On Thu, 22 Aug 2019 09:39:31 +0100, Laura Atkins via mailop <mailop@mailop.org> wrote:
>Not sure I understand this point of view. I think everyone should be rejecting >after DATA, if only to stop the abuse of the email address validation >services. For defined spam traps here, we accept the message (although we will 550 every additional account after a trap is first detected) and send it to Rev. Bayes for digestion. The server will quaintly detach and quarantine any hostile attachments. Thereafter, the sending IP address is dropped [back] into the ban list -- 1440 to 21,660 minutes, depending upon whether (or how often) we have seen this perpetrator before. >> The thing with spamtraps is that they should not be in your DB in the first >> place (especially pristine ones) or should have been trimmed from your DB a >> long time ago (back when they went from a usable user address to a bouncing >> address before being reactivated as a spamtrap). > > >This is a very limited view and one which fails to account for the realities >on the ground. > >The reality is spamtraps (pristine, recycled, whatever the term du jour is >regularly end up on lists because people submit addresses as part of their >online existence. Sometimes they give the wrong address. Sometimes thats on >purpose (I want your thing but dont want mail from you) and sometimes its an >accident (I dont know my address or I fat finger something). Some of those >wrong addresses have never been used before. Some of those wrong addresses >belong to third parties. Some of those wrong addresses used to belong to >someone but were abandoned some period of time ago. A goodly number of the traps here began receiving spam within a few hours of the domain honet.com being registered in 1997. Others were made up by somebody (the Long Family, e.g.) and others were created by some ratware appending various characters to an otherwise valid address, and then sold on. Then there is the bucket full of tagged addresses stolen from (or sold by?) some online vendors. And, of course, the addresses scraped from http://www.honet.com/Nadine. On average about 220 IP addresses walk into this propeller every day. So sad. There are several hundred of these trap addresses, and for ESPs I have confidence in, the complete list is available for use in detecting bad-faith data uploads. The original "Nadine" address is treated differently: it forwards to places like SpamCop. In our experience, if you mail to addresses that haven't engaged (subscribe, open, click) in the previous 90 to 180 days, there is a growing tendency for your IPs/domains to be classified as spammaceous and dealt with appropriately. >I dont know who came up with this whole pristine and recycled spamtrap >thing, but the terminology has not brought any clarity to discussions. In '1984' there's Newspeak. Since 1995, there's been Spamspeak. Clarity in discussion is to be avoided at any (reasonable) cost. mdr (an extremely disturbed nerd, as our old friend Sanford would have it) -- "Honest folk do not wear masks when they enter a bank." -- Unspiek, Baron Bodissey _______________________________________________ mailop mailing list mailop@mailop.org https://chilli.nosignal.org/cgi-bin/mailman/listinfo/mailop
