Proposal to remove Preferred-Value field for region YU in LTRU

John Cowan cowan at ccil.org
Mon Mar 2 18:23:58 CET 2009


Broome, Karen scripsit:

> I realize that this use case is not related to language, but in
> this case suggesting CS would not be appropriate. I almost wonder
> if countries should be considered "deprecated" in the traditional
> sense. They may no longer be current countries of the world, but that
> doesn't mean that people aren't going to continue to use these tags
> in appropriate ways for historical content.

The region subtags in BCP 47 tags are meant only to discriminate among
varieties of language that happen to be aligned with those particular
regions: semantically, they are just like variant subtags.  It so
happens that in many parts of the world language varieties *do* align
with countries, and it is also true that many languages have official
orthographies (de jure or de facto) that vary by country.  Our registry
is by no means meant to replace ISO 3166 itself, much less provide a
historical conspectus of it.

> If I were to represent our data properly in ISO 3166-1, I need a tag
> for Bophuthatswana, which existed so briefly, it never had an ISO tag.

"What is a country?" is a vexed business.  In this case, the rest of
the world decided to treat Bophuthatswana as still part of South Africa,
as it was before 1977 and after 1994.  The IETF doesn't get involved --
we just track ISO 3166, which just tracks UNSD M.49, which makes its
decisions primarily on economic grounds.

-- 
He played King Lear as though           John Cowan <cowan at ccil.org>
someone had played the ace.             http://www.ccil.org/~cowan
        --Eugene Field


More information about the Ietf-languages mailing list