Have you ever been curious what happens when you alias root on the PLUG server to your own e-mail address (this is just one e-mail account)?
http://ryansimpkins.com/spamgraph.png Up until now I've been pretty satisfied with dspam. However, now I'm strongly considering slamming my door shut and going with TMDA: http://tmda.net/ Virus filtering, three RBLs, etc. are already in play. The numbers in that graph are after getting through the first line of defenses. I didn't mind dspam when there were a few hundred messages in the quarantine a week. But now there is about that many every day. The dspam filter is about 99% acurate. Which scenario do you think is better: 1) A VIP sends an e-mail. You don't get it because it lands in your SPAM filter. You don't catch it due to the volume of SPAM you get. It is a rare event, but you could miss out on a time sensitive communication. 2) A VIP sends an e-mail. Your system sends an automated reply with a simple step a VIP has to follow to ensure they get past the filter. The risk is that the VIP won't bother reading or following the instructions sent. The benefit, if everything is followed, you don't miss any important communication. You could still have e-mails land in a quarantine box to try and catch the ones that got away. For the TMDA solution I could go through all of my mail folders and generate a massive whitelist of everyone who has e-mailed me. There are about 32,000 e-mail messages in the archive to feed it. Then, the script could run in a cronjob and frequently rescan the list to be sure it stays up to day. That would certainly help those who already have a connection, but what about new contacts? TMDA does a lot of other cool things too. In particular, tagged addresses could be very useful: http://wiki.tmda.net/ClientConfiguration -Ryan /* PLUG: http://plug.org, #utah on irc.freenode.net Unsubscribe: http://plug.org/mailman/options/plug Don't fear the penguin. */
