[Ltru] Re: "mis" update review request

Peter Constable petercon at microsoft.com
Sat Apr 14 04:40:31 CEST 2007


From: John Cowan [mailto:cowan at ccil.org]


> I think Mark's point was that "kind" is ambiguously English or Dutch...

In a case like "kind" where there's ambiguity and you don't know the specific language, the appropriate tag would be und.


> The normative text of ISO 639-2 says (section 4.1.2):

        The language code mul (for multiple languages) should be applied
        when several languages are used and it is not practical to
        specify all the appropriate language codes.

> I think that is meant to cover mixed-language documents, not documents
> whose language is ambiguous

Correct.


>> But that's a different question from how (or if) people should use
>> mis. IMO we should say that implementers of BCP 47 SHOULD NOT use mis
>> except if they have an immediate need to apply a language subtag and
>> have determined that there is no available language subtag encompassing
>> the language of the given content. E.g. I know it's Martian and there's
>> no ISO 639 ID for Martian. (Hypothetical example assumes Martians
>> exist and that this is not encompassed within art.)
>
> Or, until 639-3 codes are available in BCP 47, any language that doesn't
> have its own 639-2 code or plausible covering collection.

Sure; the determination can be made based on what's available at the time the determination is being made. That presents a stability issue, but I don't see how that can be avoided if you want to use mis. The other option is to simply use und on the assumption that the conclusion being made is "I don't know how I can possibly tag it using the available tags". That's stable, though it is making a different statement.


Peter


More information about the Ietf-languages mailing list