Indic in Mozilla

[Edward Cherlin <cherlin pacbell net>]

Translators, fonts, coders needed for full Indic support in 
Mozilla. I'm talking to several groups of font developers about 
this, including some at GNU-India working on the Akruti fonts.

----------  Forwarded Message  ----------

Subject: Re: Devanagari in Mozilla: some issues
Date: Friday 18 April 2003 08:45 pm
From: alok kumar mailjol com (Alok Kumar)
To: mozilla-i18n mozilla org

from indic-computing-users lists sourceforge net

We are looking for people to help us localize Mozilla to Indian
 languages. Rediff has agreed to hire people for this and their
 objective is to encourage the
penetration of the Internet in India. The programmers will work
 closely with the IndLinux.org team to localize Mozilla and the
 software thus developed will
be released in the public domain. This would be a good
 opportunity for those passionate about Indian language
 computing to work full-time in this area. The individuals
 selected will work out of Rediff's Mumbai office.

More details are below. Those interested are requested to write
 to us at: venky indlinux org and karunakar indlinux org

Regards,

Venky


Mozilla Indianization
==============

1) Goal: Enable Mozilla browser to display & handle indian
 languages properly, with proper font technology available.
 Localize Mozilla interface also to indian languages.

2) Project Requirements: All work done should be contributed
 back to Mozilla codebase, so that it becomes part of it.
 Solutions arrived at should make into main tree, rather than be
 done as adhoc hacks / addons / patches.

 i) i18n part (involving changing mozilla codebase) :
 a - Should be able to render Indian languages encoded in
 Unicode/UTF-8 b - Use a standardized font technology - eg
 OpenType.
 c - Support any encodings standardized for a language eg ISCII,
 TSCII, KSCLP (KGP)

 ii) l10n part (no code change - UI customization, translations
 of UI and Help text):
 a - Incorporate locale support (chrome level)
 b - Translate Mozilla UI into indian languages
 c - Making indianized mozilla theme - custom graphics, logos

3) Skill sets req
 a - Knowlegde of at least two Indian languages - reading &
 writing, script & language processing level  2.i & 2.ii.b
 b - C, C++, XML, Java for 2.i
 c - Graphics expertise for 2.ii.c
 d - For translator - good vocabulary in English & at least one
 Indian language, previous exp in translation also helpful.
 e - Proj. mgmt skills, ability to work with volunteers
 remotely, coordinate work through email/IM/group meetings.
 f - Interest in localization and passion for having indian
 languages on mozilla.
 g - Work experience of 2-3 years in the IT industry

Current status:
 Mozilla supports Unicode, so all indian languages represented
 in Unicode work at unicode level - no proper rendering, if  a
 suitable unicode font is available. ie proper vowel sign
 positioning, conjuncts etc are not avialable. Some work is on
 by Prabhat Hegde from Sun Microsystems on adding devanagari
 support, basically by using Pango (www.pango.org) rendering
 engine. He is exploring doing it using Opentype fonts. (Any one
 working for this should work with Prabhat for
guidance & review, as he already has experience in Mozilla
 codebase )

Some hindi translation work has been done by a group based in
 Mumbai, available at http://www.bttlindia.com/mozilla/ . Some
 translation work for Tamil is also on. (These translations
 should be reused, instead of starting from scratch)

Notes:

2.i has to be done in active participation with Mozilla
 developers & done at those forums ( mozilla-i18n , mozilla-l10n
  )
2.ii.b is pure translation work, can be done in centralized or
 distributed fashion.
 2.i.c - Since ISCII, TSCII etc are not registered character
 sets, it may not be possible to use them with Mozilla
 autodetection feature. Some workaround could be suggested.
 ISCII text will basically be converted to Unicode internally
 for display, enabled through use of 'iso8829-12' reserved for
 Indic - ISO is waiting for us to settle to something :). (The
 rationale to support ISCII is since ISCII chars will take
 1byte, UTF-8 indic characters take 3bytes, or  7 bytes when
 used as named
character references - The debate will pick up heat when indian
 languages are fully supported )

-0-

> > > ==
> > > http://groups.yahoo.com/group/linux-bangalore-hindi/
> > > can't see Hindi?
> > > http://geocities.com/alkuma/seehindi.html
>
> ==========================
> free hindi unicode editor at
> www.geocities.com/hanu_man_ji

-------------------------------------------------------

-- 
Edward Cherlin
Generalist & activist--Linux, languages, literacy and more
"A knot! Oh, do let me help to undo it!"
--Alice in Wonderland