Re: Last Call: SMTP Message Submission to Proposed Standard
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Last Call: SMTP Message Submission to Proposed Standard



I also believe that draft-gellens-submit-06.txt is
an incomplete protocol addressing inadequately
considered requirements. It aims too low and ends
up shooting way too low.

Since a new port number has been assigned to this
protocol, there are no backwards compatibility and
current installed base interoperability
issues. Therefore, the expectation is that this new
protocol -- which is to address a sensitive and
important area of Internet Mail -- be well thought
out, complete and extendible. Instead it is yet
another hack on top of the more than 15 year old
SMTP.

Given a clean plate (port 587) and given what has
been learned from the SMTP and X.400 experience
over the past 15 years, we should be able to do a
lot more for message submission than what
draft-gellens-submit-06.txt attempts to do. For
example: better authentication, better error
reporting, less historic options, better
efficiency, submission server referrals, ...

I second Greg's suggestion of: 

>>>>> On Wed, 22 Apr 1998 13:42:35 -0500, "Vaudreuil, Greg" <gregv at lucent.com> said:

  Greg> Lets have a working group to finish this work and consider
  Greg> the full requirements of this important function for Internet Mail.

I believe that restricting ourselves to "SMTP
Message Submission Protocol" instead of
considering a brand new "Message Submission
Protocol" is an architectural mistake. If we go to
a new port number, then we have a clean plate and
we should take full advantage of it. Judicial
re-use of proven technology is very different from
hacking on top of Simple 15 year old
protocols. The next millennium need not also be
the era of Simple Protocols.

Going to a new port number opens the door for
moving Internet Mail Protocols forward. But, we
are not planting the right seeds with
draft-gellens-submit-06.txt.

More work is needed here, both at the analysis
and requirements gathering level as well as at the
architecture and design level.

...Mohsen.



Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.

Note: Messages sent to this list are the opinions of the senders and do not imply endorsement by the IETF.