On Tue, 28 Sep 1999, Tomas Fasth wrote:
> At a certain level, MUAs have to exchange information with MTAs. One purpose of DSN is to allow MUAs to better
> visualize what is happening on the transportation level. As long as MTAs fail to conform to any kind of
> standard response format, MUA interaction with the user will continue to be awkward in this regard.
I can see the virtue of *a* standard response form. I'm not sure that
DSN is it.
> Couldn't dot forward processing be regarded as a successful delivery?
It could, but that isn't what the RFC says (if I recall correctly) and
also, I think it would be disastrous. Lots of people these days have
multiple forwarding of the type A->B->C->D. You don't want the sender to
get 4 "delivered" messages. (Personally, as a sender, I don't want *any*
such messages.)
> You compared it yourself with a mini
> mailing list, quite correctly I believe. If something goes wrong with the message once it passed a dot forward
> processing, shouldn't the postmaster be notified, not the sender?
We have 6,000 users on this system and 20,000 on our biggest. If one of
our users forwards to some new address that is or becomes broken, I
don't want to know. I certainly don't want to receive a message every
time a message comes in for this user (these people are always on a
zillion mailing lists). Let the senders be told. If they are desperate
they can find other ways of contacting the recipients and telling them
their forwarding is broken.
--
Philip Hazel University of Cambridge Computing Service,
ph10@??? Cambridge, England. Phone: +44 1223 334714.
