>Scott: we are thinking also to not only search for tipical spam-content
>but also for tipical non-spam-content and give them a negative weight as
>you mentioned in a posting yesterday. But we are not sure if there are
>any restrictions/rules that we should respect with the
>junkmail-weighting-system.
Declude JunkMail does have negative weighting -- both actual negative
weights (for example, you could have E-mail failing the MAILFROM test have
a weight of -5, although that isn't a good example), and "passing" weights
(for example, you could have it so that all E-mail that is NOT listed in a
custom blacklist file could have a weight of 5).
You could use this with the upcoming system in 1.55 where the external
program can return a weight, by setting the default weight to -10
('TESTNAME external weight "C:\IMail\Declude\program.exe" -10 0'), and then
adding 10 to whatever weight you would otherwise return (if you were going
to return a weight of 5, you would return 15). That way, you could have
the program return negative weights up to -10.
-Scott
---
[This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)]
---
This E-mail came from the Declude.JunkMail mailing list. To
unsubscribe, just send an E-mail to [EMAIL PROTECTED], and
type "unsubscribe Declude.JunkMail". You can E-mail
[EMAIL PROTECTED] for assistance. You can visit our web
site at http://www.declude.com .
