Re: [Ietf-languages] Proposal for variant language subtags for Pe̍h-ōe-jī romanization for Minnan (nan)

[Doug Ewell <[e-mail]>]

Winston Sung wrote:

> So which solution would be better?
> * Propose a single subtag for "vernacular orthography" and include all
> of initially used and adopted languages; or,
> * Propose different subtags for different vernacular orthography/ies
> in different languages?

You should propose subtags that meet the needs you have, to identify and exchange linguistic content.

If those same subtags would be reasonably, foreseeably, and accurately useful with Prefix values other than the one(s) that meet your needs, then including those Prefix values may be a good idea.

Most people who propose a variant have a particular use case in mind. I’m not personally a fan of expanding that into an effort to partition all language usage using BCP 47. The balance between the number of new tags that will actually be used in the real world, versus the amount of controversy generated, has not historically been encouraging. However, that’s just my opinion and others may feel differently.

> Should we make some comparisons for vernacular orthography/ies in
> different languages with IPA?

I don’t feel (again, my opinion) that IPA is the best analogy for this, or anything.

IPA was invented for specialists like teachers and lexicographers, as an all-lowercase notational system. Only later was it adapted as a Latin-script orthography for African and other languages, requiring the addition of some uppercase letters which do not exist as part of IPA proper. I don’t see the relationship to an orthography like Pe̍h-ōe-jī, which among other things uses the more common approach of diacritical marks, rather than the approach of wholly different letters taken by IPA and derivatives.

In any event, IPA-influenced orthographies for African languages are not normally indicated specially in BCP 47 — nor is the Latin script itself, if that is standard for the language.

--
Doug Ewell, CC, ALB | Lakewood, CO, US | ewellic.org