Re: [DNSOP] draft-liman-tld-names-04

[Masataka Ohta <[e-mail]>]

Andrew Sullivan wrote:

> I think Ohta-san has made it clear that he thinks IDNA is broken.
> Point noted.  But there's a whole list devoted to the topic of IDNA,

That's not an appropriate topic to be discussed further here
and, perhaps, not in IDNA list.

Instead, DNS operational problems caused by insufficient support
for (extended) case insensitivity or canonicalization are
discussed.

I showed examples of Latin characters for French because it
should be much more familiar to most of you than Chinese or
Arabic ones.

An operational requirement for multiple identical zones,
names of which are not canonicalized by IDNA, is zone
synchronization, for which exponentially many BNAME RR, which
does not exist, is required, which is a problem.

Another operational problem is exponentially many signatures,
which is huge, required for DNSSEC.

Unless we accept exponentially many BNAME RRs (and abandon
DNSSEC, which I recommend anyway), IDNA2008 must be revised,
details of which I am not interested in.

> and I don't believe it's in scope for this draft.

As for the draft, as some non-ASCII name wrote, locale
information is necessary to properly handle case
insensitivity of Latin/French. And it is necessary at
least for TLDs, which makes publishing of the draft should
be delayed until details of IDNA revision is available.

					Masataka Ohta