The disagreement concerns the existence of bazillions of people whose
free provider addresses are being forged into most spam every day.
I claim most of those people do not exist, unless you think using your
own address in spam is forgery.
As a (very small) provider of free email addresses, I can confirm that
we receive bounces (with the <> sender) to thousands of distinct addresses
every day.
I'm assuming that the recipient lists that spammers use will have a lot of
addresses that generate bounces and that this is what I'm looking at when I
see all
this rubbish.
I looked at a sample (9288) of these bounces:
9247 were to non-existent addresses on our system (likely "spam" bounces)
41 were valid
We notice many invalid addresses like this:
bjbblas(_at_)domain1(_dot_)example
bjbblas(_at_)domain2(_dot_)example
bjbblas(_at_)domain3(_dot_)example
bjbblas(_at_)domain4(_dot_)example
bjbblas(_at_)domain5(_dot_)example
It seems likely that some spamware is using a list of local parts
(or an algorithm for generating local parts) and a
list of valid domains to construct likely looking "senders".
These lists may very well be based on the recipient list in use.
(This is probably really obvious - sorry)
We do have occasional complaints from users regarding delivery failures
with
regard to messages that "I never sent!"
I'm inclined to think that, excluding nuisance forgery (anyone upset a
spammer?),
sender forgery might be largely accounted for by "accident" - and be fairly
low frequency.
However, in the absence of a good (large, up-to-date) spam corpus, this is
all
speculation.
--
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg