frr, fy, ngo, tt (was: REQUEST for registration of variant subtag 'grabar')

[John Cowan]

Michael Everson scripsit:

> There are *thousands* of languages that use the Latin script. I do 
> not see why we should add a tag for Northern Frisian. This is just 
> too weird. It makes no sense at all.

Let's return to first principles here.

Suppress-Script: exists because the IETF was concerned about RFC 4646
placing script subtags in the second position.  The language tagging
community argued that script distinctions are inherently more important
than country/region ones, but the IETF folks pointed out that existing
RFC 3066 matchers would not be able to successfully match "en-Latn-US"
against "en-US".

In order to evade such problems, we agreed to provide information in the
registry to strongly discourage the construction of tags like "en-Latn-US"
at all, since it hardly ever means anything more than "en-US".

I think, therefore, that we need to consider the language in three
priority groups:

1) languages allowed in RFC 3066 that have different national/regional
forms;

2) other languages allowed in RFC 3066;

3) languages allowed since RFC 4646 went into effect.

In each of these groups there will probably be non-controversial and
controversial cases; we should consider the former first.

I'll post (some of) the current state of things shortly.

-- 
XQuery Blueberry DOM                            John Cowan
Entity parser dot-com                           cowan at ccil.org
    Abstract schemata                           http://www.ccil.org/~cowan
    XPointer errata
Infoset Unicode BOM                                 --Richard Tobin