On Thu, 9 Sep 2004, Andrew Newton wrote:
If the above is incorrect understanding about conclusions from interim
meeting in May and list discussions prior to that, then please correct
me
and show at which point in the WG dicussions did we reach consensus
on dns mechanisms for HELO identity checking.
The conclusion was that CSV semantics should not put into Sender ID,
which uses a particular syntax known as SPF syntax and the TXT & SPF
records.
http://www.imc.org/ietf-mxcomp/mail-archive/msg02527.html
There appears to be problems with above statement. The problem has to do
with interpretation of how names "SenderID" and "CSV" are used. It appears
based on yesterday's statement that you imply that you were using both in
more general sense and assuming SenderID to mean anything that working
group comes up with that is using SPF syntax and CSV to mean HELO
identity checking in general.
But that is not how I or many others understand this. For me SenderID is
particular proposal made when attempting to merge SPF with CallerID and
that means its particular to the PRA identity (that is why until now
SenderID was always distinguished from SPF classic and it was understood
that SenderID does not include SPF classic but that MARID may at later
time standarding SPF classic envelope mail-from checks on its own after
conclusion of SenderID identity work. Similarly CSV is also particular
proposal made by Dave Crocker on the HELO identity checking.
In that sense merging these two completely different and particular
proposals or even discussing it would not have been productive for the
group so I agree with what you said in that post. I understood it to mean
that we should focus on SenderID identity at the time and discuss other
identities and specific CSV proposal later.
I do not see the above to imply that we agreed (or disagreed) that we can
not include checking other identities by means of SPF syntax, again it
was understood that discussions of that will take place later on. And
since we agreed not discuss CSV or HELO identity checking until SenderID
is worked on, it means there was no discussion and thus no consensus about
what dns symantics should be used with that identity...
--
William Leibzon
Elan Networks
william(_at_)elan(_dot_)net