New item in ISO 639-2 - Zaza

Doug Ewell dewell at
Wed Aug 23 16:15:09 CEST 2006

Peter Constable <petercon at microsoft dot com> wrote:

> Before jumping to conclusions, you need to remember that all you have 
> seen of the code table for ISO 639-3 is *draft*.

That's true, but it's been relatively stable for over a year, and with 
the publication of 639-3 reportedly just a few months away, it's not 
exactly a tabula rasa.  It's been through some review.

>> I guess we will add a subtag corresponding to this new 639-2 code 
>> element,
> Is there a choice at present?

Not really.  Section 3.3 says the Reviewer must "determine whether [a 
change to ISO 639] conflicts with existing registry entries," which it 
does not.  If the Registry included the 639-3-based subtags, which is 
planned for 3066ter, then the situation might be different.

> There will be indeed a change in the 639-3 code tables: macrolanguage 
> mappings from zza to dlq and kiu will be added.

OK, so the right thing to do in this case is add "zza" (as we must) and 
ignore the 639-3 situation for now.  By the time we have to care, for 
3066ter purposes, everything will be made right in 639-3.

> This is not a "mismatch" as you had in mind. It is a new 
> macro-language entity for which the component members were already in 
> 639-3. That's a possible scenario that will have to be considered in 
> developing 3066ter, but I don't think it is a particularly complicated 
> problem to deal with.

But the macrolanguage in 639-3 doesn't just create itself, does it? 
Suppose -- this is my whole point really -- suppose this happened in the 
3066ter era.  The Registry would contain primary language subtags "diq" 
and "kiu".  Now suppose 639-2 adds "zza" and we need to add it as well. 
We would have to reclassify "diq" and "kiu" from primary to extended, 

Actually, it may be a blessing in disguise that this use case occurred 
before we start working on 3066ter.

Doug Ewell
Fullerton, California, USA

More information about the Ietf-languages mailing list