I'm starting to wonder how much effort is going into the back end of SpamAssassin and if there's even any more developers working on it. I've found numerous 'false positives' in both 3.3.x and 3.2.x - and they come with varying degrees of penalization via points. Now being one who sends a lot of GI I'm constantly monitoring my seeds and pulling headers to look at what server/IP/domain combination is doing what. Even with ESPs who mitigate complaints and monitor space as well as regularly delist IPs, I still see these 'false positives' coming in. What frustrates me more than anything is the blanket response from the ESPs "it's a false positive, X is not listed and anyone using that version won't block your email"...well, I wanna call bullshit. If the email is catching points, whether it be from real or false positive rules being triggered, what the fuck does it matter? If someone has their inbox protected by filtering off anything above a 1.0 let's say, and some fucked up false positive gives me a 2.1 pt penalty, I go to junk or - worst case scenario - I get blocked all together. How would you combat this? What can you possibly do to over come a false positive rule being triggered by an otherwise super clean creative? I'd like to know who's seeing the same thing and start putting together a list of these false positive rules so maybe we can brain storm on how to work around these fuckin' things. :stickyman:
hey bro, body Phrase hey bro, FB_HEY_BRO_COMMA 3.099 2.783 3.099 2.331 http://spamassassin.apache.org/tests_3_2_x.html
If you're consistently seeing the same rules flagged, why not adjust your creative until you resolve it? It's trivial to run thousands of tests against spam assassin in a few seconds..
What rule is giving you the false positive? I thought you were kidding at first then I looked at your link. That has to be the funniest rule I've ever seen. It actually made me chuckle.
