Re: [Ietf-languages] Latin Sub tags

[Doug Ewell <[e-mail]>]

Hugh Paterson III wrote:

> Applying for variants is currently the method which presents the
> lowest barrier to entry. However, that doesn't mean that the
> acquisition of a variant tag is the most accurate solution or the best
> for the informatics community (or other stakeholders).

Again I must caution Hugh, and anyone else who may be thinking along these lines, not to regard the BCP 47 process as a “lowest barrier to entry” of any sort.

If an entity is considered to be a language, and not a dialect or other language variety, then getting it encoded *as a language* in BCP 47 is a much *higher* bar than getting it encoded in ISO 639. That is because we recognize the ISO 639 RAs as the experts in this field. We do not consider ourselves an alternative or adversary to any of the RAs, and do not normally consider it our job to overrule their judgment about language identification or coding. We have never yet registered a language subtag that has not already passed through the 639 process, and as the RFC says, “it is very unlikely that any subtags will be registered of this type.”

If an entity is NOT considered to be a language, but rather a dialect or orthography or an “early” variety that does not meet the accepted “separate language” criteria, then we can register a variant subtag for it.

Comparing the methodical, periodic ISO 639 process for assigning language code elements with the leaner and faster BCP 47 process for registering variants is like comparing apples and oranges. Languages get language subtags. Variants get variant subtags. We must never consider registering a BCP 47 variant for something that is really a separate language. That is abuse of the system.

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