Am 21.09.2016 um 17:53 schrieb Sean Greenslade:
As for your spam rejection paradigm, I can't possibly imagine that working well unless you have a very close relationship with every single person who emails you. If I send my resume to a job recruiter and they get a bounce when they email me back, I highly doubt they're going to bother to call me up and tell me my email system is broken. My resume's going in the trash and they're moving on. Just because you haven't received any calls doesn't mean there's no problems...
it's absolutely no problem to outright reject high scored spam and tag the likely spam stuff - BUT the prerequisite for doing so is to collect bayes data, watch how the systems operate and after it's classification is proven good and all sort of scores are adjusted decide what is the safe reject score
the problem of the OP is that he starts things the other side round and first reject without good evidence and don't have anything to make the system bullet profe because it's rejected
when one starts which dangerous rules like reject based on message-id, not realize that his balcklists are not working and bayes don't work this system is *not* pruction ready at all
