Re: Last Call: SMTP Message Submission to Proposed Standard

[Keith Moore <moore at cs.utk.edu>]

Greg,

In a nutshell, I think the problem is that there are too many
different things that people want to do at message submission time -
and if earlier discussions are any indication, the diversity is such
that we're not likely to get consensus on exactly how people should do
any of these things.  And yet, there's substantial agreement that it
is desirable to handle message submission separately from relaying -
but this requires protocol changes, and thus it requires agreement
between implementors if different client and server implementations
want to interoperate.

I don't want a standard which gives license to an MTA or submission
server to do harm to messages.  However, I fear that if we hold this
up until it is a "complete protocol", that we'll never be able to get
it out the door.

Keith