(iso639.551) Many languages - how to code?

John Clews Scripts2@sesame.demon.co.uk
Sun, 05 May 2002 22:09:28 GMT

Dear John [Cowan]

Actually, there is still a problem.

Earlier, John Clews wrote:

> > Currently we could face the possible scenario that there could be
> > three different schemas in operation: SIL codes, ISO codes, and IETF
> > tags for the same languages.

In message <200205042130.XAA80295@dkuug.dk> John Cowan relied:

> IETF tags are explicitly derived from, and subordinated to, ISO ones;
> any existing IETF code that is synonymous with a newly created ISO one
> is swiftly deprecated by our ever-vigilant Michael Everson.

But that only covers 2/3 of the problem. It doesn't rule out the
following scenario (1, 3 and 4) with hypothetical codes shown in
square brackets [like this], and with events happening in this order:

1. A SIL code [ABC] for language X already exists - used by many
   users, over several decades, though not in the context of
   RFC 3066, obviously. RFC 3066 specifies how codes from ISO 639 and
   ISO 639-2 might be used in IANA Language Tags, though it does not
   specify how codes from SIL might be used in IANA Language Tags.

2. (If it did do, or if a successor to RFC 3066 did, there would be no
   problem in allocating IANA Language Tags for all the codes
   requested for North American langauges).

3. A requirement for a IANA tag for language X is made. No ISO 639
   code exists, and Michael Everson, as IANA Language Tag Reviewer
   allocates a yet different code [i-def]. This might be because
   there is a more urgent need for this code (indeed perhaps a set of
   codes, as in the recent request for a great many North American
   languages) than is likely to emerge in sufficient time from the
   deliberations of the ISO 639 Joint Advisory Committee.

4. the ISO 639 Joint Advisory Committee hasn't allocated any code for
   language X, but later decides that it would be a good idea to have
   a code for language X. It doesn't like either of the codes
   allocated in 2. or 3. above, for various reasons, so it allocates
   a different code [ghi].

So the possibility certainly exists of three codes, and three code
systems existing for many languages.

Deprecation by the standards developer certainly doesn't stop
deprecated codes being used, and of course there may be various
consequences as a result.

Therefore, it would be useful - indeed essential - for the three
groups (SIL, ISO and IETF) that allocate codes or tags, independently,
and/or in conjunction with one of the other groups, to get together
and provide a simple solution.

Dealing with situation 2. above in some useful way strikes me as the
most cost effective area where time spent might produce some useful

Best regards

John Clews

John Clews,
Keytempo Limited (Information Management),
8 Avenue Rd, Harrogate, HG2 7PG
Email: Scripts@sesame.demon.co.uk
tel: +44 1423 888 432;

Committee Member of ISO/IEC/JTC1/SC22/WG20: Internationalization;
Committee Member of ISO/TC37/SC2/WG1: Language Codes