Re: [idn] UTF-8 as the long-term IDN solution

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

Dave Crocker writes:
> In 1992 we were assured that Unicode was instantly going to be in mass
> use, for example.

Really? Precise reference and exact quote, please. Who told you that
Unicode was ``instantly'' going to be in mass use? (It's widely used
now, though it isn't universal yet.)

When I say that many people proposed deprecating 7-bit SMTP a decade
ago, I can back up my statements. For example, Andre Pirard said ``I
find Robert Ullmann's proposition excellent. Require that relays pass 8
bits ...'' See http://www.imc.org/ietf-smtp/old-archive/0326.html.

The only objection to that requirement was the claim that 8-bit support
would take a long time to be deployed. Vixie said that he had some
seven-year-old sendmail binaries, for example, and concluded ``with
near-certainty'' that ``any changes to the SMTP spec will take at least
a decade to reach 90% of the critical server population.''

Well, it's now a decade later, and every critical server supports 8 bits
in the message body EVEN THOUGH YOU IETF IDIOTS STILL ALLOW 7-BIT MTAS.
If the SMTP spec had changed ten years ago to require transparent 8-bit
handling in both the header and the body, we wouldn't have sendmail's
UTF-8 failures today.

I DO NOT WANT TO BE FACED WITH 8-BIT PROBLEMS IN ANOTHER TEN YEARS. I am
astounded that anyone would object to requiring 8-bit support. Even in
the case of a horrible piece of code like sendmail, handling 8 bits is a
simple matter of converting dangerous characters to two-byte sequences
on input and doing the opposite conversion on output.

> After all, we all have such a good track record when predicting the
> future, don't we. 

Obviously _you_ don't.

---Dan