http://bugzilla.spamassassin.org/show_bug.cgi?id=2419





------- Additional Comments From [EMAIL PROTECTED]  2004-05-06 13:08 -------
There's still the issue that the Bayes tests and weighting should be applied
against against ruleset #1.

Since Bayes can weight equally between HAM and SPAM, adjusting the weightings of
other rules based on the success of the Bayes tests unnecessarily reduces their
usefulness.  When a spam message comes in that Bayes cannot judge (50%), then
the other rules are often reduced in weight so much that they won't detect it
either.




------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

Reply via email to