Re: Language environments

[Bruno Haible]

Paul Eggert writes:
> I'm not sure that a call for new language environments would be a good use
> of our time at this point.

Please note also that not every locale for which there is a
translation team should correspond to an Emacs "language
environment". For example, "Brazilian Portuguese (pt_BR)" (for which
there is a translation team) and "Austrian (de_AT)" (for which there
isn't) can use the language environment of pt and de.

> That info seems to be absent from gettext 0.10.40; perhaps it was too
> much of a pain to maintain?

No, it's simply a recent addition.

> While we're on the subject, gettext-0.10.40/doc/ISO_639 seems to be
> derived from an older version of
> <http://www1.ics.uci.edu/pub/ietf/http/related/iso639.txt>; e.g. it's
> lacking "sh" for Serbo-Croation and "Sangho" is misspelled.  Perhaps
> gettext should be updated?  Sorry, I don't know where the canonical
> copy of that file is, but apparently it's not at dkuug.dk any more.
> 
> (Other resources in this space include
> <http://www.oasis-open.org/cover/iso639a.html> and
> <http://www.w3.org/WAI/ER/IG/ert/iso639.htm>.)

The resource I use is
http://lcweb.loc.gov/standards/iso639-2/englangn.html
(labeled "ISO 639-2 Registration Authority"). It spells sg as
"Sango" and doesn't mention Serbo-Croation (which IMO was a political
fiction of the past).

Bruno