ietf-822
[Top] [All Lists]

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

2004-09-03 21:22:03

Keith Moore writes:
That's not in itself an argument against MFT - just an argument against
the notion that MFT provides a reliable "reply to list" feature.

Some MUAs have a ``reply to list'' function that sends a message to the
List-Post address, or to an address configured for this mailbox. But
this is not exactly the desired function. (Example 1: sender not on the
list. Example 2: crossposts.)

Every popular MUA has a ``reply to all''/``follow up''/etc. function.
Mail-Followup-To turns this into exactly the desired function, removing
the demand for ``reply to list.''

Of course, without Mail-Followup-To, the ``reply to all''/etc. function
isn't exactly the desired function. (Example: sender on the list.)
That's why some MUAs introduced the ``reply to list'' function---it
failed less frequently than ``reply to all'' before Mail-Followup-To.

I see no evidence of independent user interest in ``reply to list.''
Users think of it the same way as ``reply to all''---except that it has
a different set of failure modes. Increased use of Mail-Followup-To has
reduced the number of ``reply to all'' failures, and universal use would
eliminate all of the failures; in contrast, there are no proposals to
reduce the number of ``reply to list'' failures.

---D. J. Bernstein, Associate Professor, Department of Mathematics,
Statistics, and Computer Science, University of Illinois at Chicago