Re: [idn] IDNA interoperability failures, once again

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

> > Again assume LANG=en_US.UTF-8, so this can be done.
> This assumption is wrong.

Please look up the word ``assumption'' at m-w.com. I did not say that
everyone uses LANG=en_US.UTF-8. I explicitly restricted attention to
that locale, because it's a locale where the programs under discussion
_can_ display IDNs properly.

The point of this thread is that the IDNA deployment plan will cause
interoperability failures: in particular, mail will bounce. Until the
IDNA proponents can explain how to fix these problems in _one_ modern
Unicode-aware locale, there's no point in considering other locales.

> some national charset like ISO-8859-1 or KOI8-R

Converting UTF-8 to KOI8-R is unreliable, obviously. The reason we want
the world to settle on Unicode is to eliminate such problems. Has it
never occurred to you that some Russian users want to see non-KOI8-R
characters? Or that some Europeans want to see non-8859-1 characters?

Maybe you're happy never seeing anything outside 8859-1. But people
writing software still have to take time to deal with the character-set
mess---time taken away from giving you features that you want. Try not
to be so shortsighted.

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