[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Experimental countrychooser branch

On Mon, May 03, 2004 at 06:16:55PM +0300, Anton Zinoviev wrote:
> On  3.V.2004 at 13:56 (+0200) Christian Perrier wrote:
> > 
> > But they have to justify it with real arguments, not fake
> > arguments. The ones they give on their web site is currently a fake
> > argument.
> We are not in a position to decide which argument is real and which is
> not.  We already know -- there are arguments for both sides.  I think
> that for Debian it is better not to choose any of the existing
> standards as the decision about the best standard is more or less a
> political decision.  We should create our own standard.

AFAICT nobody ever complained that a package did not follow ISO-639 or
ISO-3166, I do not understand why we should be now stuck with these

> I already proposed one option for how to proceed [1].
> Another option is to create a new package gnu-locale-codes (I can do
> this if you want) mapping the language and country codes to names
> acording to the GNU locales themseves (in some sence this is the GNU
> standard).  For example acording to the locales gl_ES and gl_ES@euro
> the proper name of their language is "Galician".

All this is overkill.  On the long term, I hope that Alastair realizes
that iso-codes does not make much sense unless some other attributes
are added, like vernacular names.  When we ask user his location,
we are not interested in the official name of his country/region
as defined by UNO, but we want to display the name (s)he is supposed
to find here.
On the short term, I suggest that daily builds embed my hacked
countrychooser version for 1 or 2 weeks (if it works).  If people still
have complaints, we can go back to the current version.


Reply to: