On 02/11, Henrik K wrote: > method of whitelisting. You can't seriously expect to block on some > attribute that not everyone can or bothers to change (DNS). None of this
Correct. I am not suggesting that anyone block anything based on MTX at this time. I suggest using it for whitelisting (small negative score, not absolute whitelisting) alone until it is more broadly in use. This is clearly stated in the Implementation Sequence: Conservative people use these new tests to reduce false positives: score MTX_BL 1 score MTX_PASS -1 # Only hit if MTX_BL wasn't. score MTX_FAIL 0.001 Except for those who are willing to cause a small number of false positives, like me. I can, and do, score more harshly those emails that do not have MTX records. And the senders get an error mentioning the option of MTX. All the emails that have been hit seemed likely to be spam. For example, this list gets through just fine with these settings: score MTX_PASS -100 score MTX_FAIL 2 As to the problem of freemail, sites that send both non-spam and spam, constantly, I think that necessitates a blacklist that allows you to define a score per domain (of the PTR record of the sending IP). So, for example, you could blacklist hotmail to only negate the benefit of them having an MTX record, so for hotmail, the net result would be 0. It's funny how, for just believing I may have come up with an idea that is new and useful for dealing with spam, I am consistently attacked. Because people often believe that, and they're almost always wrong. I can't blame you, purely statistically speaking, I'm probably wrong. And I assure you that fact has not slipped my mind. -- "Let's just say that if complete and utter chaos was lightning, then he'd be the sort to stand on a hilltop in a thunderstorm wearing wet copper armour and shouting 'All gods are bastards'." - The Color of Magic http://www.ChaosReigns.com
