draft-phillips-langtags-04 /2.3 UND

Fri Jul 2 20:39:59 CEST 2004

Thanks for the comments Misha, Jon, John.

1) However, section 2.2 says:
"The primary subtag is the first subtag in a language tag and cannot be empty."

The BNF also seems to indicate that it is not optional.

So how do we indicate the empty string?
;-)

2) I have to admit when I read:
"You SHOULD NOT use the UND (Undetermined) code unless the protocol in use
forces you to give a value for the language tag, even if the language is
unknown.  Omitting the tag is preferred."

I thought "Omitting the tag" meant not specifying language at all as opposed to
using an empty string to represent it.
i.e. If we were using markup, omitting the tag is this case:
<html></html>

as opposed to
<html lang=""></html>

but that is only possible in programming syntaxes that optionally declare
natural language.
A structure that had a fixed field for language would have no choice but to use
an empty string or UND.

So I now interpret "omitting the tag" as being equivalent to "use the empty
string".
It's fine if no one else had that confusion. Otherwise, the authors might be
more explicit.

3) Anyway, I still have a concern that the spec should indicate an empty string
can have multiple meanings and side effects.

(inherit or use a default value) vs. atheism vs agnosticism.

The authors might choose not to attempt to distinguish these cases in the
syntax in this spec, but it would be good to caution readers of the spec that
receiving an empty tag could imply any of these situations.

I do have a concern that empty strings be further distinguished for web
services... but it can be addressed by that group.

tex

Jon Hanna wrote:
> 
> Quoting John Cowan <cowan at ccil.org>:
> 
> > Misha Wolf scripsit:
> >
> > > XML 1.1 specifies that the empty value of xml:lang be used for
> > > this purpose:
> > >    http://www.w3.org/TR/2004/REC-xml11-20040204/#sec-lang-tag
> >
> > Not really.  You are confounding, as it were, agnosticism and atheism.
> > UND says that the language is unknown to the tagger; the empty string
> > doesn't say anything.
> 
> I can't see anything to indicate that UND means anything more than the language
> isn't stated; whether because the tagger is ignorant or just isn't saying.
> Likewise with the empty string for xml:lang.
> 
> The use of the empty string rather than UND matches the RFC "You SHOULD NOT use
> the UND (Undetermined) code unless the protocol in use forces you to give a
> value for the language tag, even if the language is unknown.  Omitting the tag
> is preferred."
> 
> --
> Jon Hanna
> <http://www.hackcraft.net/>
> "It is the most shattering experience of a young man's life when he awakes
> and quite reasonably says to himself, 'I will never play The Dane.'"
> _______________________________________________
> Ietf-languages mailing list
> Ietf-languages at alvestrand.no
> http://www.alvestrand.no/mailman/listinfo/ietf-languages

-- 
-------------------------------------------------------------
Tex Texin   cell: +1 781 789 1898   mailto:Tex at XenCraft.com
Xen Master                          http://www.i18nGuy.com

XenCraft		            http://www.XenCraft.com
Making e-Business Work Around the World
-------------------------------------------------------------