Mapping and Variants

[JFC Morfin]

At 18:15 11/03/2009, Erik van der Poel wrote:
> From my point of view, we have plenty of
>participation in this WG from people who state how difficult it is to
>introduce these, but very little participation from those who can
>state the actual needs of the various impacted communities. So there
>may be a natural tendency of this WG to gravitate toward approaches
>that do not require difficult transitions.
>
>So my question is: Given that this may be our last chance to make
>changes of this nature to the IDNA spec, is it OK for the IETF to be
>biased against the communities' needs?

These words are the type that I should support. However, I should 
tell you how the IETF and we (<mailto:france at large>france at large) have 
worked on the matter.

1. We first posed the question to Vint: Does IDNA target to be the 
ML-DNS solution that people expect? James Seng and Vint Cerf, 
responded with a rather clear: "no". It looks to be RFC compatible. 
(This means constraints on usage, or to encapsulate it into something 
helpfull).

2. Vint proposed for me to create an ML-DNS side list. I saw that:
     - it would extend beyond the Charter area: Vint confirmed in 
asking me to not refer to ML-DNS any more.
     -with the Internet PLUS concept (Parallel Layered User 
Systemic), we came with something that could be generic.

3. I, therefore, appealed the IESG "against" Vint precisely for 
community users to be helped in dialoging with the IETF work - so to 
formally know the IESG attitude. The appeal was positively dismissed 
and the <mailto:iucg at ietf.org>iucg at ietf.org list and site were 
subsequently created. It will certainly take some time to build-up 
and for me to transfer it to a more adequate governance..

4. I then appealed to the IAB in order to obtain their respective 
views. This is now clear, and documented in the IUCG Charter and FAQs 
(http://iucg.org site).

5. I also used the French language issues to learn how to best help 
other language communities.
    - we obtained strong French people support, but I was not 
listened to very much in here.
    - since only TLD Managers (DENIC, GRNIC) are listened to, we have 
initiated the ".fra" French language gTLD project, which is to be 
manned, financed, governed, and administered along the international 
dynamic coalition/enhanced cooperation WSIS framework.

6. The DNS system should be considered in a systemic manner. We 
identify three layers.
- the names of usage (what people actually enter, remember, and publicize)
- the registry namespace policy with two case-sensitivities
- the protocol where:
   - we do not care about IDNA 2003 compatibility, since we never used it.
   - we intend to use "xs--" (for case sensitive) instead of "xn--" 
(for non-case-sensitive)

7. Every language community should do the same, especially the main 
languages and all the ISO 3166 languages. This means that there 
should be an ".eng" project (or perhaps the WG could split on English 
to study this more quickly) in order for the WG to better 
differentiate what belongs to (1) the language (2) the registries (3) 
the protocol.

The IETF should only document the protocol. That protocol should 
provide Registries and users with the tools/solutions that they need 
to implement their own policies.

Best
jfc


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.alvestrand.no/pipermail/idna-update/attachments/20090312/65c84a62/attachment.htm