OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

geolang-comment message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


Subject: [geolang-comment] ["Rebecca S. Guenther" <rgue@loc.gov>] Re: Which ismore stable, T or B codes?



I sent an email with some questions to the ISO 639 Maintenance Agency,
and got an interesting reply. See below.
--LMG



--- Begin Message ---
I will give you some quick answers without having looked at the sources
you mention below, and will then review your information.

On 5 Jan 2003, Lars Marius Garshol wrote:
> 
> Hello,
> 
> I am currently chair of the OASIS GeoLang Technical Committee[1],
> which is working on creating a set of URIs for identifying languages
> based on ISO 639-2. Our intent is that these URIs will be used in
> topic map and RDF applications to identify languages uniquely. These
> URIs stand a good chance of becoming an important part of the future
> Semantic Web, in other words.
> 
> We currently have a first rough draft of our specification[2], and I
> am right now working on creating a new draft, and in this regard I
> have a question. I am wondering which of the code sets is the most
> stable. The bibliographic codes, or the terminological ones?
> 
> In the 639-2 standard I found this text:
> 
>   To accommodate large applications that build continuously, the codes
>   in ISO 639-2/B shall not be changed if a language name or its
>   abbreviation are changed.
> 
> which seems to indicate that the bibliographic codes are the most
> stable, but it is hardly conclusive. I've also done a thorough scan of
> your web pages, without finding any answer.

There is a long history behind this situation. I will try to explain that.
When ISO 639-2 work began it was because of the need for a more extensible
list than what ISO 639-1 provided (at the time it was called ISO
639). Since that standard was for a 2-character code, there were only so
many combinations of letters possible, so it was felt that a 3-character
list was needed to accommodate more languages. Such a list already existed
and had been used in the library world since 1968; that standard was a
NISO standard Z39.53 and also used in the MARC formats. ISO 639-2 was a
joint effort between TC46 (Information and Documentation) and TC37
(Terminology). TC37 wanted codes that were somewhat consistent with those
used in the 2-character code list. (e.g. "de" for German should include
those letters in the 3-character code), while TC46 participants felt that
already defined codes needed to remain stable because of their widespread
use (and they were mostly based on the English forms of names). Members of
TC46 agreed to change codes for lesser-known languages when the impact was
not too huge (not too many records), but for some 22 languages, changing
the English based codes would have had too large of an impact in the large
systems that used them. Thus, alternative codes were established, where
639-2/T codes and 639-2/B codes differ. It is important to remember that
it is only 22 of the 450+ languages that differ. All the other 639-2 codes
are the same. We try to think of them as synonyms.

The sentence you cite above reflects the view of the bibliographic world
about always keeping the codes the same for the same entities, because of
the large number of bibliographic records that exist using the codes. 

I don't think you could say that one or the other is more stable. But it
is certainly the case that if you use only 3 character language codes that
639-2/B has had more widespread use. In fact I'm not sure how much use
639-2/T has (remember again, only 22 codes are different), because now
people seem to use the RFC 3066 that says to use a 2-character code if it
is available and a 3-character one if it doesn't exist. So I don't see
that /T has been much used. 

We certainly are committed to keeping 639-2/B stable.

> The answer is quite important for us, as it is imperative that we
> choose the most stable form of the codes in order to avoid unnecessary
> changes to the URI set.
> 
> 
> In addition to this question, now that I have gone to the trouble to
> contact you I would be interested to hear whether you, as registration
> authority, would be interested in maintaining such a URI set on your
> own? I don't necessarily need an authoritative response to this
> straight away, but it would be interesting to get some impression of
> your opinion on this.

It is actually the case that I am involved in some metadata efforts that
involve assigning URIs to elements in our metadata scheme. That may
include assigning them to values in controlled lists, such as the language
codes. However, the agreement is to do this for the MARC metadata
elements, which means we would may only take on that burden for the B
(bibliographic) codes. (Perhaps it would not be too burdensome to do all
of them; we would have to investigate). So this is quite possible since
we've already committed to doing some of this (if you're interested I
could cite for you the effort bringing this about).

In the meanwhile, I'll take a look at the documents you mention below to
see if I have any further thoughts.

Rebecca
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
^^  Rebecca S. Guenther                                   ^^
^^  Chair, ISO 639-2 Maintenance Agency                   ^^
^^  Senior Networking and Standards Specialist            ^^
^^  Library of Congress                                   ^^
^^  Washington, DC 20540-4402                             ^^
^^  (202) 707-5092 (voice)    (202) 707-0115 (FAX)        ^^
^^  rgue@loc.gov                                          ^^
^^                                                        ^^
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

> Thank you!
> 
> [1] <URL: http://www.oasis-open.org/committees/geolang/index.shtml >
> [2] <URL: http://psi.oasis-open.org/geolang/iso639/ >
> 
> -- 
> Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
> GSM: +47 98 21 55 50                  <URL: http://www.garshol.priv.no >
> 
> 




--- End Message ---


-- 
Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
GSM: +47 98 21 55 50                  <URL: http://www.garshol.priv.no >


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


Powered by eList eXpress LLC