Timetable for action: May 31 is suggested
Sean M. Burke
sburke at cpan.org
Tue May 27 20:07:17 CEST 2003
At 01:29 AM 2003-05-28 +0100, Michael Everson wrote:
>>would be to have something else like two separate entities, each
>>specified in a separate RFC, one of which is RFC 3066.
>Yes, I had suggested that a long time ago. Apparently this will fail
>because current imlementations want to conflate script within language tags.
The only expression I saw of this "lang=X, script=Y" notion was one where
one said "Accept-Language: uk, ru, sr, fr, it, zh" and then "Accept-Script:
cyrl, latn, hans" and somehow it gets figured out by elfin magic whether
the user wants his Serbian in Cyrllic, Latin, or Han Simplified.
The obdurately messy fact is that script preferences are /so/ tightly
coupled to language preferences that it's ridiculous to /not/ "conflate"
them. This is a matter not of "current implementations", but of how
hyoo-mons think. Filthy filthy hyoomons! When will they learn the joys of
orthogonality and the austere gloire of a life lived by the carpenter's
T-square and the Etch-A-Sketch!?
Yes, there is the /question/ of how to combine script subtags and country
subtags: zh-sg-hant or zh-hant-sg. But this is a question suggested by
reasonable curiosity and a whim for "completeness". It is not a question
required by urgency or need. The languages that /actually/ have scripts
and are /actually/ used in multiple countries would only produce one or two
dozen permutations total. One might as well just arbitrarily list them,
rather than pining for the coming of a glorious metaprinciple that would
make all our mundane decisions for us. This is not a Pandora's box. The
language tags are not here to harm you. They do not bite.
Sean M. Burke http://search.cpan.org/~sburke/
More information about the Ietf-languages