Hello qe3ee,
Sunday, June 26, 2005, 4:19:29 AM, you wrote:
q> This is added by one listserver. Excerpt from a neglected mail:
q> X-Spam-Checker-Version: SpamAssassin 2.63-lists.debian.org_2005_05_20_02
q> (2004-01-11) on murphy.debian.org
q> X-Spam-Status: No, hits=2.5 required=4.0 tests=BAYES_70,BODY_8BITS
q> autolearn=no version=2.63-lists.debian.org_2005_05_20_02
q> X-Spam-Level: **
q> Excerpt from another:
q> X-Lycos-AS: 38.00
q> X-Lycos-AV: OK
a) to get things straight: I suppose both of the above messages were
not Spam, or do you mean those were false-negatives ?
q> -Is this what I should try to pre-program my spamfilter to neglect?
Unless your Spam-plugin is giving you false positives, I don't know why you
should do that. Of course I'm a K9-user, I never really tried Bayes-It or
Bayes filter, since I'm using K9 for more than one mailer. So I'm not
acquainted with the way those plugins work.
I use the "X-Brightmail: Suspected Spam" header(field)(1) in a TB!-filter to
collect those messages that were not caught by K9, I do not use it to keep K9
from testing them...
I think (but others may know more about this) you should look for something
like
X-Spam-Status: Yes and
X-Lycos-AS: ?? (more than 50? I don't know at what % a message is
considered to be Spam)
From your earlier message:
,------/ \----
| -Oh! I allmost forgot. I have blended subscriptions to various discusionlists
| in different languages in subfolders. How have you seen this affect the
| filtering efficiency in your cases.
`-----8<-------
For me: not at all...
--
Best Wishes,
Mark
using The Bat! 3.5.0.31
________________________________________________
Current version is 3.5.25 | 'Using TBUDL' information:
http://www.silverstones.com/thebat/TBUDLInfo.html
