In article <[EMAIL PROTECTED]>, Bill Wohler wrote: > I just read http://www.paulgraham.com/spam.html and am convinced that > filters like spamassassin won't keep up and that Baysian filters are > the way to go.
well I've switched from spamassassin to bogofilter (I've trained it with ~600 spams and activated the automatic update mode). I've got less spam emails in my inbox, no false positive by now, no 20mb spamd process and less cpu utilisation. Other bayesian filters that I've tried were spamprobe, spamoracle and crm114. Spamprobe identified almost every mail as spam, crm114 worked best of all bayesian filters, but crashed on the third day with around 70 mails lost in nirvana. I'm currently testing spamoracle to see if it would be a good add-on to bogofilter and after I'm convinced that the latest crm114 version is also stable I will try it again. I've just put some of my test results (well, okay I've just found my results for the first three days ~ 40 mails) at http://3jane.no-ip.org/spam.html. I will adjust that homepage ASAP I've got more mailfilters running, maybe I can find some old mail logs of mine. Andreas -- In Africa, some of the native tribes have a custom of beating the ground with clubs and uttering spine-chilling cries. Anthropologists call this a form of primitive self-expression. In the West, we call it golf. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
