Request for variant subtag: western

Peter Constable petercon at
Wed Aug 30 03:38:06 CEST 2006

Please explain. Your mail to others does not AFAICT address the specific point I make here: if you revise a variant subtag by adding additional prefixes, we have no means to document what the denotation of some given prefix + variant combination is. 


E.g. suppose the entry for a variant subtag “western” were revised to include the prefix “en”, and some people start using it to denote some variety of English spoken in Cornwall while others start using it to denote some variety of English spoken in Western Australia, then there’s a mess.






From: mark.edward.davis at [mailto:mark.edward.davis at] On Behalf Of Mark Davis
Sent: Tuesday, August 29, 2006 5:02 PM
To: Peter Constable
Cc: IETF Languages Discussion
Subject: Re: Request for variant subtag: western


That's based on a mistaken assumption (see my mail to others). Mark

On 8/29/06, Peter Constable <petercon at > wrote:

> From: John Cowan [mailto:cowan at ]

> As for hy-hyewest, I think it's awful.

While it may look aweful, an advantage of this is that we can document precisely what is meant by "hyewest" (or whatever form of subtag one might want) -- that is, it is the Armeanian variety spoken by the Armenian diaspora. 

But there is no way we can document how "western" ought to be interpreted in all of the combinations it is used (unless we want to allow multiple "western" variant entries, one for each prefix to be documented). 

And going back to a paper Gary Simons and I wrote back in 2000, documenting what tags are meant to denote is important. It won't help anybody if people start using "foo-western" but there are widely differing opinions as to what that refers to. 

Peter Constable
Ietf-languages mailing list
Ietf-languages at


-------------- next part --------------
An HTML attachment was scrubbed...

More information about the Ietf-languages mailing list