Date: Fri, 13 Dec 2002 08:22:52 -0800 From: Terry Lambert <tlambert2@mindspring.com> To: dschultz@uclink.Berkeley.EDU Cc: akruijff@dds.nl, cls@raggedclown.net, chat@freebsd.org Subject: Re: Spam decisions Message-ID: <3DFA095C.3117915@mindspring.com>
next in thread | raw e-mail | index | archive | help
David Schultz wrote: > I whitelist mailing list mail and off-list responses thereto, when > I can identify it. The idea that ``I don't know anyone in country > X, so mail from there must be SPAM'' breaks down for mailing > lists. I actually think that in general, if a whitelist is > required at all, there must be something wrong with the model. The problem is "recevier filter" rather than "receiver proxy filter". When the mail is accepted into your POP3 maildrop on your behalf by your ISP's SMTP server, rather than giving an SMTP 5xx error code on the address, it validates the email addresses for you. The only way to really deal with this is to change the "DATA"/"RCPT" processing order in the SMTP protocol itself, and institute server side filters. What it really boils down to is that you need to have a means of rejecting at transfer time for delayed delivery, messages for which an address does not exist, based on both envelope and content of the message. You can't really do this, if you OK the address as existing without having the content in hand. The down side is that the misaddressed mail transfer burden will go up; but largely, that's a "sender pays the time" issue, since mail servers themselves tend to be well connected, relative to mail originators/clients. The remaining problem of a "sieve"-like filter on the receiving SMTP server is one of compute overhead: server side filtering increases server overhead. Basically, that means, minimally, a shared classification engine. So... time to jin the SMTP mailing group of the IETF, and fix the protocol, after which the other stuff becomes less important. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3DFA095C.3117915>