On Mon, May 26, 2003 at 01:36:09PM -0500, Eric A. Hall wrote:
There is also existent proof that filtering is only "mostly" successful at
even its limited role. There are plenty of problems with false positives,
delayed positives, and so forth, which cumulative conspire to make this
approach less than functional as anything but a stop-gap measure. In that
regard, filtering is analogous to defining ~delayed defeat as victory.
I find filtering quite productive. I use a free Unix gateway system developed
here called MailScanner (www.mailscanner.info) which brings together many
different anti-spam methods and does anti-virus. Since Sugust 2002 it's
filtered out 8,700 spams for me, and 570 virii, with only two false positives.
Sure, it misses a few, but 8,700 deletions would have taken me a while...
Tim