ietf-822
[Top] [All Lists]

Re: [Fwd: I-D ACTION:draft-moore-mail-nr-fields-00.txt]

2004-09-04 13:57:46

Bruce Lilly <blilly(_at_)erols(_dot_)com> writes:

Assuming that there is such a thing as a "list preference" for replies
to the list, that can be easily achieved by haying the list expander
insert a Reply-To field pointing to the list mailbox if and only if
there is no (author-provided) Reply-To field in the message.

It sounds like you're making the assumption that there are two types of
reply (individual and to-all), and Reply-To should be used to send them to
the appropriate location with the only difference being whether people in
the To and Cc headers are included.  While this may be correct for
individual e-mail, this is manifestly not the case for list e-mail.

There are similarly two different styles of reply for list mail, but
they're a *different* two.  They are exactly analogous to the two styles
of reply to a Usenet posting, namely "private reply" and "public reply."
(And there is, in fact, a third, rarer type of reply, namely the list
equivalent of reply to all that includes any additional people or lists
addressed in the message, but let's put that aside for the moment.)

The use that you're making of Reply-To works fine for individual e-mail,
but when used in the list context, it breaks private reply.  Now, I
realize that there are some lists where private replies are always
inappropriate, but again one cannot generalize from one particular type of
mailing list, and there are many where this is *not* the case.

You could, of course, argue that private reply should be implemented by
ignoring the Reply-To header, and in fact one MUA has already made that
choice, but then you get into all sorts of other trouble since that use is
not anticipated by any mail standard and the MUA cannot adequately
distinguish between personal e-mail (where this is wrong) and list mail
(where this may be arguably correct).

Different people have different priorities, but I personally consider
private reply to be vitally important and any solution that breaks private
reply to not be a solution at all.

-- 
Russ Allbery (rra(_at_)stanford(_dot_)edu)             
<http://www.eyrie.org/~eagle/>


<Prev in Thread] Current Thread [Next in Thread>