Re: [Indic-computing-users] Incorporating Indic Computing

[Hema A M. <he...@la...>]

Please bear with me, this is a little long.

> For a long while, some of us (including myself, of course) have felt the
> need for an organization that can take the mission of Indian language
> computing forward. When Prakash Advani and I started IndLinux in 2000, we
> hoped that a lot of volunteers would step forward and help us localize
> Linux, but two years later we find that whatever we have done, it is because
> we have one person (Karunakar) working full-time on localization.

Just a correction to this (this list is not exhaustive) 

i) TeNeT Group, IIT Madras: a team of  5 people (ALL OF THEM 
ARE EMPLOYED ON A PROJECT) where the job is localisation of linux. 
(www.tenet.res.in)

ii) Prof.RKK (SDL, IIT Madras) and his team (please look at the website 
www.acharya.iitm.ac.in) has quite bit of localisation under
linux.

iii) AU-KBC has work going on in Tamil Linux

iv) IIT Kanpur has a project under TDIL for Linux with Hindi
(they also have modified X) (Prof.Rajat Moona)

v) NCST, Mumbai has also Indianised X.

In my opinion, there is a general DISMISSAL of all this work
by the Indic Community - primarily because they are not unicode
compliant (RKK's work) or are too general (TeNeT) in that they
support any kind of encoding.

I for one am not too excited about Unicode - primarily because
the population that we are targetting (the rural users) donot
really care about the support for ALL languages of the world.
I would rather use the entire 16bit (as in RKK's scheme) for
a compact representation of C^*V clusters.  

i) Perhaps our rural users may want support for 3 languages.  

ii) Further, there is already a lot of content on the web: there is
a lot of inertia in the field to shift to something new - an
example again:

a) B(h)amini is a Tamil font with a typewriter
like layout used in our kiosks at Madurai for inputting
data in Tamil.  Although we have provided a proper typewriter
layout with much better fonts, our operators are not willing to
shift - alternatively, we just changed the input layout to suit
their requirements.

b) Somebodyelse in Delhi uses Baraha font and refuses to shift
to somethingelse.

Comment: If we want all these people to shift we should be able support
unicode with ALL the existing fonts and not insist on Opentype.

Some of my thoughts:

a) I do think that multiple standands will continue to co-exist
(this is India with all its DIVERSITY :-)) and all that we CAN
do as part of the Indic Community is to understand and accept this.
We have to make sure that we have appropriate tools in place 
to perform conversions across different languages, different
fonts and different encoding schemes.  These could be filters
that a browser could use for:

i) searching
ii) downloading
iii) converting (today for example, we are quite comfortable
     when the browser supports multiple formats for saving
     web pages).

b) We HAVE to work with TDIL, MIT - irrespective of
how bureaucratic this organisation is - I personally do
think UNLESS THERE IS A POLITICAL WILL NOTHING ON A MASS
SCALE can be achieved. 

c) The Indic Computing group should also include speech
and handwriting interfaces (given the C^*V clusters in
Indian languages, these interfaces will be required).

d) The Indic Computing group should get itself into a position
where it advises the Govt./National Funding agencies about 
Open Source and its advantages for Indic Computing.  This requires
that we establish credibility - I feel we should host a 
COMMON website, where all the software available from the
community are categorised and posted.

Regards, Hema