Re: [idn] new I-D: Safely Encoding of likeness information into ACE label version 0.2

[Eric Brunner-Williams in Portland Maine <brunner@nic-naa.net>]

> This is a pipe-dream. See my messages of
> Wednesday, July 18, 2001 09:45
> Thursday, July 19, 2001 08:20

Those were good reading. I enjoyed the font example. I look forward to
stuffing 1x1 pixel gifs into <urk> HTMLized </urk> email, particularly
around what appear to be domain names.

...
> We have had a lot of experience dealing with huge numbers of characters.

The "We" here is the UTC, and other character standards body participants and
character rendering/processessing implementors. Not the run-of-the-mill IETF
contributors. Which is my first point:
	this is the work of bodies which create abstract character
	repertoires and coded character sets

> It is harder than it looks.

And it isn't DNS. We could be playing this game with:

	a) any system using all of ASCII, (What do unprintable characters
	   look like?)
	b) any system using all of 8859-*, (Ditto plus Latin/Greek)
	c) any system using just about any 8-, 15-, or 16-bit encoding,
	   whether of propriatary, national, or international origins,
and
	d) any system that "discovered" and "negociated" encodings, e.g., 
	   any of a), b), and c) above,
as well as
	e) Unicode.

Which is my second point:
	A protocol that is capable of signaling the existence of name
	proximality is something other than the DNS.

Eric

P.S. Someone, in the spirit of i18n-friendliness, changed the sanity check
     in Solaris 2.3 syslogd from isprint() to isascii(). I unpatched it. 
     I'll pay one beer to the first person who correctly identifies the
     real bug. Hint: Its a real security bug, has little to do with any
     specific Un*x, and is related to C0 Controls.