Message-Id: <6.2.1.2.2.20050814165223.0307be00@mail.jefsey.com>
Date: Sun, 14 Aug 2005 16:54:06 +0200
To: ltru@ietf.org
From: r&d afrac <rd@afrac.org>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Cc: 
Subject: [Ltru] non-script support
Precedence: list
Sender: ltru-bounces@lists.ietf.org
Errors-To: ltru-bounces@lists.ietf.org

Dear all,
It is interesting to note that this WG claims to deliver general purpose 
language tags. Several times members underlined that it did not only 
support the printing industry but every language related issue, whatever 
the medium/media (except the odd restriction concerning artificial and 
machine languages - what will be a major a problem since the WG does not 
want to define what is a language in a distributed network context (this 
leaves specialists perplex). Is a "computer script" part of human language 
(architexts are computer languages, HTML and XML are computer languages)? 
Is a "Pink Minitel" automated program using a human (at least an adult) 
language? I also regret it does not want to be bilingual: French uses two 
different terms ("langue" and "language"] what certainly helps the analysis.

However the Draft ABNF says:

range         = (language
                  ["-" script]
                  ["-" region]
                  *("-" variant)
                  *("-" extension)
                  ["-" privateuse])

"script" here is confusing for the average layman, if it also to mean 
voice, bel canto, alarms, content mix.  All the more than all this fuss 
about langtags (rather than voiding illegal RFC 3066) is for XML (not 
CLDRs, says Mark Davis; not official W3C says Addison Philips) while the 
W3C calls for positions on internationalizing the Speech Synthesis Markup 
Language (SSML). The topics under consideration seem to belong this debate, 
as the very equivalent for non written language usage to what script is for 
a written usage: pronunciation, volume, pitch and rate among others like 
sex, accent, age, speed, style, referent, etc.

If Members are interested the can read about the W3C effort at:
http://www.w3.org/2005/08/SSML/ssml-workshop-cfp
http://www.w3.org/2003/08/Workshops/
http://www.w3.org/Voice/

The WG-ltru Draft is only written by individuals not representing the 
position of the W3C (however it does not include a disclaimer like the K/H 
RFC). Key non-Member SDOs such MPEG, ITU which are concerned or influencing 
the international network technology, and key non-Members concerned 
organisations such as UNESCO, MINC, ISO, GAC,are not represented in this 
debate or their members ignored. Considering this, I think it would be 
inappropriate at this stage to impose the premature load of reviewing an 
incomplete and contested Draft , even in helping with its document 
write-up, to the IESG. At least,

- the Draft should also documents the support of the voice subtags in 
parallel to script subtags, and create an ad-hoc registry until ISO or 
another SDO can come with an acceptable registry.
- there is a position request sent to the W3C/IETF joint committee.
- there is a coordination exchange with ITU, MPEG, UNESCO, MINC, ISO, GAC, 
ccTLDs, ICANN, as the current most concerned entities I can think off.

I note  that, should this Draft be accepted and confirmed in appeal by the 
IESG, my appeal to the IAB will include among many other things, a call for 
an IAB guidance over the missing analysis and descriptions of what is a 
text, a script, a language, a mode, a referent, a context in a computer 
network environment [which is essentially differ from printed and past 
centuries document by usage, by technology, type of exchanges and relations 
and by the functions of each of these elements in the communication]. More 
generally guidance will also be called concerning the exact bearing 
(identification, definition, authentication, negotiation?) of the proposed 
langtags, the required competence and authoritativeness to discuss their 
registration, the management of their possible conflicts with other 
elements (parameters, names, tags, ccTLD, voice descriptions, icons, 
charsets, etc.), etc.

I am glad the W3C approaches these questions seriously. But the W3C is an 
application oriented consortium, like several others: they need to rely on 
a stable and open network architecture basis, the IETF is to provide. I am 
afraid that IETF and IAB are not supporting their effort adequately at that 
network architecture level. We must provide the building blocks (protocols, 
concepts, tools, elements, etc.) they will build upon. I am afraid the 
script oriented Draft and its IANA Registry Management is only a small part 
of their whole need to describe and usefully support lingual channels.

This calls for a global consensual (the whole intergovernance of the world 
digital ecosystem) framework, for the [at least 5 main subtags plus date] I 
ask, for the analysis this WG refused to carry through a serious reading of 
the Charter (which should be submitted to an IAB review since the WG-ltru 
did not prompt by itself the amendment and extensions this Charter 
require), for an ISO 11179 approach permitting the extensibility and the 
scalability the users need, whatever the ISO inter TCs feuds which are of 
no interest to IETF.

The Draft is now probably mature in its ABNF part (except its attempts at 
format exclusion). The corresponding filter should be finalised. This 
should satisfy the librarians, the political censors/police (cf. security 
section), the corporate strategists and some developers. I doubt however 
this theoretical simplification of the lingual reality will match the needs 
of the users, of the search engines, of the sales people (they want to 
satisfy their customers, not to curb them to unrealistic standards). So, 
the authors have completed the work they were assigned by the Chairs. The 
WG should now initiate the real work required by the IESG and expected from 
the expertise it claims to gather.

For that we only need to make sure the Draft only claims to complete RFC 
3066, and to jointly write a consensual BCP 47 Multilingual Internet 
framework which will support it - and the many other more pragmatic 
propositions we can expect.
jfc 


_______________________________________________
Ltru mailing list
Ltru@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/ltru