Message-Id: <199710030401.AAA23912@spot.cs.utk.edu>
From: Keith Moore <moore@cs.utk.edu>
To: Dave Crocker <dcrocker@brandenburg.com>
cc: drums@cs.utk.edu, moore@cs.utk.edu
Subject: Re: wishful thinking vs. reality 
In-reply-to: Your message of "Thu, 02 Oct 1997 09:46:58 PDT."
             <3.0.3.32.19971002094658.036bbadc@ng.netgate.net> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Fri, 03 Oct 1997 00:01:04 -0400
Sender: moore@cs.utk.edu

> >Yes or no: Are you trying to prohibit reply-to-From?
> 
> 	replying to the From field, when Reply-To is present, is a
> violation of the explicit directive given by the originator of the
> note, specifying use of the reply-to field, in place of the From
> field.

I prefer to think of Reply-To as advice, rather than a directive.  If
the author of the reply chooses not to take that advice, it might be
impolite, but it shouldn't be a protocol violation.

IMHO user agents SHOULD use the reply-to header as a default
destination for replies, but also SHOULD (perhaps even MUST) allow the
author of a reply to override the reply-to header field, and specify a
different set of recipients, when composing the reply.

The author of the reply must have the final say about where the reply
goes.  This is true even for a 'reply' -- a message sent in response
to that particular subject message -- and not just a 'response'.

Keith