Sender: jwz@netscape.com (Jamie Zawinski)
Message-ID: <3430616C.CDFC49AA@netscape.com>
Date: Mon, 29 Sep 1997 19:18:20 -0700
From: Jamie Zawinski <jwz@netscape.com>
Organization: Netscape Communications Corporation, Mozilla Division
MIME-Version: 1.0
To: drums@cs.utk.edu
Subject: forwarding (was Re: Reply-To)
References: <199709300123.VAA04405@CS.UTK.EDU>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Roger Fajman wrote:
> 
> One case that I encounter fairly frequently, but haven't seen
> mentioned here explicitly is the following:
[ ...getting a message and passing it on in such a way that the
  2nd recipient will, by default, reply to the original author...]

Yes, I agree that that's a common situation/desire.  (I think the
latest marketing buzzword for this is "workflow", but I usually try 
to avoid using words like that in polite company.)

> Another case I don't recall being mentioned is replying to a forwarded
> message.  Sometimes I want to reply to the person who sent the
> original message, sometimes to the person who forwarded it, and
> sometimes to both.  Sometimes the message was forwarded to a list and
> I may or may not want the list to see my reply.

It's true that one sometimes wants to do any of those things, but I
think that there's no problem here; the existing ways of encapsulating
messages give enough information for the recipient to do any of those
things.  MUAs can of course make this easier or harder, depending on 
the tools they provide...

I also suspect imprecision in the use of the word "forward".
I consider this and this only forwarding:

	From: foo
	To: bar
	MIME-Version: 1.0
	Content-Type: message/rfc822

	From: baz
	To: foo

	forwarded body here

There are any number of ways the MUA could deal with this unambiguously;
it's not necessarily complicated for the user to express their desire
about which message they want to reply to (the inner one or the outer
one.)

(Ok, I'd still consider it forwarding if the MIME headers were missing,
but in that case, it's a lot harder for an MUA to do the right thing...)

But when some people say forwarding, what they mean is

	From: foo
	To: bar

	baz wrote:
	>
	> forwarded body here

This is generally caused by using MUAs that don't provide any other
easy way to do it, other than cut-and-paste.  It's true that it's
not as simple a matter for a receiving MUA to allow the recipient to
direct replies to "baz", and that perhaps "foo" could make "bar's" 
life easier by putting in a Reply-To; but my answer is, "don't go 
like that."  That's not a forward.  If you want to forward, then
forward.  Don't reply-and-quote-and-change-the-To-field and call it
forwarding.

Sometimes people mean this when they say forwarding:

	From: baz
	To: foo
	Resent-From: foo
	Resent-To: bar

	forwarded body here

I think the more common usage refers to that as "resending" or
"redirecting" these days (though 822, despite the header names,
seems to call it forwarding.)  I don't see how Reply-To would
necessarily play a part here; again, the user (and the MUA) have 
enough information to make an informed decision about which author 
a reply should go to, depending on the nature of that reply.

Of course, my least favorite of all possible forwarding abuses is:

	From: baz (by way of foo)
	To: bar

	forwarded body here

which is just horrible beyond compare... 

-- 
Jamie Zawinski         http://people.netscape.com/jwz/      about:jwz