On Thu, 24 Sep 2009 01:00:20 +0200 Karsten Bräckelmann <guent...@rudersport.de> wrote:
> On Wed, 2009-09-23 at 15:54 -0700, Ted Mittelstaedt wrote: > > RW wrote: > > > Incidently the point about backscatter is wrong. The traditional > > > approach of classifying, and then discarding or filing to a spam > > > folder, produces zero backscatter from spam. Backscatter is > > > actually caused by rejecting at the SMTP level - when it's done > > > on the wrong SMTP transaction. > > > > Say what?!?!? > > > > http://en.wikipedia.org/wiki/Backscatter_(e-mail) > > > > "...The recipient mail servers then use the (potentially forged) > > sender's address to attempt a good-faith effort to report the > > problem to the apparent sender...." > > > > There's 2 separate and independent SMTP transactions here. > > > > The first is the spammer to the recipient mailserver. > > > > The second is the recipient mailserver to the apparent sender. > > > > "rejecting at the SMTP level" makes no sense at all in your context. > > It does. What you just described, however, is *bouncing*, a.k.a. > "rejecting after accepting". That's not rejecting -- by the MX, mind > you. It's not just straightforward bouncing, poorly set-up backup mx servers, intermediate forwarding servers, open-relays etc can all contribute to extra backscatter. If you compare it with the tradition delivery approach which produces zero dsn backscatter the claim of reduced backscatter seems bogus to me.
