Re: [Indic-computing-users] RE: [Bengalinux-core] Gnome glossary translation started

[Guntupalli K. <kar...@fr...>]

On 23 Sep 2002 23:36:36 +0530
Sayamindu Dasgupta <unm...@So...> wrote:

> On Mon, 2002-09-23 at 23:01, Ahmed, Taneem wrote:
> > Hi Sayamindu,
> > 
> > Sorry to email again, but something just hit me. As I mentioned in
> > my last email that we may want to send out the list of words to
> > people, I was trying to figure out who would be our target
> > audience. Then I realized that the response will be quite
> > different between Indian Bengali and Bangladeshi Bengali people.
> > Do you think it would make more sense to do the translation under
> > bn_IN and bn_BD like en_GB and en_US? 
> > 
> 
> hmmm.... important observation - cc-ing to all
> but, going by that argument, we would have to do not only bn_IN and
> bn_BD, but also, bn_CHT (chattagram - my ancestors are from that
> region), bn_MD (midnapore), bn_NB (north bengal) and what not.
> I believe that Dhaka University still recognises the Bengali grammar
> and dialect proposed by Calcutta University (though this is highly
> unfair - sort of cultural and linguistic dadagiri) 
> So, let us go together now - once a bit matured, we can branch out
> to handle the dialectical differences
> 
  You need not worry on all that.  ISO locale naming scheme is (from
man setlocale) is

       A locale name is typically of  the  form  language[_terri-
       tory][.codeset][@modifier],  where  language is an ISO 639
       language code, territory is an ISO 3166 country code,  and
       codeset  is  a  character  set or encoding identifier like
       ISO-8859-1 or UTF-8.  For a list of all supported locales,
       try "locale -a", cf. locale(1).

so bn is language code for Bangla,  territory is the country codes (
IN for India, BD for Bangladesh )
Any further variants within a territory can be denoted by modifier.
 so you could have  bn_IN@chattagram or bn_IN@northbengal , We would
actually be having the largest no. of locales ( around 1600, one for
each dialect , glibc guys will go crazy :)
 Only the minor variations need to be considered  in locale variants.
Assume everything is common first , then branch off when you know the
differences.
 So in India we would be using  bn_IN and in Bangladesh bn_BD, both
may have same locale database & translations. Going by that logic
English users in India should be having default locale  en_IN not
en_US ( its already define in glibc ).

Regards,
Karunakar