On Mon, May 10, 2004 at 03:10:38PM +1000, Daniel Pittman scribbled: [snip] > > And, naturally, you have hard numbers to back up your claim, for > > situations analogous to the Debian mailing lists. Because it would be > > terrible to make sweeping statements like that with no evidence. > > ...and personal experience goes a long way. Testing on around 10,000 > mails showed me that dspam was reasonable inaccurate and would simple > stop catching spam of any sort for about three hours after processing > every false positive report... > > So, YMMV, obviously. :) As with everything. I am going to give dspam a try one of these weeks when I have a little more time and train it with my corpus of spam/ham to see how well it works on the lists I'm subscribed to. regards, marek
Attachment:
signature.asc
Description: Digital signature