recent meeting of ISO 639/RA JAC

[jcowan at reutershealth.com]

Michael Everson scripsit:

> It would be appropriate to summarize on this forum.

The text is short enough to just copy, I think:

Report on meeting of ISO 639/RA Joint Advisory Committee, Jan. 15-16, 2004

Peter Constable

This meeting was held at Library of Congress in Washington, D.C. The
purpose was (i) to deal with routine management issues for ISO
639-1/-2, and (ii) to make decisions relating to future directions
in the development of ISO 639, including ISO 639-3. The meeting was
attended by six members of the Joint Advisory Committee (Havard Hjulstad,
Gerhard Budin and Christian Galinski from TC 37; Rebecca Guenther, Glenn
Patton and Anila Angjeli from TC 46), two representatives from SIL,
the designated RA for ISO 639-3 (Gary Simons and Paul Lewis), and me as
project editor for ISO 639-3.

This meeting went very well)B—much better than I had anticipated. For
instance, I had expected there might be difficulty on some issues related
to development of ISO 639-3, particularly from the representatives of
TC 46 (the librarians), who in the past have shown some reluctance to
consider changes from MARC practices. There was very much a sense of
cooperation and collaboration, however, and a level of trust that has
not existed in the past.

Progress on ISO 639-3

As project editor for ISO 639-3, my concerns were two-fold:

- to establish consensus on an appropriate model for maintenance of
ISO 639-3, since the current practices for ISO 639-1/-2 in which all
decisions are made by the JAC would not scale well

- to get a commitment from the JAC to resolve quickly several questions
relating to the existing identifiers in ISO 639-1/-2

Both of these were achieved. (The JAC will be dealing with the open
issues in the existing code tables in the coming weeks; the process will
involve circulation of a proposal on resolution of the issues with a
deadline for comments and balloting only on individual points as needed,
so closure within a specific time frame is ensured.)

As a result, I am very encouraged about progress on ISO 639-3. The
result of the CD ballot that was completed last month was a vote in
favour of advancing to DIS. There were various comments submitted, and
the SC chair has approved my proposed disposition of those comments. A
new draft will be circulated for balloting next month, in time for
the ballot to be completed prior to the TC 37 meetings in Paris next
August. An optimistic schedule, if there are no major comments on that
draft, is that an FDIS ballot would be circulated this fall.

Other developments in ISO 639

There was also discussion on other future parts to ISO 639. Parts 4
and 5 are in progress; part 4 will provide an overall framework for
language identification; part 5 will provide identifiers for language
collections (an extension of ISO 639-2, in a sense, much like ISO 639-3
is an extension of ISO 639-2 for non-collections). The UK is planning
to submit a new work item proposal for a part 6, which (in rough terms)
would provide identifiers for dialects, but it is not yet clear what
the results of that will be.

The overall picture for the future of ISO 639 is that there will be an
overall three-letter codespace, and an overall set of entities defined
containing languages and collections of languages; the two types will
be divided between parts 3 and 5, and eventually may get merged into a
single standard.

Higher-level protocols and implementations will determine which members
of the total set are to be used in a given application context. ISO
639-2 will become a profile of the overall set of entities (and of the
three-letter codespace), a subset deemed useful for certain application
contexts (with general-purpose libraries such as LOC likely continuing to
be a primary intended usage context). ISO 639-1 will be another profile of
that set of entities intended for yet different usage scenarios, though
it will make use of two-letter identifiers. Each of these application
profiles will maintain their own criteria for determining which members
can be included (as is currently the case). Additional usage profiles
may be definable in the future, possibly as parts to ISO 639 or through
some form of registration process.

ISO TC37 and the ISO 639/RA Joint Advisory Committee are very concerned
about meeting the needs of different user communities in the area of
language identification.  They would like to see a greater level of
interaction from various sectors, and would welcome liaison relationships
from bodies representing different groups of stakeholders.

-- 
Deshil Holles eamus.  Deshil Holles eamus.  Deshil Holles eamus.
Send us, bright one, light one, Horhorn, quickening, and wombfruit. (3x)
Hoopsa, boyaboy, hoopsa!  Hoopsa, boyaboy, hoopsa!  Hoopsa, boyaboy, hoopsa!
  -- Joyce, _Ulysses_, "Oxen of the Sun"       jcowan at reutershealth.com