Language codes

[Doug Ewell]

Mark Davis ⌛ wrote:

> What we do is act like "UK" was in the registry, but deprecated in 
> favor of "GB", since that solves the problem.

By "we" I assume you mean Google.  Certainly, any implementer of a 
matching engine can add their own private extension such as this.

> (What would have been a good general solution.)

No, it wouldn't.  Are we going to carry over this battle from LTRU? 
Adding a pre-deprecated, *wrong* subtag to document and legitimize some 
users' *wrong* usage would not have been a good general solution.

RFC 5646 points out (correctly) that deprecated subtags are "valid in 
language tags" and can even be "preferred in certain contexts."  This 
would have sent the wrong message entirely in the case of 'GB' vs. 'UK'. 
The place for users to find valid BCP 47 subtags is in the Registry, not 
at the end of e-mail and Web addresses, not on the white elliptical 
decals on the rear bumpers of cars, not in the abbreviate forms that 
people use in speech.

--
Doug Ewell  |  Thornton, Colorado, USA  |  http://www.ewellic.org
RFC 5645, 4645, UTN #14  |  ietf-languages @ http://is.gd/2kf0s