ietf-asrg
[Top] [All Lists]

[Asrg] Bounce criteria...

2003-03-30 07:24:23

Should they engage in the conversation of this list (as opposed to
sending spam), then I would say "yes".

How do I know whether Ralsky has subscribed?  Whenever I subscribe to
a mailing list, do you say I should remove all spam defenses in case
Ralsky has already subscribed and wants to send some private non-spam?

No,  but if you send an email to ralsky rather than the list the email
infrastructure should not automatically filter out the reply on the basis
of a block on the IP block Ralsky is in.

Nor should you create bounce messages in this case, its spam. If you 
don't want to read the message fine, but sending autoresponses to 
someone sending you have decided not to read is schoolyard 
nerner-nerner type stuff.

The problem with security is that it is very easy to solve if you
define the problem to be an easy one rather than solving the real
problem. If you don't care about false positives then go ahead and
block all email from freeserve.co.uk, even though pretty much
everyone in the UK who is on the Internet has at least one account 
with freeserve, virgin or another 'free' provider. The reason that
spamers choose these domains as flags of convenience is because
there are so many legitimate users.

The argument I am making here is that email should not be bounced
or discarded on the information from the MTA alone.

Furthermore perhaps people could explain why it is necessary to
have bounce messages in an alleged 'end-to-end' architecture.
I can see the need for the four server model, but why do we
have to cope with people with chains of fifteen relays?

I think that a lot of the problems we are having here are
due to the imposition of a pseudo-intelligent store and forward
architecture rather than true end to end.

                Phill
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg



<Prev in Thread] Current Thread [Next in Thread>