From: "Bart Schaefer" <schaefer@brasslantern.com>
Message-Id: <970927100150.ZM1253@candle.brasslantern.com>
Date: Sat, 27 Sep 1997 10:01:50 -0700
In-Reply-To: <8272.875376553@munnari.OZ.AU>
Comments: In reply to Robert Elz <kre@munnari.OZ.AU>
        "Re: Reply-To (was Re: Issues from the past few weeks)" (Sep 28,  2:09am)
References: <8272.875376553@munnari.OZ.AU>
To: drums@cs.utk.edu
Subject: Re: Reply-To (was Re: Issues from the past few weeks)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii

On Sep 28,  2:09am, Robert Elz wrote:
} Subject: Re: Reply-To (was Re: Issues from the past few weeks)
}
}   | Both interpretations ("reply to the list", and "if you reply to 
}   | the list then don't CC me") are advisory.  Of course the user is 
}   | free to ignore either piece of advice.  I don't think anyone has 
}   | ever suggested otherwise.
} 
} They do, all the time (in other ways) - 822 does, by telling people they're
} not supposed to use the From: address when a Reply-To: header exists.

I'm interested in seeing your response to Dave Crocker's message stating
that it was in fact the intent of 822 that Reply-To indicate the sender's
preference that replies not be sent to the From: address.

I'm interested in JWZ's response to Dave, too, since Jamie's currently the
most adamant proponent of the reply-to-sender/reply-to-all UI.

} That's insane - there's nothing at all that suggests anywhere in 822
} that I can't create a new message, and send it to any address I like,
} including an address that happens to be the the From: header of the
} message I am staring at, but 822 tells us that the "reply" function is
} not allowed to do that if the Reply-To header is there.  100% dumb.

I think the problem there is a misreading of the word "should".  In 822,
"should" doesn't carry the weight of "SHOULD" in more recent RFCs.  If I
understand Dave correctly, the admonition in 822 does *not* mean that a
UA is "not allowed" to send a message to the From: address; merely that
the Reply-To: address is the preferred destination for replies.

Given Dave's statement of the intent of 822, and given that the ietf-mailext
list is already discussing Personal-Reply-To and Group-Reply-To headers (see
  http://www.imc.org/mailext/mail-archive/
and
  ftp://ftp.dsv.su.se/users/jpalme/draft-palme-newsmail-00.txt
for details), can we agree that Reply-To should be clearly defined to have
the "sender preference" interpretation, and move on?

(BTW, I advise viewing the mailext archive sorted by date rather than by
thread because many of the messages are not threaded correctly for whatever
reason.) 

-- 
Bart Schaefer                                                   Zanshin
http://www.well.com/user/barts                   http://www.zanshin.com