[Indic-computing-users] [Fwd: Re: [GKD] The $100 Computer: A Polite Scam]

[Tapan P. <ta...@cs...>]

-------- Original Message --------
Subject: Re: [GKD] The $100 Computer: A Polite Scam
Date: Tue, 1 Mar 2005 17:43:15 +0000
From: Pat Hall <p.a...@bt...>
Reply-To: gk...@ph...
To: gk...@ph...

Dear Colleagues,

We have had many thoughtful and thought-provoking postings about the
hardware end of that mythical $100 computer, so let's focus back at the
software end without which the hardware is useless. We all seem to
assume that Linux plus Gnome or KDE plus Open Office is the way forward,
and maybe that is so, though I have some reservations that I will
attempt to unpick some other time. For the moment I want to come back to
the interface language, picked up by Lishan Adam and Edward Cherlin, and
explain why I think this is not that simple.

Lishan Adam says in his posting of 23rd February:

> It would be good if N. Negroponte sends a couple of the $100 computers
> to the universities in developing countries, especially in Africa where
> young people turn them around to speak local languages and talk to
> radios.

Computers should be given to colleges and universities, and the young
people there should be encouraged to make the translations and
localisations necessary. Most African languages are written in the Roman
script, or extensions of it, though there are important exceptions in
the north and east. This means that the first step, that of representing
the writing system in the computer, is unnecessary, or at most only
necessary for some diacritics that can be worked around. But then to
translate the interface text, help files, etc, requires agreement on
translations for the technical terms, and an agreed orthography
(spelling) for the language. In South Africa some languages have rival
orthographic systems, one set up by Dutch missionaries, the other by
English missionaries, each basing their spelling on the different
phonetic use of the Roman system in their own European languages. Some
agreement must be reached in the wider community long before these get
further embedded in software. And later, it you want spell checkers, you
need word lists and maybe some grammatical (morphological) understanding
of the language. You cannot leave all this to a group of enthusiastic
students; some community level process is essential. If you want to see
the kind of chaos that can result you need look no further than South
Asia, where during the age of True Type fonts all sorts of ad hoc
solutions for rendering the local complex writing systems were arrived
at, all mutually incompatible and incapable of exchanging information.

Looking to the Simputer, as Edward Cherlin does in his posting of 23rd
February, we see how grossly you can misrepresent what is necessary.
India has in excess of 500 languages, some say more than 1500 languages
(look on the CIIL or SIL websites), mostly unwritten. By contrast the
Simputer is enabled for a few of the dominant languages of India in
which there is a long established tradition of writing. This does not
mean that what the Simputer is doing is not worthwhile, just that it is
only scratching the surface of what is needed in India. Unwritten
languages need to be given appropriate writing systems, there needs to
be an agreed orthography, and so on. One thousand times over in South
Asia, six thousand times over worldwide.

However the underlying reasoning of Lishan is quite right, you cannot
impose localisation from outside, and for most of the world's languages
you cannot leave it to market forces either. The communities that use
these languages are too poor and as was argued in so many postings, have
other preoccupations about where to spend their money and time. The
example of Irish, excavated by Edward from the LRC website, is an
excellent example; there is no market for computers working in Irish so
the only way Irish gets into computers is through enthusiasts and
activists like Michael Everson. So if all these languages are to be
supported by computers, then the effort must be found in the local
communities where the interest is in doing so and the tacit expertise
lies. But this needs some gentle coordination and knowledge sharing,
like Localisation Dev is doing.

And that is also where the Global Initiative for Local Computing (GILC)
comes in. We want to help those critically important language and
technology activists around the world help themselves. GILC will be
formally launched at the "LRC-X Conference: The Development Localisation
Event" in Limerick, Ireland, 13-14 September 2005. In recognition of the
financial constraints that are placed upon many of the people that this
conference would be relevant to, the LRC have decided to implement the
following measures:

* Free conference registration for a limited number (50) of delegates
subject to individual financial constraints. Please contact <LR...@ul...>
should you feel that you might be eligible for a free registration. The
LRC will also assist individual delegates with accommodation expenses,
where needed.

* The authors of the 10 best papers submitted will receive free
conference registration and a cash re-imbursement to be put towards
their travel expenses. A review board appointed by the LRC will select
the papers.  For details and updates visit <www.localisation.ie>

Alternatively, if you are privileged enough to be able to attend the
Unicode conference in Berlin in April, come to our session. Otherwise,
email us:

Pat Hall <p.a...@op...>
Reinhard Schaeler <rei...@ul...>
Global Initiative for Local Computing



------------
***GKD is solely supported by EDC, a Non-Profit Organization***
To post a message, send it to: <gk...@ma...>
To subscribe or unsubscribe, send a message to:
<maj...@ma...>. In the 1st line of the message type:
subscribe gkd OR type: unsubscribe gkd
Archives of previous GKD messages can be found at:
<http://www.edc.org/GLG/gkd/>