[Fwd]: Response to Mark's message]

[John Cowan]

Mark Davis scripsit:

> written language ID
> country ID of residence
> country ID of citizenship
> country ID of bank account
> timezone ID
> 
> [This is actually the defintion needed by a customer I was talking to just
> yesterday.]

OT remark: I hope the customer knows enough to use real GNU/Olson timezones
and not just UTC offsets, especially if they need to handle anything other
than current times.

> ISO-639 fails miserably as unambiguous specification of written language. I
> realize that the proponents of ISO-639 don't even want it to apply to
> written language. 

Au contraire: the requirements for ISO 639 languages are precisely the
existence of written documents in public repositories (prototypically
libraries).  ISO 639 is basically a blessing of a MARC standard devised
by the Library of Congress.

> RFC 3066 is somewhat better, but has the problems as discussed on this list.
> As to the issue of whether RFC 3066bis should include SIL codes directly or
> not, technically I don't much care. I suspect it would be slightly cleaner
> if 3066bis just included some ISO standard.

It would, if it weren't for the massive problems with ISO 639 even as a
standard for written languages.  There are far too many codes whose precise
referents aren't really well defined.

> A. 3066bis adds script codes, SIL codes

What I favor.

> B. 3066bis adds script codes, ISO adds SIL codes to ISO 639-3, (later)
> 3066bis#2 adds ISO 639-3 codes

That could take a looooong time.  3066ter (that's the jargon) would be
an awful long time coming.

-- 
A mosquito cried out in his pain,               John Cowan
"A chemist has poisoned my brain!"              http://www.ccil.org/~cowan
        The cause of his sorrow                 http://www.reutershealth.com
        Was para-dichloro-                      jcowan at reutershealth.com
Diphenyltrichloroethane.                                (aka DDT)