Re: [Ietf-languages] Latin Sub tags

[Doug Ewell <[e-mail]>]

IMHO, this section needs to be reproduced in a prominent location on the public ISO 639 website. It might help alleviate confusion over the concept of a macrolanguage, or over the difference between macrolanguages and collection codes. The present thread showed one example of such confusion; the 639-3 change request archives show many other such instances.

Of course, it would be best if the entire text of ISO 639:2023 were made publicly available, but I know ISO likes to charge for most standards.

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


From: Sebastian Drude <[e-mail]> 
Sent: Monday, December 4, 2023 14:42
To: Doug Ewell <[e-mail]>; Mark Davis ☕ <[e-mail]>
Cc: Hugh Paterson III <[e-mail]>; IETF Languages Discussion <[e-mail]>
Subject: Re: [Ietf-languages] Latin Sub tags

I remember having read that the "macrolanguage"-scope was invented and applied merely in the context of harmonizing the Ethnologue database with the existing ISO-639-2 codeset when Part 3 was implemented.  So I would be surprised if many new code elements with that scope would have been created since then, and I believe that in the future such requests, while not excluded (see below), would need very good support, especially if they change the meaning of existing code elements and threaten to break existing applications, as Mark rightly pointed out.  And I also agree, the inconsistency (comparing, for instance, ar with de) that Mark points out is worrying, and I do not have any coherent and practicable solution.
Excerpt from the new ISO 639:
6.2.4    Eligibility of a macrolanguage for language coding
To qualify as a macrolanguage, a proposed macrolanguage code element shall meet the following criteria:
• The encompassed individual languages shall be closely related.
• Some applications require that a language be treated like an individual language in public data interchange.
• Other applications require that the same language be subdivided into two or more individual languages in public data interchange.
• The relationship between the macrolanguage code element and the encompassed individual-language code elements are explicitly defined, i.e. there is a defined mapping from the macrolanguage code element to two or more specific individual-language code elements.
• A macrolanguage code element shall not encompass another macrolanguage or language group.
• The creation of new macrolanguage code elements or the widening of an existing individual-language code element to the scope of a macrolanguage shall be based on evidence of application requirements for public data interchange, as well as appropriate linguistic relationships between the individual languages involved.