Re: [Declude.JunkMail] Question about Some Spam Tests

[Matt]

Andy, They're currently listing 9 servers on the same block.  At least two of these machines were listed earlier this week as well.  According to SpamCop's statistics, most of these machines are sending out around 99.9% non-reported spam.  The subjects from the samples appear to be primarily small time senders who are E-mailing from their own address books.  It's not working. I do weight, I'm not sure how I might have given a different impression.  I weight it the same as you in fact.  The messages that I caught both FP'd on BASE64 because of the way that AOL handles attachments/how Declude assesses hits for BASE64.  I probably passed almost all of these FP's, though I don't like seeing personal E-mail scoring so high, especially when coming from the largest ISP around. If this keeps up for another several days, SpamCop drops another point and maybe more.  This shouldn't be happening, and in general, it hasn't as far as I could tell, but now it is reproducible and widespread at least with AOL's mail servers.  This leads me to believe that something is broken. Matt Andy Schmidt wrote: Hm:   "In the past 772.9 days, it has been listed 19 times for a total of 3.1 days"   Sound's like it works as designed - because of the volume of legit mail, it is only listed for a few hours each time.  A total of THREE days out of roughly 800?   Again - why not use a weighting system? If this AOL machine is listed by SpamCop but is used to send AOL mail, then what other tests is legit mail failing to cause it to be held or deleted?    Im my case, SPAMCOP is 70% of the "hold" weight - and I have yet to see legit AOL mail being reported as "lost". Best Regards Andy Schmidt Phone:  +1 201 934-3414 x20 (Business) Fax:    +1 201 934-9206 -- ===================================================== MailPure custom filters for Declude JunkMail Pro. http://www.mailpure.com/software/ =====================================================