Doug Ewell wrote:
>
> Antoine Leca <Antoine.Leca@renault.fr> wrote:
>
> >> 1) Normalize to single form when possible. Use ISO 639-1 code
> >> instead of 639-2 if one exists.
> >
> > Are you forced to re-tag every bit of data when ISO 639/RA issues a
> > new code?
>
> From what I have heard, ISO 639/MA will not be issuing any new 639-1
> (two-letter) codes for languages that already have a 639-2 (three-
> letter) code. So this re-tagging scenario should not occur and Carl's
> solution, which is the same as that proposed in RFC 1766 bis, should
> work fine.
I *should* have missed something.
In the last publication of new codes, there was "bs" for "Bosnian".
My understanding of the situation of the former Yugoslavia is that
the language which is intended to be tagged is a form of Serbo-Croatian
that is spoken in the country named Bosnia-Herzegovina (not sure
about Herzegovina), and outside this country by the natives or relatives
of natives of this very country.
Now this language is not a sudden invention: it was known before. And
as I understand things, this language was tagged "hr-XX-Bosnian", or
something like that (XX being the relevant country of the speaker).
So now the (probably fictious) document is supposed to be re-tagged
as "bs-XX". Or have I missed something?
Another example: a text in Avestan was, before the last change, tagged as
"x-Avestan" or "x-Avesta" or "x-zend" or a number of others, according to
the tagger. Now, should they be re-tagged? (and don't miss me, that will
be certainly a Good Thing; remember, the question is about the requirement).
Antoine
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:13 EDT