http://bugzilla.spamassassin.org/show_bug.cgi?id=3683
[EMAIL PROTECTED] changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|RESOLVED |REOPENED
Resolution|WORKSFORME |
------- Additional Comments From [EMAIL PROTECTED] 2004-08-13 13:03 -------
I can understand that you don't want a high bayes score to influence which
additional emails get learned as spam, but ignoring that an email has a high
bayes score and allowing it get get learned as ham is not a good idea. This
means that all manual bayes learning from emails that were not caught by the
non-bayes rules slowly gets undone as these emails get repeatedly marked as ham.
A better rule would be to only ignore the bayes score when deciding to autolearn
it as spam, but to not ignore the bayes score when deciding to autolearn
something as ham.
Any chance of getting this changed?
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
