Preliminary proposal for 225 language-country pairs

Mark Davis Mark Davis" <
Mon, 18 Feb 2002 20:54:45 -0800

It seems arbitrary to generate all the possible variations, on the off
chance that some are usefully distinguishable. In terms of speech, for
example, mine is closer to a en-CA then it is to an en-NYC.

This is also not divorced from the notion of locale; the two are
closely intertwined in practice.


Γνῶθι σαυτόν — Θαλῆς
[For transliteration, see]

----- Original Message -----
From: "John Cowan" <>
To: "Mark Davis" <>
Cc: <>; "John Cowan" <>
Sent: Monday, February 18, 2002 20:27
Subject: Re: Preliminary proposal for 225 language-country pairs

> Mark Davis scripsit:
> > Why do this? en_US is already definable. Is it to be able to have
> > locale like en-PN_US, a locale with the conventions determined by
> > variety of English used in the US as spoken by immigrants from
> > Pitcairn?
> Locales have nothing to do with this.  RFC 3066 language tags come
in four
> flavors:  those registered with ISO 639, those registered with IANA,
> private-use tags beginning with x-, and tags of the form
> xx-yy or xxx-yy, which do not need to be registered with IANA but
can be.
> I am attempting to compile a list of (potentially) useful
combination tags.
> In theory all 436 * 239 = 17004 combinations are usable, but most
> of them are plain silly: there just is no distinctively Australian
> of Navajo.
> But by all reports en-PT is pretty different from en-US or en-GB.
> SIL goes so far as to call it a separate language.
> --
> John Cowan 
> To say that Bilbo's breath was taken away is no description at all.
> are no words left to express his staggerment, since Men changed the
> that they learned of elves in the days when all the world was
>         --_The Hobbit_
> _______________________________________________
> Ietf-languages mailing list