From: Robert Elz <hostmaster@munnari.oz.au>
To: djb@koobera.math.uic.edu (D. J. Bernstein)
Cc: drums@cs.utk.edu
Subject: Re: re PIPELINING 
In-Reply-To: Your message of "06 Mar 1996 03:18:33 GMT."
             <19960306031833.11466.qmail@koobera.math.uic.edu> 
Date: Thu, 07 Mar 1996 02:30:19 +1100
Message-Id: <1289.826126219@munnari.OZ.AU>
Sender: kre@munnari.oz.au

    Date:        6 Mar 1996 03:18:33 GMT
    From:        djb@koobera.math.uic.edu (D. J. Bernstein)
    Message-ID:  <19960306031833.11466.qmail@koobera.math.uic.edu>


    It is, however, obviously not understood by the
    people who wrote the relevant paragraph in RFC 1123 and the people who
    uncritically copied it into the DRUMS spec.

It would have helped had you indicated which paragraph you're
objecting to...
    
    Here is a proposed rewrite of the paragraph:
    
    When a client sends a single message to two or more recipients over
    a single SMTP connection, it SHOULD NOT use the sequence MAIL, RCPT,
    DATA, MAIL, RCPT, DATA. The sequence MAIL, RCPT, RCPT, DATA uses less
    time and bandwidth.

This is extraordinary!   Not that you woulld want that wording,
but that you could believe that it says anything different now.

Here is 1123...

            When the same message is to be delivered to several users on
            the same host, only one copy of the message SHOULD be
            transmitted.  That is, the sender-SMTP should use the
            command sequence: RCPT, RCPT,... RCPT, DATA instead of the
            sequence: RCPT, DATA, RCPT, DATA,... RCPT, DATA.
            Implementation of this efficiency feature is strongly urged.

            Similarly, the sender-SMTP MAY support multiple concurrent
            outgoing mail transactions to achieve timely delivery.
            However, some limit SHOULD be imposed to protect the host
            from devoting all its resources to mail.

That is, it already says not to use RCPT DATA RCPT DATA ...
(which I would have had a hard time imagining anyone doing
in the first place, but someone must have for it to have been
felt necessary to say anything at all).   It also expliticly
says that sending multiple copies in parallel is OK.   The
only thing I can imagine that you're objecting to is the
SHOULD to send all copies to one host in one message,
as RCPT RCPT RCPT .. DATA, rather than sending them in
several parallel connections, which personally I strongly
object to deleting.  One mail connection is all that's needed
to deliver mail to all users on my hosts thanks, and further
to all users on hosts MX'd at one of mine as well.   I don't
care that you want to get the mail through a fraction faster,
I don't want the extra processing load, nor the extra bandwidth.

kre