This is the mail archive of the mailing list for the glibc project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: locale for Uzbekistan


I'm a Serbian Gnome translation coordinator.

It was decided to use "sr@Latn" for the latin transcription of *default* Serbian cyrillic translations. We've got more than 50k strings translated, and around 150 modules, which means 150 files named "sr@Latn.po".

For that purpose, I've also updated the sr_YU locales and named them sr_CS (country code of "Serbia and Montenegro", formerly "Yugoslavia", has changed on July 24th to "CS" from "YU").

I've submitted both sr_CS and sr_CS@Latn locales to libc-alpha a while ago, but there was apparently not enough interest (they are available at at all times, I'd be working on updating
a latin collating order, because cyrillic is fine with UCA/ISO14651 sorting).

There's nothing ugly about naming. If the case of the letters is a problem, I have to wonder how is a different case in country codes treated, so as not to cause any problems.

Btw, since ISO 15924 lists script codes with first letter uppercase, there is *no* ambiguity with other modifiers such as "@euro" or anything else.

That's also advantage, because identifying locales by script is a bit more common than other things (Uzbekistan and Serbia-Montenegro are already practicing that usage; others may follow [like zh@Latn, other Asian countries, etc] especially when there's already a pattern one can use).

So, for all mentioned problems there is a solution. If I need to submit any papers to FSF in order to get modified locales in GNU libc, I'd be glad to do so.

Please don't create problems where there are no problems :-)

Cheers, Danilo

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]