New draft submitted of 3066bis...

McDonald, Ira imcdonald at
Tue Nov 2 18:15:45 CET 2004


I thoroughly agree with John Cowan below.

In both RFC 1766 and the successor RFC 3066,
the ABNF limits the values to using ALPHA
or DIGIT (per RFC 2234).

'xml:lang' may or may not be broken in W3C
specs.  Not our problem.

But RFC 3066 is a BCP (Best Current Practice),
which is a normative statement of IETF policy.
There's no question of 'advisory', as Chris
Lilley reports hearsay of below.

And there's no possibility (or utility) for
ever extending the allowed characters in
RFC 3066 or successor language tags.

Case folding for ASCII _in_this_domain_
is strictly locale independent.

- Ira

Ira McDonald (Musician / Software Architect)
Blue Roof Music / High North Inc
PO Box 221  Grand Marais, MI  49839
phone: +1-906-494-2434
email: imcdonald at

-----Original Message-----
From: ietf-languages-bounces at
[mailto:ietf-languages-bounces at]On Behalf Of John Cowan
Sent: Tuesday, November 02, 2004 11:40 AM
To: Chris Lilley
Cc: Martin Duerst; ietf-languages at; mark at;
Elliotte Harold; Norman Walsh; mark.davis at
Subject: Re: New draft submitted of 3066bis...

Chris Lilley scripsit:

> Right, that is my analysis also. However, it has been stated that
> RFC3066 is advisory, and you can put any string you like as a value of
> xml:lang. I think that is wrong too, and the XML spec clearly says
> either an RFC3066(or successor) value, or "".

Ah, I understand now.  The plain language of the spec is to be ignored.
"Fiddle, we know, is diddle; and diddle, we take it, is dee."
        --Swinburne (see

Winter:  MIT,                                   John Cowan
Keio, INRIA,                                    jcowan at
Issue lots of Drafts.                 
So much more to understand!           
Might simplicity return?                        (A "tanka", or extended
Ietf-languages mailing list
Ietf-languages at

More information about the Ietf-languages mailing list