Vernon Schryver <vjs(_at_)calcite(_dot_)rhyolite(_dot_)com> wrote:
No, filtering is cheap even if you use Perl schemes
such as SPamAssassin and it's almost invisible with others.
I'll be honest, and admit that I haven't even bothered with those
kind of filtering solutions, so I don't know what their expense would
be.
Why? My filtering involved firewalling port 25, and sending ICMP
"port unreachable" messages to anyone not on a whitelist of IP's. The
expense of doing that has turned out to be too much for me to handle.
(bandwidth, time, cpu power, email from clueless admins complaining
about my machine "attacking" them). I hate to think of what would
happen if I opened port 25 again, and ran DCC or even Perl over every
incoming email. My legitimate mail would *stop*.
That's my point. I don't *care* how cheap it is to run any kind of
filtering software. It's still too expensive. Look at the numbers
Chris Lewis posted for Nortel. A domain I administer gets about as
much spam as a company with 50k employees! In what crazy world is
that acceptable? Why would it be my problem to filter out that
garbage?
With the amount of legitimate email I get, I should be able to use a
486-66 on a 56k dialup to handle the traffic. Any expense over that
is *directly* attributable to spam.
I've spent $1000's upgrading commodity hardware to handle spam. I'm
tired of buying a new, faster, PC every year, just to handle the
increased load. It's not my problem, and I want it to go away.
Maybe you have $1K every year to spend on upgrading your commodity
hardware. I don't. And I don't see why I should have to.
Alan DeKok.
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg