Re: header-munging

"D. J. Bernstein" <djb@koobera.math.uic.edu> Tue, 17 September 1996 08:53 UTC

Received: from cnri by ietf.org id aa07527; 17 Sep 96 4:53 EDT
Received: from list.cren.net by CNRI.Reston.VA.US id aa21858; 17 Sep 96 4:53 EDT
Received: from localhost (localhost [127.0.0.1]) by list.cren.net (8.6.12/8.6.12) with SMTP id DAA08541; Tue, 17 Sep 1996 03:42:22 -0400
Received: from koobera.math.uic.edu (qmailr@KOOBERA.MATH.UIC.EDU [128.248.178.247]) by list.cren.net (8.6.12/8.6.12) with SMTP id DAA08523 for <ietf-smtp@list.cren.net>; Tue, 17 Sep 1996 03:42:12 -0400
Received: (qmail 6475 invoked by uid 666); 17 Sep 1996 07:46:50 -0000
Message-Id: <19960917074650.6474.qmail@koobera.math.uic.edu>
Date: 17 Sep 1996 07:46:50 -0000
Sender: owner-ietf-smtp@list.cren.net
Precedence: bulk
From: "D. J. Bernstein" <djb@koobera.math.uic.edu>
To: ietf-smtp@list.cren.net
Subject: Re: header-munging
X-Listprocessor-Version: 8.0 -- ListProcessor(tm) by CREN

> The problem is that 
> it's too easy for a site to not bother (either because they miss the
> subtlety or because the people running the site don't care) and use 
> the same SMTP server to do both submission and relaying.
> (as well as configuring it to do fixup)

If the site is happy with the current situation, why do you say there's
a problem?

For a site where sendmail always does fixups, and where there's not
enough mail for speed to be a problem, and where the users are happy
with how sendmail interprets their messages, why should they change
anything?

If they're _not_ happy, do you seriously believe that they're going to
``not bother'' correcting the situation?

> The danger of defining a new SUBMIT abstract module is that people
> will write a new SUBMIT program and try to place it outside of the 
> user's (or his site's) control -- where it cannot in general function 
> properly.

Generating correct messages is a _cooperative_ effort of both sides of
the MUA---the user's side and the ISP's side.

What you're saying is that the ISP's side can't carry the entire burden.
Indeed, some users want to send some rather weird messages. Obviously,
if a user wants something that the ISP can't do, the user has to do it
himself.

---Dan