Registration request for new variant subtag; Anii, Balanka

[Doug Ewell]

Mats Blakstad <mats dot gbproject at gmail dot com> replied to Peter
Constable:

>> If we have a variant subtag for this, would there also be a need for
>> a variant subtag to denote the Bassila / “standard” variant?
>
> This will not be necessary. There is a standard orthography (based on
> Bassila), so it should only be necessary to specify which the Anii
> variant if it is not according to the standard.
>
> This is how I see it at least. However, as far as I know, IANA don't
> have any guidelines if registering one variant also means that the
> other variant should be registered?

It wouldn't be an IANA guideline as such. It would be a combination of
what is set out in RFC 5646 and the precedent(s) set by the Reviewer,
more or less influenced by discussions on this list such as this one.

RFC 5646 doesn't seem to specify anything about this; any such passage
would probably be in Section 3.5 or 3.6.

Previous discussions and decisions have gone both ways on this. Usually
the discussion starts like this one: with a request to register a
variant for the less-common dialect or variety, the one that really
needs to be distinguished. Sometimes the decision has been made to
register all known varieties (Belarusian, Russian, Volapük
orthographies). Sometimes only the less-common variety has been
registered (Basque, Catalan, English varieties). We don't, for example,
have variant subtags for Modern English, for non-Scottish or
non-Liverpudlian English dialects, or for non-Biscayan Basque dialects.

Last November, for Kabuverdianu dialects, the original assumption was
that only 'barla' (the less common variety) would be registered, but
eventually an equal and opposite 'sotav' subtag was requested and
registered as well.

My opinion is that variants should be registered to solve actual tagging
needs, not just to tile the plane. Variants for the "common" or
"default" variety just create a divergent way to tag the most common
content, such as tagging today's edition of Pravda as "ru-luna1918".
However, there are plenty of other opinions on this, and ultimately the
Reviewer's is the one that matters.

--
Doug Ewell | Thornton, CO, USA
http://ewellic.org | @DougEwell ­