Vernon Schryver wrote:
From: Andrzej Filip <anfi(_at_)Box43(_dot_)pl>
...
For me /dev/null redirection of messages "classified as spam" breaks RFC
(at least spirit of it).
...
We all agree that even marginally legitimate mail needs a DSN
and that SMTP 5yz status codes are better than DSNs.
Our views get closer :-)
I am trying to convince you and other participants that 5yz MUST be used
to reject spam by MTAs. I know that "real life" will force some
exceptions from the above general rule but we disagree deeply about what
justifies such exceptions.
However, it is all and always the responsibility of the people running
the SMTP server. Mail senders and third parties have no standing to
complain if mail targets send some or all mail to /dev/null. They
are free to do that with software in MTAs and MUAs or manually by just
hitting delete. Whether they act wisely or foolishly, or whether it
is a mistake or a careful wise or foolish plan, it is the right and
responsibility of the mail target, not the mail sender.
I can reluctantly agree with you in case of a company running its own
email server providing mailboxes to "this company" employees only. The
possible and unavoidable mistakes will hurt the company itself. I will
fight such approach in case of every ISP providing service to general
public.
Besides the recently mentioned example of AOL choosing /dev/null
instead of DSNs, there are other cases where automated /dev/null is
the best choice for mail. For example, I have more than 1000 spam
traps of invalid (almost all have never been valid) addresses. Any
and all mail sent to them is unsolicited. Mail sent to them is spam
with probability larger than 99.999%, which is far better than the
technical reliability of email today or in the good ol' days.
Sorry, I am unable to accept reliability above 99.9% without proof.
How have you measured and VERIFIED the accuracy ?
If mail to those addresses generated DSNs or 5yz responses, some spammers
would stop sending to them. That would hurt my spam filtering efforts.
Thus, it is best that mail to those addresses goes to /dev/null
(except for representative samples that are archived on CDROM so
that I can respond to lying spammers demanding changes to my blacklist.)
I wrote that the recipient has the right to send any message to
/dev/null, I have not requested any exceptions for spamtraps (The
spamtrap is the recipient).
--
Andrzej [pl>en: Andrew] Adam Filip http://www.polbox.com/a/anfi/
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg