[A12n-forum] Re: Audio (Re: Elements of a strategy for African language web content)

["Don Osborn" <dzo@xxxxxxxxxxxx>]

Andrew, Here's a response to the rest of your letter (apart the mention of
translators, already treated).

First a general comment to the group:  in discussing ICT and African
languages here, it is hoped that A12n-forum can be a meeting place of
questions from people implementing projects and businesses, practical
experience in the field, ideas from all over, news of relevant technologies.
In this way the list will have a different enphasis from the older
sister-list, A12n-collaboration, which continues to focus more narrowly on
the more technical issues surrounding African languages in text on computers
& the internet (fonts, input methods).

----- Original Message ----- 
From: "Andrew Cunningham" <andrewc@xxxxxxxxxxxxx>
To: <dzo@xxxxxxxxxxxx>
Cc: <a12n-forum@xxxxxxxxxxxx>
Sent: Wednesday, September 17, 2003 10:16 AM
Subject: Re: [A12n-forum] Audio (Re: Elements of a strategy for African
language web content)


> A few thoughts on facilitating african language web content:
>
> The quality of audio material is very important as indicated in Don's
> email. The email provides a brief but well expressed summary of the
> issues of recording content on a small scale.
>
> Another option is the digitization of African language radio programs,
> although the move to digital recording studios that are required for
> high capacity audio streaming would be expensive, but a valuable
> investment in the future.

I hope we can get some discussion (and if necessary reposted info that is
pertinent) on this dimension of ICT in Africa.  A lot of time and effort has
been and continues to be (necessarily) focused on the text dimension, but it
seems that various audio (and video as Pam just brought up) possibilities
could be further explored.

[ . . . ]

> Although from a more practical point of view, there are a few suggestions:
>
> 1) there needs to be wider knowledge of unicode and web
> internationalization techniques.

Very true.  To this end I have been communicating with Patrick Andies and
some others re strategies for that in Africa.  More soon as this discussion
enlarges.

> 2) tools, including keyboard layouts, fonts and text and html editors
> suitable for African languages.

Also necesary.  Much of this is the focus of A12n-collaboration

> 3) web services need to be made available that are suitable for african
> languages and easily localisable into African languages. This may
> include scripts for emailing contents of html forms, web based email
> services, discussion forums, blogs, wikis, etc. Localise electronic
> community tools.

Some of these you've been pioneering in, and we've also discussed with Steve
Yost of QuickTopic.  Definitely essential parts of the equation.

> and develop localised search engine interfaces and langauge specific web
> directories. Resource location tools are important.

Again good points.

> With respect to content development:
>
> 1) character set conversion tools to convert documents in legacy
> encodings into unicode. We're doing this with some older non-unicode
> Nuer word documents. I've created some conversion tables that we use
> with TECKit to convert between the encodings.

I think a fair amount of this has been going on.  The only other effort I
can think of offhand, though, is that of RIFAL which did some work in Niger
and several other countries.  It might be helpful to compile a list of
legacy to Unicode conversions by country.

This however would bring us back to an extremely useful idea we've discussed
before - country pages on languages, character requirments, locale issues,
etc.  Add to that these fonts issues.  Very time-consuming I think.

> 2) OCR software development for OCR in African languages, allowing
> printed documents to be scanned and converted to electronic documents.

Yes.  It was a minor headache going through the scanned version of the
"Practical Orthography of African Languages" (now on the Bisharat site) to
insert all the extended characters that the software didn't know how to
interpret.  It would be a bigger headache for longer texts.

It starts to look like an agenda, as well it should.  I am planning on
airing for discussion another agenda - the one that came out of the Bamako
2002 WSIS prepcon workshop on African languages and the internet. Perhaps
out of this message and that consideration a broader set of agenda items
could be generated.

Don

Don Osborn         dzo@xxxxxxxxxxxx
*Bisharat! A language, technology & development initiative
*Bisharat! Initiative langues - technologie - développement
http://www.bisharat.net