Stability of valid IDN labels

[JFC Morfin]

At 20:50 20/04/2008, John C Klensin wrote:
>More important, there is a discussion about IDNs that Jefsey and I 
>have had many times.  Often we seem to be in complete agreement; 
>sometimes I get notes that appear to indicate that we are not.  I 
>find the combination bewildering, to put it mildly, because I don't 
>see the difference between the discussions in which we agree and 
>those in which we do not.

Dear John,
if we are in agreement and then we are not, it is because there are 
two different problems:
(1) to make the existing Internet architecture to support linguistic 
needs. We are in agreement on limitations.
(2) to provide the existing world digital ecosystem convergence with 
a Multilingual and Semantic Network contininuity that fits the needs 
of the linguistic diversity. There we agree that a presentation layer 
is needed. However, you think we can only stick to the IETF 
internationalisation. While I think that the existing technology (as 
in the ROAD case) is not so limited, and can support more through 
another paradigm (applied vision) which has to be found, documented, 
developped, tested, and dynamically deployed over the 10 years to 
come - and multilinguism is a great opportunity for (it should 
support far more than the language diversity). This is why IMHO the 
problem is not in the 2008 machines/code, but in the brain of the 
designers' community of 2008 who still think the nets as in 1983. 
Brainware is far more difficult to update than software.

We are in a real world, with technical, commercial, and political 
duties. And we have an opposition between:
- a globalization a few develop and profit from and most do not 
understand and will be unhappy with,
- a multilinguisation that most demand but have not understood either 
how it works, as it is a part of the Theory of Everything (we are at 
the man/machine interface)

Today, to introduce my ideas in _competition_ with IDNA would be at 
best delaying and more probably divisive and much counterproductive. 
This has to be in _coopetition_. This is not easy to analyse, 
explain, make accepted, and worked on.

In the RFC 4646 case, there was an urgent need to protect cultures 
and IANA from a Unicode leading consortium English based language 
cataloging. Because, as Aristoteles says "when an idea has started 
flying it will fly for ever along its premises". Premises is where 
one oppose a wrong idea. In this case they were the IETF WG debates. 
We desesperately needed the text to be more precise (to be more 
interoperable) on points authors, ISO TC37 and WLDC people most 
probably still do not identify but that some French school linguists, 
industries, politics, and TC46 Members understand very well. This is 
why ISO documents are supposed to be written in parallel in French 
and English - for a multual pragmatic filtering and a semantic 
complementarity - as we discussed it on the Unicode list.

In the IDNA case the first main pressure are the Beijing Olympic 
Games (where every Chinese athlete will have a Chinese name for his 
site, and the whole non-ASCII world to see it), and in Paris: what 
france at large, ISOC and TC46 people will hear at the ICANN meeting 
while there still are no French IDNs. So, we have a little more time. 
A few French @large lurkers are on this list, and start discovering 
the linguistic architectonical difference between IETF and them. 
However, these are long haul, very long haul and very important 
issues: it is the path to the semantic internet (not the semantic 
web) and most probably the network interoperating system.

I said that my plan was to document an acceptable framework for our 
work, so we would avoid interfering and there would be less risks 
that people confuse what we says and try to oppose it to IDNA. I 
apologize for being late but there are many thing to work on, and 
understand at the same time. Plus local political interferences by 
people not understanding for example why "AFNIC is so slow 
implementing French names". There is a real teaching which is 
missing. Both sides. Also, I am not that much interested in the 
Internet strata which has to work well. I am interested in the 
digital convergence as it - alt last - leads to the semantic 
emergence. I also feel it is better this WG-IDNABIS be well established first.

One of the things the MLTF identified is that multilingualizing the 
name space is not a job for linguists, but for multilinguists. 
Linguists try to analyse languages from the linguistic diversity. 
Multilinguists try to analyse diversity and document its support. 
IDNs are an epiphenomena. To be accepted, a Multilingual Internet has 
to be obvious to people from the whole world, in the same way as it 
is today to American users. Linguists can help a lot, but their job 
is to know human languages, not machine meta and multilanguages 
working in parallel.

Also, we are not so much interested in the engineering details of a 
possibly weak architectural approach. This is why Louis would like to 
see it confirmed. The less IDNA is constrained IRT the real moving 
world (what calls for a lot of thinking), the more chances it has to 
succeed. In addition, I do not think IDNA and an architectural 
evolution oppose. I think there is a full possibility for a 
transition. To understand if this is possible, one has to imagine the 
future (this is not your target) and try/permit to connect it to what 
you work on (interoperability).

I would like not to repeat this. After the PR-action, Debbie Garside 
believed she could win easily against me, because she had not 
understood what I had won with who. When I proposed her several times 
to cooperate, she thought it was technical and political weakness. I 
managed a meeting with ISO etc., her, ICANN and me. I tried to make 
them understand we were not opposing, but we had to work on a 
transition scheme and work proposition from their 
internationalisation to "my" (actually WSIS's) multilinguisation. 
ICANN smartly stayed in its role, but others refused and maintained 
Debbie's NWIP. They lost 3 to the world.

For four years I explain that we will not remove what exists even if 
poor (IMHO) globalization. The hysteresis is very long (more than one 
million nameservers to update). This is why we have to help 
developping better globalization solutions, even if globalization is 
not the eventual solution. However, our priority should be 
transition, i.e. not to block multilinguisation, to the countrary 
working on it and starting deploying some now.

Also, please remember. There is only one hierarchical globalization. 
By nature there are multiple heterrchical multilinguization. Let keep 
things simple, moreover they are complex.

best,
jfc