[Fwd: [FSF India] Localization of GNU/Linux to Indian languages]]

[Ramakrishnan M]

Satyakam Goswami wrote:
> 
> 1. unicode is a stupid. it doesnt exist. it doesnt work. if you wrote
> something for unicode, you can keep it. what works in india are two major
> standards  a) - the ISCII (the indian standard code for information
> interchange). the iscii standard, inspite of its name, doesnt cover all the

If my knowledge is correct, ISCII uses lower 128 codes for ASCII, and the upper
128 for *one* of the Indian Languages. This is stupid! An ISCII encoded
document can have only English and another Indian language. I cannot have
Sanskrit and Malayalam in one file?? (correct me if I am wrong)

Unfortunately Unicode is based on an early draft of ISCII, after that many
newer versions have come out. (I have a 1999 copy of it) So whatever
deficiencies were there in the older versions would have reflected in the
Unicode too. But definitely, Unicode is the way to go, because it addresses all
the issues that people were facing all these days, and brings in some standard
way of doing things, rather than using a bunch of crude techniques for
different languages.

> point him towards FSF. probably he is already a part of the movement but
> snoring. b) the PASCII (the perso-arabic standard code for information

He is runing a company, and might not be interested in Free Software, and also
I think he is into cable modems and things like that now, not Typography.

> 2. the biggest user of indian language computing is the government, and they
> want ISCII and PASCII. they dont understand Unicode (finally a sane

We can persuade them. Given a Keyboard and a terminal which can understand
Indian Languages, I don't think, they wil care for whether it is using Unicode
or ISCII.

> 3. write to rms for handa's contacts. handa got devanagri going on emcas.
> there is also a version floating around for arabic.

Yes, Devanagari is there in Emacs, but still GNU Emacs doesn't use unicode
(Unicode is not the right term here) internally, it uses a concept of 24 bit
atoms, I don't know anything more than that.

> 4. finally, there is already a large amount of code available for ISCII. i
> have seen a complete xterm run a devangri shell under linux. most of this
> work is due to rajeev sanghal (goose, put in his email id here...).

I know IIT-Madras has done a lot of work in this area
(http://www.tenet.res.in), unfortunately they too were aiming at ISCII, which
is shortsighted and stupid. I don't know whether they have released the code.

> 5. on the net, a large number of devnagri fonts are available in TTF (and
> they can be easily converted to PostScript. download them.

Are they *free* fonts? (Free in the GNU sense?)


-- 
Ramakrishnan M      WWW home: http://www.symonds.net/~rkrishnan/
--