Message-Id: <v04003a08b140c297f3a8@[130.237.150.138]>
In-Reply-To: <19980326210158.9510.qmail@cr.yp.to>
References: <v04003a19b13f538f9cd9@[130.237.150.138]>
 <Pine.SOL.3.95.980325230731.4280B-100000@elwood.innosoft.com>
 <v04003a1db13fd88bde17@[130.237.150.138]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Fri, 27 Mar 1998 03:54:43 +0100
To: "D. J. Bernstein" <djb@cr.yp.to>, drums@cs.utk.edu
From: Jacob Palme <jpalme@dsv.su.se>
Subject: Re: Reply-To Harsh realities

At 22.01 +0100 98-03-26, D. J. Bernstein wrote:
> Jacob Palme writes:
>   [ MLMs ]
> > Adding a "Mail-Followup-To" to a message which does not have any such
> > header is not any large problem
>
> 1. It produces different behavior for different copies of one message.

That cannot be avoided for any header field which is added by a list
expander. And we cannot stop list expanders from adding header fields,
this is shown by the practical experience with "Reply-To". You must
give them something, either "Mail-Followup-To" or "List-Followup-To".
>
> 2. It eliminates extra copies for users who do want them---some lists
> are incredibly slow!

Not if the sender has supplied "Mail-Followup-To".
>
> 3. It forces people to insert Mail-Followup-To in self-defense, even
> when they don't need it. It destroys the default. ``Insert me into all
> your messages---or else!'' What would happen if every field did that?

That is only for messages sent to multiple recipients, and is because
only the sender knows enough to produce a good such header for
messages with multiple recipients. And it can be done automatically,
so there is no extra burden on the sender.

------------------------------------------------------------------------
Jacob Palme <jpalme@dsv.su.se> (Stockholm University and KTH)
for more info see URL: http://www.dsv.su.se/~jpalme