A few days ago someone had a question about spam filtering problems but it drifted off into a conversation about untrimmed replies, so I thought I'd raise it again.
BayesIT had been working pretty well for me up until upgrading to the current TB version with the built-in plug-in. I tried to bring it up to speed by running through all my old junk mail as well as a collection of good mail, but it still wasn't catching much after about a week of operation and I was wondering what was wrong. I finally set my "score greater than" box to 5 out of 100 (!), and now things are looking up--it's catching maybe eighty percent of the spam with no false positives. But 5 seems an awfully low number to use, and I'm always checking for good mail showing up in my junk box. Am I doing something wrong or is this how it works? Will I be able to raise the score over time to give myself more peace of mind? And just what do those choices about Average, Maximal and Minimal really mean? I've fooled around with changing the settings but can see no difference in performance. -- Using The Bat! Version 2.10.03 under Windows XP 5.1, Build 2600, Service Pack 1 ________________________________________________ Current version is 2.10.03 | 'Using TBUDL' information: http://www.silverstones.com/thebat/TBUDLInfo.html