Adding variant subtags 'aluku' and 'nduyka' and 'pamaka' for dialects

Doug Ewell doug at
Sun Aug 23 18:50:44 CEST 2009

As Pascal and others have said, there seems to be no question that these 
three variant subtags should be registered.  The devil seems to be in 
the following details:

1.  Given that two Description fields are proposed for two of these 
records, should the Comments field mention that these are two different 
names for the same dialect, and not two different dialects?  Peter feels 
there should be such a comment, because the names refer to "dialect" and 
(unlike "Dutch" and "Flemish") are unfamiliar to many users, causing 
potential confusion.  Addison seems to agree.  I feel that the comment 
is not necessary because a single record always refers to a single 
entity; the Registry contains no examples to the contrary.

2.  Should the name "Busi Nenge Tongo" (or alternative) for the base 
language be used instead of "Aukan," which is also a name for one of the 
dialects?  Addison suggested using "Aukan" for the base language in a 
proposed comment; Pascal objected strongly because this name is more 
commonly associated with the dialect than with the base language, and 
would imply the wrong language-dialect hierarchy.

3.  Related to issue 2, should the name "Busi Nenge Tongo" (or 
alternative) be proposed as an additional Description field for language 
subtag 'djk'?  This has been mentioned in the content of ISO 639-3, but 
as Addison has pointed out, an additional description could be added to 
the Registry regardless of what ISO 639-3 does, as long as we also keep 
the one that ISO 639-3 uses (for cross-referencing purposes).  There is 
currently no formal proposal to do this.

There may be other issues which I've forgotten.

It might be worth noting, for comparison, that the Registry includes a 
language subtag 'zza' which has six Description fields, among which are 
"Dimli" and "Kirmanjki," which are also the names of individual 
languages that have 'zza' as their macrolanguage.  Fortunately, ISO 
639-3 added the "(macrolanguage)" and "(individual language)" to their 
name fields as appropriate, so the confusion is not total.  Does this 
establish any sort of precedent for how we might want to handle the name 
"Aukan"?  These languages and names are also probably unfamiliar to many 
users, so the analogy might be a legitimate one.

Pascal wrote:

> "Boni" and "Aukan" [resp.] are alternate names, a little bit 
> outfashioned these last decades because they actually are more ethnies 
> names than language names.

If these name really have fallen into disuse, is it desirable to include 
them as second Description fields, or should they *only* appear in 
Comments fields?  That might solve issue 1 above.  Remember that many 
applications will ignore the Comments fields, so users would not see 
these names unless they dig very deeply.

If we use the name "Busi Nenge Tongo" (or alternative) anywhere at all, 
Pascal needs to provide us with a single definitive spelling.  I have 
seen various posts which used the spelling "Bushi" and which omitted the 
word "Tongo."

As Addison points out, we are now under the RFC 4646bis rules, so the 
review period needs to be extended for at least another week after 
updated records and registration forms are posted to this list.  Also, 
if it is proposed to add a second Description field to 'djk', we need a 
two-week review period for that record and registration form.  I don't 
see a problem with taking this additional time, since the discussion 
shows that there is controversy, and it should be more important to get 
this right than to rush it through.  (After all, these requests had to 
wait for seven months until BCP 47 was able to support 'djk'!)

Doug Ewell  *  Thornton, Colorado, USA  *  RFC 4645  *  UTN #14  ˆ

More information about the Ietf-languages mailing list