Assignment of private use language subtags

Anthony Aristar anthony at aristar.org
Fri May 9 21:42:48 CEST 2014


As the person who ran MultiTree until last year when I retired, let me 
make some comments on these codes:

1.  The fact that MultiTree uses a local use code should not be taken as 
evidence that the MultiTree project believes that the language exists.  
The system it uses requires that every language entered into its system 
have a unique 3-character code, and since the project is designed to 
collect all hypotheses proposed about language relationships, even those 
which are wrong, it sometimes assigns a code to a language that does not 
and has never existed.

2.  Some of the codes listed on the Wikipedia page are out of date even 
on the MultiTree site, and are either not in use any more, or are going 
to be removed shortly.

3.  The local use codes collected on the Wikipedia page are only a small 
subset of the codes MultiTree uses.  The local use space was found to be 
nowhere extensive enough for all the languages which had to be coded, so 
most of the codes it now assigns are alphanumeric.  So, for example, 
Puruha was assigned the code qq3, and Old Portuguese the code 079.  It 
might be a good thing to get an updated list of all these codes at some 
point.



On 5/8/2014 10:01 AM, Doug Ewell wrote:
> Gordon P. Hemsley <me at gphemsley dot org> wrote:
>
>> I was just informed about the existence of this mapping of private use
>> language subtags to languages, maintained by LINGUIST List and used by
>> MultiTree:
>>
>> https://en.wikipedia.org/wiki/List_of_ISO_639-3_language_codes_reserved_for_local_use
> Thanks for the link. I hadn't seen it before, and still can't figure out
> how to get this list directly from Linguist List or MultiTree.
>
>> Does anyone have any opinions as to the importance of this list, and
>> how much others should be concerned about clashing with these
>> assignments on the wire? (I know there are similar lists maintained
>> for the Unicode private use area that have some importance for
>> unencoded script users.)
>>
>> How safe is it to ignore these assignments?
> I agree with the comparison to the ConScript Unicode Registry
> (http://www.evertype.com/standards/csur/) and similar Unicode PUA
> allocations. BCP 47 users might do well to honor or avoid (as
> appropriate) these private assignments -- particularly if they gain
> traction outside of Linguist List -- but they are under no obligation to
> do so, and BCP 47 should not formalize this arrangement.
>
> The quality of the list is debatable, and discussion of that is left for
> another message (maybe). That said, since it appears in a Wikipedia page
> (maintained almost exclusively by one person), it would be a welcome
> improvement if Wikipedia authors and editors were to move toward using
> code elements from this list, compared to making up non-private code
> elements as they go along. The Wikipedia page doesn't specify this as a
> goal, though.
>
> --
> Doug Ewell | Thornton, CO, USA
> http://ewellic.org | @DougEwell
>
> _______________________________________________
> Ietf-languages mailing list
> Ietf-languages at alvestrand.no
> http://www.alvestrand.no/mailman/listinfo/ietf-languages



More information about the Ietf-languages mailing list