gep2(_at_)terabites(_dot_)com wrote:
To expand, if your incoming mailserver rejects all inbound messages with a=
transient error and notifies the recipient. The recipient can then elect to=
allow any similar messages through if the remote server tries again within=
a set period of time.
This may have to second guess how many times the remote server will retry=
before giving up, and either allow or reject the message with a permanent=
error if the recipient does not express their will in time.
This kind of lame idea is a good example... so you end up increasing the cost
of
handling E-mail, wasting bandwidth and transmission latency on artificially
necessary retries, and requiring new recordkeeping (queueing, etc) on both
the
recipient and sender ends.
Wrong. The sender doesn't need any new recordkeeping, that's how
standard MTAs (not spamware) work already, which is _why_ the idea
works.
The recipient has to do more work, but the benefit is spam reduction.
The amount of delayed mail can be kept quite small.
And the idea about "queueing it closer to the sender"... again, a
large percentage of spam TODAY is sent by spambot zombies. So
you're NOT clogging up SPAMMER uplinks... or increasing THEIR
costs... you're clogging, abusing, and increasing the cost to VICTIM
ISPs and VICTIM users.
Except that in that case, _if_ the ISPs can be convinced they need to
take action, the spam goes away before most of its victims will see
it.
Seth
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg