Errors-to: owner-drums@CS.UTK.EDU
Message-Id: <199508161511.LAA24081@wilma.cs.utk.edu>
From: Keith Moore <moore@CS.UTK.EDU>
To: conklin@info.cren.net (Jim Conklin)
cc: Eric Allman <eric@CS.Berkeley.EDU>, ietf-drums <drums@CS.UTK.EDU>,
        moore@CS.UTK.EDU
Subject: Re: "Reply-To" (and lists) 
In-reply-to: Your message of "Wed, 16 Aug 1995 10:20:18 BST."
             <199508161417.KAA22449@info.cren.net> 
Date: Wed, 16 Aug 1995 11:10:57 -0400
Sender: moore@CS.UTK.EDU

>   I must confess that I've not thought my way through how I think this
> could be done, but I think the list use (or misuse, if you're of that
> persuasion) of Reply-To will continue until an alternative is provided,
> because there is a large class of people for whom it's convenient and
> acceptable to direct replies to list mail back to the list, and that
> segment of the user community will demand a facility for doing that.

While I think Jim is right, I want to caution against us being persuaded
to break a useful feature of the Internet mail protocols simply because
there is a "large class of people for whom it's convenient".  In general,
for almost any misuse of the mail protocols that causes the net problems, 
there is a significant group of people who believe it is the right behavior.

For instance, those who grep up on mail systems that had only one reply 
address, often think that the "right" way to do lists is to have them 
rewrite that reply address to point to the list.  While it was perhaps
the best thing they could do given the constraints of their environment,
it doesn't make it right for the Internet as a whole.

We should feel free to insist that certain kinds of behavior are broken.
On the other hand, it's a also good idea to write standards and code that
are tolerant of such behavior, since some people will do what they want 
no matter what we say.

Keith