What really worries me is the growing number of messages between 4.5 and 5. Many of these already have a Bayes score of 90+.
Agreed, this is why rule development for SA always has been, and always will be, an arms race.
Spammers are aware of SpamAssassin. Spammers are tuning their mails to try to take advantage of it.
If they were not, SA 1.0 would have been the end of SA development.
Unfortunately developing good rules is a very labor-intensive task. The 2.5x family is, at least in my opinion, slightly weaker in the rules department than some of its predecessors. This weakness in the rules is the result of the effort involved in adding bayes, but bayes makes it ultimately a much better spam combatting tool.
I think that in 2.6x and 2.7x there's going to be an increased focus on adding more rules, pruning out the bad ones, and trying to make existing rules hard to forge.
I also think that SA needs to start having more rules that target and penalize abuse attempts, like the "mutiple mua" rule.
To be honest, I'm hoping to have some time available to put in and help out the SADevs work on this kind of stuff. But lately I've been pretty busy with other things (as is evidenced by my lack of time to do a last-pass on the rule-writing guide... soon, I mean it!)
------------------------------------------------------- This SF.Net email is sponsored by: INetU Attention Web Developers & Consultants: Become An INetU Hosting Partner. Refer Dedicated Servers. We Manage Them. You Get 10% Monthly Commission! INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
