Training the Octopus
For some reason, hunting down spammers and torturing them to death is still illegal. Go figure.
As an alternative, I finally gave in and installed POPFile last night. Ever since I first read Paul Graham's A Plan for Spam, I've been meaning to hunt down and install a Bayesian spam filter. POPFile is a statistics-based automatic e-mail classification program, and said classifying can include recognizing spam and dumping it into its own little oubliette. Eventually, that is; it takes time to learn your particular classification scheme. Once POPFile has classified a sufficient quantity of mail, it's supposed to do a remarkably good job, not only of disposing of spam, but also of sorting your mail in useful ways.
Installing was surprisingly easy, at least on Windows; it's a pretty standard self-extracting installer. I wasn't expecting it to have such an easy time sorting through my mail settings and reconfiguring Outlook Express (I know, I need a real mail client) to pass incoming mail through the filter. I was expecting to have to do a lot more manual configuration. Translating the filtered e-mail into OE's folders required a bit more effort, but didn't take too long.
The difficult part will be training the program; it's going to take a while for me to get enough non-spam e-mail to build a sufficient statistical database to classify things properly. I'll just have to see how it works out over time.
I also read Paul Graham's "A Plan for Spam".
I tried more spam filter for Outlook. Now I use SpamBully for Outlook. I think, actually the best bayesian spam fighting and stopping product for Outlook I have seen is the one Spambully from http://www.spambully.com. I use it now for Outlook. I must say, the program is easy to use and is extremly effective at killing almost all of that bad spam. Plus they have some cool features like bouncing and challenge.