Re: Two translations for Uzbekistan

[Pablo Saratxaga <pablo mandrakesoft com>]

Kaixo!

On Fri, Oct 03, 2003 at 05:44:13PM +0200, Mashrab Kuvatov wrote:

> Mandrake Uzbek translations exist in both scripts. uz and uz@Cyrl
> is used for latin and cyrillic scripts respectively. uz@Cyrl locale
> was suggested by Pablo Saratxaga - person in charge for i10n and
> i18n @ MandrakeSoft.

I did that following the precent of Serbian translations (wich uses latin
and cirillic too).
Serbian uses "sr" for cyrillic and "sr@Latn" for latin script, as cyrillic
is the somewhat official script. For Uzbek the official script is the
reverse, but I followed the same logic to suggest the names to use for
translations.
 
> Recently, Uzbek-cyrillic was included into glibc
...
> Finally, it became uz@cyrillic.

Too bad it hasn't been standardized to follow current usage on translations;
that means that values of LC_*/LANG and LANGUAGE will differ...
Well, the situation with Uzbek isn't as bad as with Serbian, as the unadorned
name ("uz") has the same meaning in both cases; for Serbian however, "sr"
has opposite meanings in default glibc locale and in LANGUAGE as used
by Gnome...
 
> I would like to introduce Uzbek-cyrillic and -latin translations to KDE
> and later on to GNOME.
> 
> Is it possible to have Uzbek-cyrillic and -latin translations in KDE / GNOME ?
> If yes, how should Uzbek-cyrillic locale look like ?

It can be anything (well, almost anything); however I strongly feel there
should be some consistency in naming (for the same reason that there isn't
a mix of two- and three-letter codes; or a mix of language only and 
language_COUNTRY codes (for cases were it isn't needed; eg in the past there
were in Gnome Bulgarian translations named as bg.po, bg_BG.po
and bg_BG.CP1252.po; hopefully they all normalized on bg.po now).

Until now, the only case I know of a script distinction for a same language
that is actually largely used in translations, is for Gnome translations
for Serbian (and in mdk too, that standardized on what has been done
in Gnome).
AFAIK KDE still has Serbian support in latin only.

Now, there is nothing against using "uz.po" and "uz@cyrillic.po" in Gnome,
but then it will use a different logic than Serbian, which will lead to
uncertainty when another language with the same problem will appear.
By using the same logic as Serbian currently uses, eg: "uz.po" and
"uz@Cyrl.po" it is more consistent.

About KDE, as it hasn't done it yet, it could choose anything, however,
it would be a wise decision to choose the same naming for po files in all
projects.

> 
> Regards,
> Mashrab.
> -- 
> Mashrab Kuvatov
> Ph.D student
> University of Bremen, IUP
> www: www.sat.uni-bremen.de/members/mashrab

-- 
Ki ça vos våye bén,
Pablo Saratxaga

http://chanae.walon.org/pablo/		PGP Key available, key ID: 0xD9B85466
[you can write me in Walloon, Spanish, French, English, Italian or Portuguese]

PGP signature