http://bugzilla.spamassassin.org/show_bug.cgi?id=2910
------- Additional Comments From [EMAIL PROTECTED] 2004-01-12 11:59 ------- No, not a magic token (in the current use of "magic token" in SA). The suggestion is to take the tests that hit and use them as tokens for Bayes, then let Bayes deal with the probability of spam vs ham, etc. As I recall, JM tried this and found pretty bad results. ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.
