backpeddling on suggestion
petercon at microsoft.com
Sat Aug 14 21:59:12 CEST 2004
A couple of rounds back, I suggested that the syntax allow for extended
language subtags, which would accommodate potential request and, in
particular, some existing registrations, such as zh-min and zh-min-nan.
There's a couple of problems with this.
First, in my mind, I was conceptualizing the possibility of allowing
multiple extlang subtags as utilizing the notion of macro-language being
introduced in ISO 639-3 (I won't go into details on that at the moment).
Effectively, when zh-min-nan was registered, it was applying the
macro-language notion: zh is both a single language identity but at the
same time includes multiple languages; min was used in the registrations
to represent the union of the various Min languages. It was premature of
me, however, to assume that the possibility of macro-languages
encompassing other macro-languages should be permitted in ISO 639-3;
it's something that invites abuse and so should be considered carefully.
What is really a major problem here is including extlang in the syntax
and having it reference ISO 639. It creates an immediate conflict in the
case of zh-min and zh-min-nan (I should have checked this straight off a
few months ago): "min" is assigned in ISO 639-2 to the Minangkabau
language, an Austronesian language spoken in Indonesia and completely
unrelated to the Min languages of China. (There isn't a problem with
"nan", though: I have assigned it in the draft table for ISO 639-3 to
Min Nan.) If we leave things the way they are, then the Min Nan /
Minangkabau conflict will need some kind of resolution.
As soon as I get this draft for ISO 639-3 finished, I'm off on vacation;
I'll have to let you all thrash this one out for the next several days
Globalization Infrastructure and Font Technologies
Microsoft Windows Division
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Ietf-languages