Re: Are international characters allowed in email addresses?

Ned,

I accept that you belive that I act in bad faith. I don't seek to change
your beliefs.

I don't understand that something as central to the purpose and process
of a working group as its requirements gathering can be both irrelevant
and recommended reading.

I don't think Mr. Harrison was asking a question that shouldn't have been
asked in discuss@apps.ietf.org. I don't think discussing the underlying
byte processing is uninteresting, or something that should not have been
made in discuss@apps.ietf.org. At Minneapolis (previous time) the Apps
Open meeting resulted in a (very) modest attempt to ask or even answer
some common architectural problems in the Apps area. You should recall
this, you were present.

Having done two i18n OS upgrades, which touched everything from sort to
sccs to csh and ksh in apps, and lib{c,w} in the system libraries, and 
the locale definitions, and display width in the tty subsystem, from an
explicit ASCII and an explicit ASCII+HP15 process and file encoding to a
code set independent (implicit UTF-8) process and file encoding, I'm not
seeing the width and depth of engineering experience on this problem now
in the IDN WG. It isn't present in Keith alone, or you alone, or both of
you together.

I don't understand that something as broadly consequental as implicit
(and subsetted) ASCII can't be discussed except in a working group with
utterly asbysmal process and progress, and where things need fixing.
I can understand that over the past year the IETF has become a meaner,
even nastier milieu in which to attempt to work. I hope it is a local
problem.

As to my expectations, they are that technical discussion prevail. Having
some experience designing and implementing system-wide expansions of the
base character handling from ASCII to a larger range of functionality, I
am at a loss to understand the uncivil, even abusive, conduct that passes
frequently without comment when the issue of i18n arises in an IETF context.
2026, 2282, ...

It isn't necessary to conclude anything about dns labels (see bcp42) to
look at the utility for changing sendmail.

Keith's note was previous, not subsequent.

> Even if I were interested in discussing header processing issues at this
> point (I'm not), it is now clear that you have no intention in engaging in
> a sensible or reasonable discussion. So as far as I'm concerned this
> discussion is now terminated.

Yes, an end has been reached. I am glad I didn't write that paragraph.

Eric

Received on Friday, 29 June 2001 03:46:01 UTC