On Sunday, February 8, 2009, 2:02:56 PM, Karsten Bräckelmann wrote: > Just got this in an actual replica watch spam.
> This spam was sent using an innocent third party as the fake sender address > who will pick up bounces and misdirected spam complaints. It went out via a > third party host (broadband host in the USA), i.e. stealing someone else's > service. It was sent to addresses harvested off websites at random to people > who have no interest whatsoever in fake watches, stealing their time. > Whoever is sending this spam has no regard for other people whatsoever. > It was good for a laugh, really. :) Until a strange feeling crept over > me, realizing the words... > Confirmed. That paragraph *severely* affected Bayes for me. No Bayes > training with that mail up to that point, results reproducible. > BAYES_99 probability 1.0000 WITHOUT that text > BAYES_50 probability 0.5905 with that poisonous snippet Perhaps helping get the first message through was the desired effect? I get the impression that getting the first one through successfully is a major goal. It seems they expect the later ones to get blocked. Cheers, Jeff C. P.S. Gotta tell Joe about the reference. :) -- Jeff Chan mailto:je...@surbl.org http://www.surbl.org/