[Reader-list] Monthly posting

[surekha at servelots.com]

Hello all,
    I am sending a monthly posting for the project titled "Multilingual
support in web applications using server-side java".

Surekha.

----------------------------------------------------------------------------
                           Monthly Report May’2004
         Multi-lingual support for Web applications using Server Side Java
----------------------------------------------------------------------------
-----------

A collaborative work with K Srinivasa Raghavan

Abstract of proposal:
--------------------

   Whether it be e-mail or a project like PANTOTO [http://www.pantoto.com],
   internationalization [i18n] allows the application to work in a specific
   local language. Multi-lingual applications however, go a step further and
   enable the user to switch locales (language) dynamically at run-time.

   The proposal is to investigate the issues involved in localizing a web
   application and providing multilingual support using server side Java and
   popular browser based scripts.

Current status of project:
-------------------------

   1. Input Method Editor (IME):

   An Input Method Editor (IME) in JavaScript has been developed which
allows users to
   type in any of the Indian Scripts (Devanagari [Hindi, Marathi,
Sanskrit], Bengali,
   Gurumukhi, Gujarathi, Oriya, Tamil, Telugu, Kannada and Malayalam) in Web
   applications (Browsers). This Input Method is platform and browser
independent.
   It has been tested on Netscape7 and Mozilla 1.1 on Linux and IE6.0 and
Netscape7
   on Windows. There is a problem in rendering Indic content in Netscape7
on windows
   and Mozilla1.1 and Netscape7 on Linux.

   The IME is developed for two keyboard layouts namely “WX” and “Inscript”.
   “WX” is a notation developed by NLP team at IIT-Kanpur.

   We have also integrated this Input Method with the PANTOTO Communities
Software.

   2. Data storage:

      Indic data can be stored at the backend either in ISCII encoding
scheme or in
      Unicode encoding scheme.

      Presently we are storing data in ISCII (Indian Standard Code for
Information
      Interchange) at the backend to analyze the problems associated with
this
      encoding scheme.  In case of multi-lingual content, we need to prefix
a language
      tag whenever there is a change in language. This required us to make
changes to
      the conversion utility 
(http://www.sarovar.org/projects/codeconverters) developed
      earlier, to store multi-lingual content in ISCII encoding scheme. The
updated
      conversion tool will soon be made available in the public domain.

      We have been able to store data in ISCII encoding scheme for two
Indian Scripts
      (Devanagari and Kannada) successfully.

Road ahead:
-----------
 1. Input Method Editor (IME):

      We are also planning to make this IME as an independent tool that
sits on the
      browser at the client end and can be invoked on any web page.

 2. Data storage :

      We will try storing data in Unicode at the backend and will analyze
the
      issues involved in this encoding scheme.



--------------------------------------------------------------------
mail2web - Check your email from the web at
http://mail2web.com/ .