Re: Armenian (was: Re: ISO 639-3 changes, part 3)

Strong agreement. (I think it were better had collection IDs never been permitted.)


Peter

From: Ietf-languages <[email protected]> On Behalf Of John Cowan
Sent: Thursday, March 8, 2018 7:55 PM
To: ietflang IETF Languages Discussion <[email protected]>
Subject: Re: [Ietf-languages] Armenian (was: Re: ISO 639-3 changes, part 3)

Yes, I was wrong to say that 639-5 codes are not usable.  However, they are certainly not advisable.

On Thu, Mar 8, 2018 at 9:16 PM, Arthur Reutenauer <[email protected]<mailto:[email protected]>> wrote:
> 'hyx' is not a language code (it represents the Armenic language family,
> just as 'gem' represents the Germanic language family), and like other
> 639-5 codes, it cannot be used as a BCP 47 language tag.

  Really?  I can’t find anything about it in RFC 5646; -5 code elements
are mentioned explicitly as primary language subtags in section 2.2.1.
They can’t be extlangs, though.  In any case, [hyx] is present in the
language subtag registry, with the description “Armenian (family)”.

>                                                           If the RA decides
> to create a macrolanguage code for Armenian (macrolanguage), that's another
> matter, but they haven't done so.

  (Michael seems to have pre-empted their decision, by the way.)

        Best,

                Arthur

_______________________________________________
Ietf-languages mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ietf-languages