[Ltru] Re: Small bibliography error in RFC 4930

Martin Duerst wrote:

 [I've stripped "ietf-provreg" from the Cc: list]
> Randy made the procedural argument (both RFC 3066 and RFC 4646
> are BCPs).

Just to clarify, I think you or Randy mean "RFC 3066 used to be
a BCP before RFC 4646 obsoleted it".

> I have absolutely no clue what could break in an implementation
> when using a tag that became legal in RFC 4646.

Matching is tricky in the presence of script subtags (and other
less likely scenarios), it was one line in RFC 3066, now it's a
separate document.

For 4646bis no implementation will be able to match nl-BE with
what-was-it, vlm (?), by only looking at the registry, unless
we deprecate it manually in favour of nl-BE.

It makes perfect sense to exclude some constructs.  When I try
the language option of a "popular" search engine I won't find
de-1996 pages while looking for de, or en-GB-oed pages while
looking for en (I'm too lazy to check what it was, maybe both).

And I won't find frr pages while looking for fy-DE (or v.v.).
But that's no 639-3 or 4646 issue, it would be the same with
3066.  For 3066 stability was an implicit assumption, and 4646
made it explicit, trying to guarantee it for region tags.

RFC 3066, 4646, and 4646bis fail miserably when language tags
are narrowed, or when language-region combinations get their
own language tag.

Maybe we should keep RFC 4646 as is, and later replace it
wholesale by ISO 639-6.  The fy-DE and nl-BE examples (plus
the "extlang" discussion) are a warning.

This nl-BE confusion is in essence the same issue as the old
CS chaos, we're trying to mix standards designed to be used
for something else.  For ISO 639-3 it's logical to replace
nl-BE by something "better" as soon as somebody seriously
wants it, they don't know any "nl-BE", they only know "nl".

But we know that nl-BE exists, and we've no way to deprecate
it in favour of whatever comes, we could at best manually
deprecate the new subtag in favour of nl-BE.

For fy-NL, fy-DE, frr, and fy, let alone frs, I thought that
it's okay to ignore the implications, after all I tried to
find a single page tagged as fy-DE and found nothing.  But
for nl-BE (and zh-*) we might be in serious trouble.

Frank

_______________________________________________
Ltru mailing list
Ltru@ietf.org
https://www1.ietf.org/mailman/listinfo/ltru