[SAtalk] Bayes balance problem

Hi, I'm afraid my Bayes filter will learn too much spam. I'm getting 90% Spam, so I can't give an equal amount of ham, even if I sa-learn every single ham I have. I'll be facing problems, right? --- This SF.net email is sponsored by: VM Ware

[SAtalk] sa-learn & attached spam, scanning Courier IMAP

s are still untouched. With false positives, things change. SpamAssassin attached the original mail to it's report, so sa-learn would have to detach it and learn the original ham message only, not the SpamAssassin report. Does sa-learn work well with IMAP files and does it lea