The idea of
using dynamic composition here too has been brought up, especially since
Ajami usage is not really standardized and local usage may have variations.
But here again, Unicode as is works well for Ajami.
> If the characters aren't there, then Unicode-based internationalization
> is hopeless. . . .
You are right, but it's not an all or nothing issue. Unicode is definitely
complete enough to work for African languages written with Latin-based,
Ajami, and Ethiopic/Ge'ez scripts. (The lack of Tifinagh is problematic in
the short term but will be remedied.) But it makes no sense to wait until
Unicode it is proven perfect (as if any project of such scale having to do
with language could be flawless) to recognize the hope that it does offer
and encourage people connected with African ICT to work with it.
> . . . And internationalization not based on Unicode would, at
> point, be heading down the wrong path.
At this point I think it is important to encourage more African involvement
in use of Unicode to enhance its options/potential for participation in
internationalization (and localization), and also to identify where in
Unicode there may be needs for improvement for its needs. A beginning point
is "awareness-raising" of policymakers and training of ICT experts. I've
been communicating informally with some people about training materials on
Unicode (that could be used for "add-on" components to other computer,
internet, and network trainings) and some forms of workshops or "roadshow"
on Unicode and African ICT - is anyone else working in this area?
Don Osborn, Ph.D. [log in to unmask]
*Bisharat! A language, technology & development initiative
*Bisharat! Initiative langues - technologie - développement