<br><br><div class="gmail_quote">On Thu, Mar 20, 2008 at 4:02 PM, shuttlebox <<a href="mailto:shuttlebox@gmail.com">shuttlebox@gmail.com</a>> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div><div></div><div class="Wj3C7c">On Thu, Mar 20, 2008 at 8:26 PM, Devon Harding <<a href="mailto:devonharding@gmail.com">devonharding@gmail.com</a>> wrote:<br>
> I seemed to be getting obvious spam lately and the root cause seems to be<br>
> BAYES_00, which is giving the message a score of -2.60. Now I recently<br>
> rebuilt bayes from the starter db from <a href="http://fsl.com" target="_blank">fsl.com</a>. What could be causing this?<br>
<br>
</div></div>It's pretty easy to circumvent Bayes by including some lines of text<br>
from a book for example. Either you can spend time training Bayes all<br>
the time or you can lower the negative scores.<br>
<br>
I have something like -0.5 for BAYES_00.<br>
<br>
--<br>
/peter<br>
<font color="#888888"></font></blockquote></div><br><br>Gotcha, and just change this in spam.assassin.prefs.conf?<br>