I just wanted to add a few comments about what I've seen here regarding
filters.
For one, I don't think any big fan of filters will say that something at
the protocol level doesn't need to be done to reduce and/or eliminate
spam. If he does, well, I won't go into that. I think it's blindingly
obvious that something has to change, and filters are just a small part.
The point regarding the escalating arms race is valid IMO, but not a
good enough reason to ditch the use of filters. I found the point
regarding cost of false positives vs. hitting 'd' very interesting, but
there is more than just a time cost here. I believe (based on experience
dealing with the masses in tech support 8+ years ago) that to see
something as offensive as pornography show up on your screen when you
haven't even clicked anything yet has a very high cost. Higher than an
occasional (if we do our work right) false positive? That's a personal
matter I think.
Anyway, I think filters are here to stay. Any solution we come up with
here is going to take substantial time to tweak and make available. Even
when it is in place, I am sure that some spam will get through somehow,
so we need something to catch what falls through the cracks.
Consider the "everyone uses SMTPAUTH" scenario. Sure, you can see what
account sent the spam, and you can track him down and spank him, but the
damage is done. Filters are a way of dealing with this. Even the large
spam filtering services that use fingerprinting suffer from this
problem; there is always the poor bastard who got the spam before the
spam rule was propagated.
Anyone that doesn't want to use filters can just turn them off, if their
ISP is doing their job right.
Of course filters really have nothing to do with the charter of this
group, so oops, sorry.
Peter
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg