Keld Jørn Simonsen <keld(_at_)dkuug(_dot_)dk> writes:
1. I think the best way is just to have the MUA not display the
diplicate mail, if that is what the user wants. I think the extra
traffic and the extra diskspace needed is negliable. Anyway the MUA
could delete duplicate mail too.
I'm afraid previous messages to the list have explained why this doesn't
really do what many people want.
2. Another way could be for the list exploder software to recognize that
a member of the list is also in the To: or Cc: fields, I think this test
is commony already done for the From: field. (metoo) in MTA's. just an
idea. Maybe some people really want to get two copies. If so,
preferences could be set in the list exploder's membership lists.
Likewise, while this has already been implemented in Mailman, this doesn't
do what people want.
The problem is that the people like myself who don't want to receeive the
duplicates generally also want the one message that they receive and see
to be properly sorted with the rest of the mailing list mail, which means
that (in order to do this reliably) it needs to be the copy sent through
the mailing list, possessing all of the apporopriate List-* headers. This
cannot be done reliably using either of the two methods, since they will
both normally suppress the list copy and retain the personal copy, which
is then filed in the person's normal inbox instead of the mailing list
folder (unless they are using less-reliable and less-convenient filtering
methods that don't take advantage of the standardized List-* headers).
I wish it were as simple as just configuring MUAs to suppress duplicates;
anyone who can run procmail can do that trivially, and many MUAs,
including Gnus (which I use) already have that capability. The problem
really is more fundamental -- I don't *want* a personal e-mail message,
and I *do* want the message from the mailing list. Those two messages
differ more than just cosmetically -- consider, for example, that I may be
on the mailing list in digest mode and not want to get even personal
replies until I receive my next scheduled digest.
The mailing list version is the one I want to receive, and it may have
undergone various transformations based on my preferences recorded with
the mailing list management software. I can come up with more examples of
how they may differ if people are interested.
--
Russ Allbery (rra(_at_)stanford(_dot_)edu)
<http://www.eyrie.org/~eagle/>