However if you are having false-crossing problems due to a bad average, see the --add-to-blacklist and --add-to-whitelist options. They are there as a correction mechanism for the AWL, and don't add static whitelist entries as their names might suggest.
As for negative and positive score assignment it is important that you know and understand that fundamentally the AWL is NOT a whitelist NOR a blacklist at all. It is a score averaging system. It pushes the score of an email towards the historical average for that sender. It isn't even aware of wether or not the message is spam at all. Because it is an averager the AWL would be broken and not work properly if it never assigned negative scores to spam messages.
So for example if a new spammer sends me a message that scores a 10 he will establish an average of 10.
If this spammer sends me a message that scores 0, the AWL (with default settings) will add 5 to the score of the message.
If the spammer instead sends a spam with a score of 20, it will subtract 5. But note that this is NOT a bug. This SHOULD happen.
The reverse cases can happen for nonspam email.. (if I have an average of 0, and send a email with a score of -10, it will add 5 to the score of the message, but the resulting score is still negative).
At 12:30 AM 5/28/2003 +0200, guenther wrote:
hey folks,
I still get some SPAM daily from the same domains, that get a negative AWL rating between -2.0 and -7.0 (!).
I never got HAM by them and very likely will never. Learning those messages with 'sa-learn --spam' triggered the BAYES on them, but does not get rid of the _false_ AWL rating.
How can I tell SA to not remember them as candidates for negative AWL?
btw: 'sa-learn' even gets rid of the 'surrounding' mail created by SA, does it? I mean, is it smart enough, to only learn the actual SPAM mail inside the MIME attachment, or am I wrong here?
Still digging through the documentation and searching for some particular answers...
...guenther
-- char *t="[EMAIL PROTECTED]"; main(){ char h,m=h=*t++,*x=t+2*h,c,i,l=*x,s=0; for (i=0;i<l;i++){ i%8? c<<=1: (c=*++x); c&128 && (s+=h); if (!(h>>=1)||!t[s+h]){ putchar(t[s]);h=m;s=0; }}}
------------------------------------------------------- This SF.net email is sponsored by: ObjectStore. If flattening out C++ or Java code to make your application fit in a relational database is painful, don't do it! Check out ObjectStore. Now part of Progress Software. http://www.objectstore.net/sourceforge _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
------------------------------------------------------- This SF.net email is sponsored by: ObjectStore. If flattening out C++ or Java code to make your application fit in a relational database is painful, don't do it! Check out ObjectStore. Now part of Progress Software. http://www.objectstore.net/sourceforge _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
