Re: [idn] WG last call summary

["D. J. Bernstein" <djb@cr.yp.to>]

John C Klensin writes:
> Well, we can keep repeating this until it becomes true, water
> flies uphill, and other convenient things happen.

There is no excuse for 8-bit failures under the DNS specifications. More
importantly, there is no evidence of 8-bit failures in real DNS servers.

> As this working group has discovered multiple times now,
> case-insensitivity is not precisely defined for many non-ASCII
> scripts.

RFC 1034 says ``case-insensitive manner, assuming an ASCII character
set.'' The obvious interpretation, and the interpretation universally
adopted by DNS cache implementors, is that (1) bytes 65-90 are treated
identically to bytes 97-122 respectively and (2) there are no other
equivalences.

If the wording isn't clear enough for non-implementors, perhaps it could
be improved, but that wouldn't justify your false claims that DNS has
problems with 8-bit names.

---D. J. Bernstein, Associate Professor, Department of Mathematics,
Statistics, and Computer Science, University of Illinois at Chicago