Re: [idn] Time to reconsider

[Eric Brunner-Williams in Portland Maine <brunner@nic-naa.net>]

> Since I
> haven't asked the author of the quoted material/ questions, I've
> anonymized those remarks

John, I suggested that we take the scope issue back to the main list.
Quote me till the cows come home. We also now know the origin of [30].

Semantic scope isn't encoding selection, encoding is just one possible
policy that could have scoped semantics, or could be scopeless. In the
context I mentioned in our side-bar, now back in the main list, rules
for equivalency over the same encoding (utf8) were the subject of
discussion. In the origin-of-[30] note, rules for equivalency over the
same encoding (ASCII) were the subject of discussion.

> Quick summary: I believe that we should not agree to either
> Nameprep or the Requirements document as they stand because the
> underlying assumptions behind what we are doing have been shown,
> by the WG's work, to be fundamentally flawed. 

This far I agree. Where I expect we disagree is what the underlying
assumption(s) are which appear (to each of us) to be fundamentally
flawed, which I'll restate as affirmative goals.

	JK:	extend "meaning" (languages, scripts, and fuzzy matching)
	EB-W:	extend "labels" (0x2d, 0x30-0x39, 0x41-0x5a, 0x61-0x7z)

Is consensus possible?

In practical terms, "apparently not":

	VGS is going live with its RACE, the mechanism for non-ASCII
	in the .com namespace.

	CN went live with UTF8, the mechanism for ASCII and non-ASCII
	in the .cn namespace.

There are a lot of other choices to pick from, but of these two, I'm
going to deploy UTF8, subject to possible ICANN and possible Registry
Trade Association regulatory oversight and technical coordination.

In working group terms, the answer isn't much better.

There are 6,182,148 (Walid), and 5,410,306 (Ydisg), encumberments, and we
have simply not had the bandwidth to conclude independently as VGS did that
neither applies to IDNA, a lack of knowledge I happen to know is shared by
at least two other namespace vendors.

We've a nameprep/idna poll (my reply was "no" and "rfc2044") outstanding.

All this, and about 20 drafts in various stages of aging are before this
working group, along with the prior organizational art, rfc2044 and rfc2130,
and rfc1034/1035.

It will take a wiser head than mine to sort out the WG roadmap to rough
consensus in finite time, and it appears that time is very finite indeed.

Eric