Re: [alac] IDN document, draft 5 again

[Thomas Roessler <roessler@xxxxxxxxxxxxxxxxxx>]

On 2004-12-29 12:00:54 +0100, Vittorio Bertola wrote:

> Changes:
> - revised second para wording on cultural priorities (Thomas);

Better.  I'd prefer, though, to change the second sentence into
saying that IDNs are "not only" (as opposed to "not") "a technical
or business affair" ("matter" would be better"), and to strike the
third sentence entirely.

> - weakened recommendation #1 to make it a feasibility study (Thomas);

ok

> - added further clarification at the end of recommendation #4 on the
> opportunity of bundled registrations (we're not saying that bundles are
> compulsory, we're giving two options anyway) (Roberto);

"Should be sold and registered in bundle" still sounds compulsory.

If the point is that no two registrants should be able to register
names in the same equivalence class (i.e., if you don't want the
entire bundle, then the rest is blocked for anyone but you), then
write that.

The more I think of this, the more dubious the entire bundling
business appears to me.

BTW, the "liberte" vs. "liberté" example is still as bad as it was
in the first draft.  The two strings are not graphically equivalent
in any conceivable sense.  Please remove this example.

> - added an alternative to stopping the registrations, which is making
> them provisional, ie subject to a final check against the policies once
> approved;

I continue to object against this recommendation.  Declaring
existing IDN registrations provisional would be (1) unjustified and
(2) a fatal precedent [ICANN retroactively declaring domain name
registrations "provisional"], and certainly the wrong thing to do
for ICANN.  I have already explained in the past why holding up
registration of IDNs is a bad idea.

Making such a far-reaching recommendation on the basis of a gut
feeling that "the legitimacy of registrations sold before the
approval of final technical and policy standards is dubious" (which
is justified by another gut feeling, "or it might create an
extremely dangerous case for the future" -- what's meant by that?)
would, in my view, be irresponsible.

Please strike recommendation 8, or add a note that I object against
it.

> - added further clarification at the end of recommendation #9, to
> clarify that complete i18n of URIs does not necessarily have to happen 
> at the protocol level, but could happen at the application (Roberto).

> I did not change anything on #7. I am convinced that "universal
> access" to gTLDs in terms of available scripts should be forced
> onto registries, exactly like other forms of universal access are
> forced onto telecommunication operators. 

I'm not so convinced that attempting to introduce telecom-like
universal access/service obligations on higher levels of the
Internet's protocol stack is such a good idea.

> I don't think you can go to Norbert Klein and say "you know what?
> you can't have .com domain names in Cambodian, because it's not a
> relevant market in business terms" - I don't think this is right.

I wonder to what extent you're attempting to work around the
problems created by ICANN's overly restrictive approach to new
TLDs...  After all, one business's irrelevant market is another
business's opportunity.

> I'm not sure whether this would reinforce or weaken my argument :) but
> this morning I was giving a glance at the statement by the WSIS working
> group of ITU: 

> http://www.itu.int/council/wsis/Geneva3_04/intgov-contribution-wg-wsis.doc

> and I found this on page 13:

> "For example, deployment of Internationalized Domain Names (IDN) needs
> to be accelerated. If Internet style addresses are to become a
> predominant global addressing scheme in the future for end point
> identifiers or subscriber identity (e.g. for VoIP), then they must
> support all language scripts of the world."

And as the next step, we Unicode-enable the dials on the world's
VoIP-enabled telephones?  And then the persons in front of these
telephones? I'm increasingly suspicious that IDNs will, in some
areas, be a "cure" that's much worse than the "problems" it's
supposed to solve.

Concerning section 9, I would suggest to remove the word "promptly"
from the last sentence.

-- 
Thomas Roessler · Personal soap box at <http://log.does-not-exist.org/>.