>No. :-) I don't use DSPAM, so I don't know anything about it. http://www.nuclearelephant.com/projects/dspam/
Looks like an enhanced Bayesian system. Written in C and requires a db, either berkely, mysql or oracle. Judging from it's docs the '--enable-delivery-to-stdout' should allow it to feed scanned data back to MD. Alternatively it could be run after MD by calling it from mail.local as stated by the DSPAM docs. It does not seem to have a milter interface. Would you want to build support for it into MD? Perhaps, according to its' website it's supposed to be better at bayes like scanning than SA. But I am wondering what overhead it will add. _______________________________________________ Visit http://www.mimedefang.org and http://www.canit.ca MIMEDefang mailing list [EMAIL PROTECTED] http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
