Archaic scripts (was: Re: New version:
jefsey at jefsey.com
Fri May 9 00:37:07 CEST 2008
At 14:10 08/05/2008, John C Klensin wrote:
>Put differently, some of the issues that MLTF wants to address, at
>least as I understand them, build on IDNs but do not need to be
>--and should not be-- part of them. Far more of the MLTF issues do
>not involve IDNs at all. To confuse IDN work with those MLTF
>agendas involves what, in networking contexts, is called a layering
>violation. In more ordinary language, it is just confusion.
As the MLTF Moderator, I need to clarify this.
1. MLTF is a small, mostly French-speaking, research non-profit unit
about multilinguistics and facilitation technologies. It is based
upon the identification that "linguistic diversity" calls for
linguists to deal with language aspects, and needs system architects
to deal with the respective technical support of diversity. We named
this new discipline multilinguistics.
The first observation that we made is that linguistic diversity does
not exist as such: it is the linguistic perspective of semiotic
diversity. Some cultures and peoples may express things through
language that others express through postures, gestures, icons,
signs, etc. This means that what is of utmost interest is not the
text (Chomsky) but rather the utterance process (Culioli) and the way
that the locutor represents the underlying informational semantic to
an identified collocutor in considering or /through pragmatic,
linguistic, grammatic, vocabulary, diplomatic, semiotic, stylistic,
mediatic, interpretation, etc. mutual adjustments. We label
facilitation as what adds value in one or several of these areas.
In this process,
- the Internet is one of the dumb asynchronous proctocol sets
permitting the digital convergence of the logical continuity strata
(OSI, 7 layers model) to deliver end to end interoperability,
- over electromagnetic plug to plug interconnections.
- the MLTF are interested in the semantic strata atop it, which is
concerned with brain to brain interintelligibility.
Therefore, you should be right: the MLTF should only build atop the
IETF, using and not interfering with its deliverables. The same as
IETF builds atop the ITU and does not interfere with its
However, there is always a risk
- either of a border definition, in turn leading to layer/strata
violations and the confusion that you describe.
- or incomplete, or even inappropriate, support of the functions that we need.
The situation that we are in is, specifically:
- names are semantic and universal the MLTF has delayed the work,
development, testing, and deployment in the naming area for years,
waiting for the IETF in order to stay interoperable.
- IP addresses are limited to the Internet topology and still mostly IPv4.
- the DNS is an interface between semantics (naming) and digital (IP
- the way that the DNS works, in which the true border is the
keyboard: (a) what is in the user's mind is semantic, (b) what is
processed should be digital [hence symbols on keyboards or in tables
Your good suggestion to calrify this and the corresponding debates
were: the IETF should only support scripts and characters, and should
not consider languages and meanings. Its today's problem is
particularly an IETF debate about languages and script filtering (two
areas where the IETF does not claim RFC3935 Technical Competence).
As a result, the MLTF suspects the IETF may not deliver the solution
that it has been waiting so long for. It means it would have to shop
somewhere else or make it by itself.
However, (1) you state that there cannot be two IDN for the DNS. This
does not mean that there cannot be several IDNAs (ex. 2003 and 2008),
(2) Louis suggested a compromise, and I proposed to build upon it.
All the best.
More information about the Idna-update