Einar Stefferud <Stef(_at_)ics(_dot_)uci(_dot_)edu> wrote on Tue, 25 Jun 91
15:56:54
MDT...
I have become convinced, through silence and avoidance of reading all
this XXXX effluvia, that it is hopeless to deal with the 8-7 encoding
anywhere but at the original UA posting point. PERIOD. FULL STOP.
OK, no problem with that one.... it is where I started from.
So, what we need now is some strength of resolve on the part of the
IETF and the IAB to declare that THIS IS INTERNET OPERATIONAL POLICY,
and that anyone who does otherwise is in fact IN VIOLATION OF THE
INTERNET STANDARDS.
Again, agreed, with the understanding that I've never seen symptoms
of such resolve about anything at, or near, the applications level.
So, we will have set the ultimate design
goal for those who want to do it anyway.
"Just don't ever get caught!"
Hesitant ok. The hesitation is about the fact that several of us seem
to have concluded that the 821 rules about trace fields (Received:) are
too weak, and that there should be a firm rule expanding the requirement
for adding these fields at relays to an additional requirement (and
probably some additional subfields) that anything modifying a message
body or munging headers *in any way at all* add trace fields to document
what was done and by whom.
I think that is good doctrine; maybe someday someone will get around
to writing it up (if the ietf-smtp list remains quiet, it might even be
me). But, if it is good doctrine, then the above design goal would need
a little rephrasing.
--john