Peter Constable petercon at
Fri Feb 18 16:02:32 CET 2005

> From: ietf-languages-bounces at [mailto:ietf-languages-
> bounces at] On Behalf Of Michael Everson

> >Also, I think the reluctance to register a tag like iu-Cans-CA is
> >mistaken on other grounds: we are not obligated to determine that
> >valid tag denotes something distinct from every other valid tag.
> We are not? ISO 639 does, and this RFC is an extension of that.

This RFC is not simply an extension of ISO 639. If all it did was allow
people to register tags of the form 

"i-" 1*8alpha

for languages not listed in ISO 639, with no other subtags adding
regional or other qualifiers, it would be an extension of ISO 639 and
subject to similar criteria.

> >I suspect that there's no distinction between fr-CI and fr-GH, but
> >both are valid tags, and probably in use somewhere.
> Probably?

Given the tendency of people to make ll-CC combinations, it seems
reasonably likely to me. 

> >And note that, while I think there's no distinction, someone else
> >may determine, for whatever reason, that they think they need to
> >distinguish something in this way.
> At least there's a distinction between -CI and -GH.

So then there's a distinction between mn-Mong-CN and mn-Mong-MN.

> There is no
> distinction between Cans and Cans-CA, since Cans isn't used anywhere
> but CA.

I conceded for the particular case iu-Cans-CA that I could not provide a
scenario in which it might be needed.

> >The important thing for us is not to establish precisely what every
> >distinction is (an endless task involving an ever-changing domain
> >which different interpretations are possible), but rather to ensure
> >the intended meaning of any tag is understood by all and for which it
> >clear, to some minimal level, how to utilize it.
> I thought the point was to register things that people need to make
> distinctions. You know. Distinctions.

My point has been that tags like mn-Mong-CN vs. mn-Mong-MN make a valid
distinction even if we cannot point to a corresponding dialectal or
spelling difference.

> Like Scouse isn't standard
> English, so it needs a tag. Like Inuktitut in Syllabics isn't just
> Inuktitut, so it needs a tag. But there's no difference between
> Inuktitut in Syllabics and Inuktitut in Syllabics in Canada, so why
> would you need that tag?

The only reason, as I've conceded, for that case is that, if we had to
consider many others like it, it would not be feasible for platform
developers to anticipate when the country would or would not be needed,
just as John and Tex found that for ll-CC combinations it was far from
trivial to determine when the country ID would or would not be needed.
Only a user of the tags is in a position to know when there's a reason
for including the country ID, whether we're talking about ll-CC tags or
ll-Ssss-CC tags.

Peter Constable

More information about the Ietf-languages mailing list