Re: iesg comment re message submission in draft-ietf-grip-isp-expectations-03.txt

["Mike O'Dell" <mo@uu.net>]

the disproof of non-existence ("i know of several servers...")
is hardly a statement of production-quality capability,
and the operational change you demand impacts many millions
of users.  one major issue is that many email CLIENTS do not
support an alternative submission port - it's rare enough to
find one that can post email using a server different from the
one it fetched the email from.  and while I applaud Eudora for
its flexibility, it is, alas, rather unique in that respect.

the fact that this part of the "Internet Architecture" has long
muddled the various components of email systems has taken a dreadful
toll which will unfortunately take a long time to fix.  the document
should probably reflect this particular reality while suggesting 
a course of action when one has the luxury of blank paper.

Note: i don't for an instant dispute the desirability of authenticated
email submission - it is clearly a very valuable tool, but it also
relies on the strength of the identity mechanisms which vary
considerably, and making unilateral demands of a large user base is
known to be limited in its short-term utility.

this gets to the very heart of my concerns with this document.

instead of offering reasoned advice in an reasonable tone, it 
makes pontifical demands in a very strident voice, ostensibily
distinguishing the saved from the damned.  

the style is innapropriate for the IETF, and
does little to support the credibility of the document,
the validity of any one issue not withstanding.

    -mo