Final Sigma (was: RE: Esszett, Final Sigma, ZWJ and ZWNJ)

[Mark Davis]

Logically speaking, there is no difference between a UTF-8 model and a
Punycode model. Both can be thought of as transfer-encodings of Unicode.
Punycode is arcane, but it is simply and fully a lossless encoding of
Unicode.

The problems we are dealing with are not because of that, but because of the
fact that we are mapping and restricting on the client side. That is, the
encoding is orthogonal to the mapping/restricting. If we had UTF-8 in URLs
we would be faced with the same issue: how do we we map/restrict labels, and
where is it done. For example, the Greeks really want *more* mapping than is
done by IDNA2003, not less.

The one difference, which is orthogonal to the encoding issue, is that the
core DNS server-side lookup could require mapping/restricting, instead of
burdening the clients and/or registries with it. But the mapping/restricting
issues don't magically disappear... And it is unclear that every DNS server
should be required to implement all of the mapping/restricting rules. And we
still wouldn't want that to differ by language (for reasons outlined
earlier).

Don't get me wrong; it would be great to have UTF-8 in the DNS at some
point. It would make a number of things easier. But it would have no affect
on the issues that are causing so many disagreements (and/or
head-scratching).

Mark


On Fri, Feb 27, 2009 at 12:29, Andrew Sullivan <ajs at shinkuro.com> wrote:

> On Fri, Feb 27, 2009 at 01:17:47PM -0500, John C Klensin wrote:
>
> > Yes.  And what you didn't call out,
>
> […]
>
> Oh, sure -- I can think of _lots_ of ways this is complicated by the
> actual way the world is.  What you said, what Jaap said, the problem
> of getting all the right metadata in the first round (does anyone
> really think that once we open the "metadata for domain names" can,
> the only worm that will pop out is IDNs?) -- it's just a bad idea if
> we want to solve this problem quickly.
>
> > Conversely, if one can contemplate that radical a chance to the
> > DNS, then it is plausible to rethink the entire IDNA model and
> > start thinking about a transition to a UTF-8-native model
>
> Right.  I seem to recall asking some time ago whether that's what we
> need to start contemplating.  It sounded to me like the answer was,
> "No."  But if we're not willing to bite off that much, then we need to
> accept that we have to live with a bunch of limitations that are
> inherent in DNS.
>
> > Anyone who is willing to accept the 10 year (minimum) that
> > transition would take before we had stable IDNs should certainly
> > advocate just such a solution.
>
> I agree completely.
>
> A
> --
> Andrew Sullivan
> ajs at shinkuro.com
> Shinkuro, Inc.
> _______________________________________________
> Idna-update mailing list
> Idna-update at alvestrand.no
> http://www.alvestrand.no/mailman/listinfo/idna-update
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.alvestrand.no/pipermail/idna-update/attachments/20090227/6aac49d8/attachment.htm