Message-Id: <6.1.2.0.2.20050413144215.0de1e0c0@mail.jefsey.com>
Date: Wed, 13 Apr 2005 15:27:09 +0200
To: "L.Gillam" <L.Gillam@surrey.ac.uk>
From: "JFC (Jefsey) Morfin" <jefsey@jefsey.com>
Subject: RE: [Ltru] Great Script Debate Part II: Formats...
In-Reply-To: <4A7C6FA2AB31194E80E13FE585F6A21292A38C@EVS-EC1-NODE1.surre
	y.ac.uk>
References: <4A7C6FA2AB31194E80E13FE585F6A21292A38C@EVS-EC1-NODE1.surrey.ac.uk>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Cc: ltru Working Group <ltru@ietf.org>
Precedence: list
Sender: ltru-bounces@lists.ietf.org
Errors-To: ltru-bounces@lists.ietf.org

Dear L.Guillam,
this certainly is a right of everyone - and a possibility and a certain 
work. This is the same as for 10646: building a list of name of usage 
correcting or adapting the 100646 names to the reality of usage. And be 
sure this is exactly what we are working on as a prototype CRC.

What I oppose in "your" position, is four things: dominance, lack of QA, 
complexity, inconsistency, decency. Since you certainly do not see where 
let me explain again.

- dominance is not good because it is not appropriate to the network 
architecture. Since Paul Bahan in the 60s and the work carried in the 80s 
we know that networks can be centralized by a dominance, decentralised by a 
governance, distributed by an intergovernance. These are natural 
interest/forces/solutions which are the very architecture of every network. 
They mostly result of their granularity, which means for wide systems of 
their density. Today the internet itself which started in a centralised 
development phase (mid 80s) to a decentralised deployment phase (mid 90s) 
enter its distributed scalability phase. Dominance concepts (IANA, ICANN, 
ASCII, English) were until now acceptable as only "old". They become 
progressively totally outdated and an architectural flaw more and more 
people and concept will oppose and already suffer from. The idea of a 
reviewer in English for the language of the world, was a bizarre oddity, it 
is becoming an unacceptable bug.

- lack of QA results from abandon of cross culture control and protection. 
Look at the ieft-languages list and at this list. How many people of 
English/non English culture/mother tongue when we are precisely discussing 
languages and cultures. Look at the number of represented nationalities 
when you know (and this is what is precisely discussed and acknowledged 
through the ISO 3166 country codes) a language is the first and basic 
social contract of a nation. I am not interested here in politics but in 
quality. Look at the number of wrong 10646 names in spite of all the 
accumulated efforts. Look at the ietf-languages lists which doubled the 
considered Greek languages tags when I called upon the Greek ccTLD and the 
University of Athens. There is a need for cross cultural control. This 
control is provided by French for centuries and at ISO. Be sure I will not 
accept that all of us we lose that QA.

- complexity. The hook of a language is not an easy thing because 
printf("$stringNN") is not a programer reflex. In permitting to have a 
single reference name, you do not oblige programers to support several 
languages (when there are two, there are hooks for all).

- inconsistency. We know that the reference names are just because France 
accepted not to delay ISO 639-3 (I concurred) in having not the 
budget/people to carry the work. The response was "we will use autonyms". 
Autonyms would be nice. But "fra french" is not really an autonym. Now, the 
problem is may be something we could fix. We have to discuss.

- decency. A Draft has been introduced twice, and failed. We (ietf@ietf.org 
opponents) proposed a small rewrite of two or three point or a real WG. A 
WG and a charter have resulted, which uses the same Draft. And while the 
Draft is not even submitted to the IESG the Draft is enforced. This is 
playing a lot on a very well known idea: people today cannot be internet 
builder anymore, so they are afraid of being remembered as Internet 
wreckers, and they do not dare to oppose. It happens that I do not fear 
that: my past is behind me already. But I do not want to be remembered as 
the one "who could have" ...  As pointed out in another mail, the problem 
here is that it is a call to extremism. Everyone serious outside of this 
long working together group, and interested (but IAB, IESG and IETF is not 
really interested in multilingualism), sees that the Draft is a clumsy 
attempt at proposing an XML patch (Addison well documented it) and will 
oppose it to be made a Multilingual Internet confusion. In the process 
these "everyone" will most likely fight and forget Addison's legitimate 
need for solution to the real, urgent and important (but not fundamental) 
need of an XML patch.

all the best.
jfc

PS. Since you are going to tell me "and why IAB, IESG and IETF is not 
interested ???", what result from observation (for example the very odd RFC 
3869 which is a call of IAB to Governments for Internet R&D financing and 
does not even allude to languages after 192 unanimous Governments made it 
their priority and it is the US year of multilingualism). I started a draft 
last summer on the way the Internet standard process could support the 
documentation of a Multilingual Internet. It was interrupted by the RFC 
3066 bis saga. From this I think it is the syndrome I quoted above and Vint 
made his leit-motive "why to fix what ain't broke". The architectural 
change is huge. The problem is that the question is not "to fix what would 
be broken", but "to replace what is outdated".

PS. In case you ask what the hell can be a "Multilingual Internet", I 
suggest to use http://minc.org . You may not know it, but it terms of users 
it is a substantial one. Not so far from the majority if not already the 
majority. But the growth of the ASCII Internet is so slow (800 millions 
against 1.4 billion mobiles). Vernacular is not supported. Architecture is 
too rigid.



  13:22 13/04/2005, L.Gillam said:

>Jefsey,
>
>For once I find I might be able agree with you - I might
>have misunderstood you however, so please don't comment
>it as a certainty.
>
>Reference name provides the "hook", as you refer to it.
>You (2nd person singular/plural) can provide whatever additional
>translations you (2nd person singular/plural) desire for such
>reference names, in whatever languages you (..) desire,
>keep whatever local copies you (..) wish, and so on. The
>problem you are making for people on this list is that
>you (2nd person singular) seem to be insisting that the registry
>MUST contain names in EVERY language for EVERY language, or should
>produce translations where they are not available from the source
>standards (e.g. reference names in 639-3). I see little purpose -
>we are taking about "names" that help to identify tags, not
>imposing restrictions on "languages", per se, which would
>be somewhat different.
>
>Nobody is prevented from creating French names for these
>7000 reference names, German names, Slovakian names, and
>so on. Making it a necessary and essential task is
>unrealistic - look at the army of translators needed by
>the EU, for example.
>
>As I said, feel free to create your own hooks in other
>languages of your (2nd s/p) choice for your (2nd s/p)
>presentation - provided that in interchange the presented
>item transforms into the correct tag there will be no
>difficulties.
>
> > -----Original Message-----
> > From: JFC (Jefsey) Morfin [mailto:jefsey@jefsey.com]
> > Sent: 12 April 2005 17:48
> > To: Gillam L Dr (Computing); ltru
> > Subject: RE: [Ltru] Great Script Debate Part II: Formats...
> >
> >
> > I think you are globally right.
> >
> > At 16:50 12/04/2005, L.Gillam wrote:
> > Where did I make something forbidden Jefsey?
> >
> > Ah! I thought you wanted to forbid French and hooks for other
> > languages.
> >
> > On the contrary,
> > >I think you'll find I say there are certain things we should
> > >not impose (just above here for example).
> >
> > This implies that default references should be autonyms? I was wrong
> > reading you differently.
> >
> > >Quite the reverse. Please do not twist what I, or indeed
> > others, write in
> > >an attempt to validate your position.
> >
> > I have no position. Just observation. When I say that the
> > lack of support
> > of English and French is a blocking element. This is because
> > I observe that
> > it will be real world blocking, because applications are
> > under current
> > development and projects under financing which say otherwise.
> >
> > >There's a "forbid" if you want one.
> >
> > How rigth you are.
> > jfc
> >
> > PS. I suppose you will want to respond I a wrong or
> > unintelligible. No
> > need, I have already taken your response into account.
> >
> >
>
>_______________________________________________
>Ltru mailing list
>Ltru@lists.ietf.org
>https://www1.ietf.org/mailman/listinfo/ltru


_______________________________________________
Ltru mailing list
Ltru@lists.ietf.org
https://www1.ietf.org/mailman/listinfo/ltru