One approach that I've thought casually revolves around a feature in
Sendmail that would open an SMTP connection back to the sending host
(assuming the sender had smtp services) and test it for relaying mail back
to you. You could then accept or reject mail from that host based on the
finding. Not that I've thought about it in depth, but you could hash status
for incoming mail hosts after probing them. But this would encourage people
to close host relays.
There are still things it won't address. Among them are the spammers who
initiate their own SMTP sessions from things like ppp dialups.
Just some thoughts.
Allen
-----Original Message-----
From: James M Galvin [mailto:galvin(_at_)acm(_dot_)org]
Sent: Monday, January 22, 2001 10:45 AM
To: Jacob Palme
Cc: ietf-822(_at_)imc(_dot_)org
Subject: Re: Statistics of "intelligence" in e-mail spams
Conclusion: Based on my statistics between two time periods
whose average date was about one year, both these kinds
of "intelligent" spamming had more than doubled its percentage
of all spams.
My statistics thus supports the idea that spammers are getting
more intelligent.
Well, strictly speaking, I suppose I can accept that your data indicates
at least some intellectual improvement in spammers. I agree with you to
the extent that you want to give credit to spammers just because they
can put your name or email address in the headers or body of the
message.
However, personally, I'm not impressed. The concept of "mail merging"
existed in the paper world long before it was attempted with email, so
all the spammers have shown is a new application for an old principle,
which I don't consider particularly significant.
What I was really after when I spoke of the intelligence of spammers is
the mechanisms by which they execute delivery. "Pure spammers" have
exactly one: find an open relay and exploit. The problem is finding an
open relay is like "shooting fish in a barrel," and this is unlikely to
change in our lifetime. Adding mail merging to this delivery mechanism
is like putting icing on a stale cake: it looks good and "feels" good on
the outside, but the inside is disgusting.
Beyond this point there is a lot of gray area regarding spam. There's
some distance between pure spam and marketing using email, and each
person's threshold for what is spam will vary along that continuum.
Email marketing is certainly becoming "smarter" and if you consider that
spam then spammers will look smarter. I don't.
Jim