xn--p1ai

[J-F C. Morfin]

At 18:31 24/02/2011, John C Klensin wrote:
>--On Thursday, February 24, 2011 09:31 -0500 Lyman Chapin
><lyman at interisle.net> wrote:
>
> > Simon,
> >
> > I think that many of us expect the case-folding behavior of
> > IDNA2003 to be preserved by applications (esp. web browsers)
> > at least during the transition to IDNA2008 (e.g., following
> > UTS #46, http://unicode.org/reports/tr46).
>
>And some of us believe that the recommendation of RFC 5895,
>which include mapping to lower case (see its Section 2, item 1),
>are more appropriate than UTS #46.

Let me clarify this also from an IUser point of view, as we initially 
and determinedly oppose case-folding. (NB. An IUser is a member of 
the IUse emerging community who is interested in the people centric 
convergent use of the Internet along with the other (smart or not) 
available bandwidths from the world digital ecosystem).

1. We accepted John Klensin's proposition

During the WG/IDNABis work, we were able to identify that IDNA2008 
casefolding, as and where proposed by John Klensin, permitted a clear 
and stable response for the Internet bandwidth that we could 
typically interface with our own ML-DNS (multi-naming-layer 
language/technology/context independent) concept. Such an ML-DNS, as 
we explore and work on it, could be documented as another RFC 
5895-like example on the user side of the network side IDNA2008.

2. The WG did not care about French

This ML-DNS of us addresses our "Projet.FRA" (http://a-fra.org) needs 
for a Francophone open ontology using its name space as an ontology 
and a semantic network, which I reported to the IESG Last Call. 
<http://tools.ietf.org/id/draft-iucg-afra-reports-00.txt>http://tools.ietf.org/id/draft-iucg-afra-reports-00.txt.

2.1. we reported it to IESG/LC

It emphasizes that IDNA2008 does not support French majuscules (first 
letter in a phrase or in an important word) that Unicode does _not_ 
support either, except as uppercases and, therefore, our obligation 
to build a solution on top of IDNA2008. This is because Unicode 
tables document a mechanical typography, and registrants need to have 
their orthotypography rules (and not tables) supported 
(orthotypography is the syntax of the way one writes a language - 
something that is out of the IDNA2008 and Unicode scope). French 
language speakers unwaveringly demand such a support as well as the 
other Latin language speakers. To explain: "Etat" means 
"State/Government," état" means "status", etc.

That situation calls for:

1. a mapping of everything to lowercase so that everything is stable, 
clear, and secure.
2. us to add the majuscule information as a metadata that we 
investigated in 
<http://tools.ietf.org/id/draft-iucg-punyplus-03.txt>http://tools.ietf.org/id/draft-iucg-punyplus-03.txt. 


2.2. our position is qualified as "research"

The IESG considered our position and our plea for a quick validation 
of the IDNA2008 consensual document set (so that we would know the 
rock to build upon). IESG also considered our proposition as 
"research". Since we have other topics for semantic research on this 
project, we are considering different choices with members of the 
MAAYA network (linguistic diversity) and will introduce ".fra" (along 
with other projects we call "xTLD"s, i.e. experimental TLDs, once 
ICANN finishes its gTLD rigmarole, in order to avoid any unnecessary 
naming conflicts). We will first document a charter on the way to use 
the Internet as its own test bed through an RFC for information on 
"intertesting" in gathering different ICANN and IETF requirements.

3. Strength of IDNA2008 and IDNA is weakness

Another reason why we pleaded for IDNA2008 to be quickly accepted was 
that the AD (once we had completed the WG/LC and IETF/LC) started to 
question, and rightly so, the IDNA principle itself. As RFC 6055 
partly also does: will different applications present the same 
A-label to the DNS? Our ML-DNS vision addresses that concern because 
it extends the DNS and, therefore, utilizes a single Internet DNS Use 
Interface with unchanged applications that are transparent to the 
U-label actually entered by the Users.

3.1. The presentation layer.

This is the role of the presentation layer. "The Presentation Layer 
is responsible for the delivery and formatting of information to the 
application layer for further processing or display."  The Internet 
presentation layer was virtual. It was introduced by i-DNS and by 
IDNA2003 in using the "xx--" ACE prefixes. Yet the idea to dedicate 
it to the sole linguistic names at user application layer (i.e. 
outside of the network area) was only a patch that the ML-DNS (i.e. 
on top of the DNS in the network area) corrects. Applications like 
browsers have nothing to do with domain name massaging, we just want 
them to pass our entries to the domain name management systems 
(plural, as we may or not be using the Internet or the Internet 
technology, and the DNS or not).

3.2. the IUI

However, this introduces the notion of a fringe IUI (Internet Use 
Interface) that faces no problem in becoming an "Intelligent Use 
Interface" with multiple technologies, and to consider an IUDNS 
(intelligent use domain name space) of which the Internet domain 
space is a small yet open part, of which the class IN is 1/65,000-th 
and of which the ICANN root space is a tiny commercial chunk.

4. Technical vs. technico-commercial disagreement

IAB RFC 3869 explains as to why the commercial reasons of 
applications developers, who are also service providers, such as 
Microsoft and Google, and their Unicode consortium, may lead some of 
their employees to plea/lobby against common technical interest, for 
an internet that pays better rather than works better. The same, 
paying services or immaterial goods merchants may prefer a 
client/server architecture tying users to their proprietary offer 
(such as Apple etc., which is another Unicode consortium leading member).

However, some IETF leaders and smart Google people have now swallowed 
and are digesting the optional change introduced by IDNA2008 with:
- the "inside" Internet
- encapsulated into an uncoupled peripheral extended (intelligent) 
services "IUI" (Internet Use Interface), interfacing the user universe,
- itself accessing and operating the Internet and other systems of 
the world digital ecosystem though their own "IUI" (Intelligent Use 
Interface).

5. A strictly conformant architectural enhancement

This change totally conforms with the Internet architectural principles:

5.1. RFC 1958 principle of constant change: the change is dramatic 
but does not require a single bit change in the code. It consists in 
looking at the Internet from the outside rather than from the inside.

5.1. RFC 3439 principle of simplicity: why give huge and unlimited 
DNS responsibility to browsers and applications. Let's keep it 
simple. ASCII just works, let not fix it with complexity (cf. the 
exemple of this thread)

5.3. the RFC 5890 to RFC 5895 set unlimitedly "multiplies by 
division" the power of the Internet in installing an endlessly 
diversified intelligent capacity to match external diversity (such as 
linguistic diversity) where RFC 1958 and all the IETF culture puts 
it: at the fringe. By doing so, it acknowledges that the Internet 
technology matches a third fundamental networking principle, the 
principle of subsidiarity.

  6. The IUse community primary targets

Based on the ML-DNS/IUI principle, we mainly target seven fringe to 
fringe areas:

6.1. active/abient content:

The internet currently only supports passive content (what I receive 
is what was sent). We need active (what I receive is what the sender 
intended me to receive) and ambient (what I receive is what 
corresponds to my current context) contents to be supported.

6.2.smart extended network services:

We want to get local slots supported (smart local operative tasks), 
i.e. local OPES or peers to the remote connection peers able to 
uncouple the client/server interoperations and extend the user's 
experience of the used network's technology.

6.3. semantic networking:

We consider three main communication strata serving

- signal/data basic services,
- content (passive [value-added services], active and ambient 
[extended services]),
- and semiotic/semantic (facilitation services). Semiotic means the 
incorporation of many sources of perception and utterance other than 
script and dumb voice and image. Semantic means the exchange of 
meanings that can be extracted and processed by the facilitation 
services along with the perceived context and the receiving mood and 
options. Semantics' coherence is as precise as geometry's 
correspondance and mathematics equivalence. Their figures only can be 
draft, computed or discussed.

6.4. multilingualisation:

We consider an anthropobotic society where machines and people 
interspeak. This means that several new linguistic disciplines must 
be considered:

6.4.1. "multilinguistic": as the way several languages may coexist in 
use, machines, process, and society, and be treated as equal 
(multilingualisation is a layer above Unicode's globalization in the 
sense that it could be termed as a specific value added equal 
globalization of every natural language). A multilinguistic Internet 
would probably focus on a score of languages, and then 150 main 
languages in an operational community effort, and a standardized 
approach for volunteers and/or local authorities to get other natural 
languages (22.500) identified, documented, and supported.

6.4.2. mecalinguistic: as the way to manage machines' natural 
language and their interinfluence with the language version of the 
language. Example: English and MecaEnglish and MecaFrench.

6.4.3. metalinguistic: up to now metalanguages were rather limited 
depending on the natural language being considered, French being 
probably the most metalanguage embedded. However, facilitation is 
going to call on ontographies and ontologies that will lead to a 
generalization of the mostly French used "metaduction" (what is 
wrongly termed "Cartesian") as a simplification of deduction, 
induction, abduction, and hypothetico-deduction in front of the 
apparent networking complexity (RFC 3439 is a network oriented 
prerequisite in this area).

6.5. convergence of the digital ecosystem use:

The World Summit on the Information Society has clearly defined the 
humanity consensual target of a digital societal support being 
"people centered, à caractère humain, centrada en la persona". We 
assign the IUI the role to provide the user with an equal alternative 
basic, value added, extended and facilitated, and stable enough 
networking experience with any digital technology and digital 
communication network environment.

6.6. personal digital empowerment:

Through the above, the definition of a user oriented (graphic sign) 
international network character set (e.g. Bulgaria IDNccTLD problem), 
the establishment of an MDRS (metadata distributed registry system) 
to be understood as a distributed personal super-IANA cognition 
center for the three strata that we defined and their related 
services, etc. we want to ensure the full e-empowerment of every 
person (we make the difference between a user in a (de)centralized 
user (i.e. customer) centric approach and a person in a person 
centric distributed and/or intricate environment.

6.7. diktiologies support:

Since "diktyos" in Greek means network, diktyology has two meanings:

6.7.1. the science of studying networks, from quarks to the Internet 
through to political and commercial lobbies.

6.7.2. the multidimensional, diversified, open, and intricate network 
equivalent to an ontology. What "Projet.FRA" actually targets is a 
Francophone diktyology associated to a diktyography. Some network 
level and distributed/intricate integrated semantic web equivalent to 
be used and completed by people through the MDRS.

7. Market evolution

Now, some clever service provider's people will understand where the 
market lies for their corporation in this IUse usage evolution, and 
they will be the network commercial leaders of tomorrow. Others will 
not, and they will see the progressive obsolescence of their market share.

What is interesting is that the coming Brussels talk between ICANN 
and GAC may trigger a fast evolution if they obstinately (both sides) 
continue to ignore Vint Cerf's proposition, as the WG/IDNABis Chair, 
that the user oriented debate in the continuation of IDNA2008 should 
be managed by ICANN. I opposed that because ICANN does not represent 
users, but that was unnecessary: ICANN is not even interested in the 
matter (don't ask me why), even if it leads to total confusion in 
their part of the IUDNS.

8. Current situation

I/we delayed the introduction of ML-DNS for personal health (in my 
case) as well as for strategic reasons, to give time to the IESG, 
IAB, and Unicode members to consider where/how these issues should be 
addressed and documented, and to permit ICANN to reconsider, due to 
its new BoD Members. So far,

8.1. IESG and IAB were clear enough, the IUI does not belong to their 
"inside" Internet scope (moreover, it will deal with many other 
technologies), I am taking this into account to try to structure the 
IUse community, using the iucg at ietf.org non-WG mailing list as an 
open liaison between the two scopes.

8.2. Unicode has not introduced any proposition to support 
orthotypographic metadata.

8.3. ICANN has continued to enlarge their GAG (gTLD applicant Guide).
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.alvestrand.no/pipermail/idna-update/attachments/20110225/9a2f18e3/attachment-0001.html>