RE: [idn] I don't want 8-bit failures in 2011

["Brian W. Spolarich" <briansp@walid.com>]

| One of the great things about the EDNS proposal is that it provides long
| labels and names. Since this would be a favorite feature of a lot of the
| non-English locales, it would be widely-adopted.

  I don't find this to be a convincing statement.  My impression was that
the conventional wisdom with respect to CJK was that names would generally
be shorter.  Host labels are used by humans to identify and reach hosts and
network services, not to compose short novellas. :-)

  Verisign folks (or anyone who might know), from the ML testbed, do we find
that the CJK labels in terms of number of codepoints are longer, shorter, or
about the same as the non-ML ones?

  If one is really worried about the 'long label problem', what data do we
have on whether ACE or UTF-8 is better overall?  Some ACEs are better at
compression than others of course.

  All of this has been discussed in the past, but it doesn't hurt to re-open
this issue if for no other reason than to facilitate consensus.

  I still think Klensin's point is right on:  there are things we're doing
with the DNS today that don't fit well with its purpose.  We can either
choose to change the DNS or provide a new service layer that is designed to
meet those needs.

  -bws