Eric,
I did not say that forgery is not a problem.
I said that it is not essential to spamming.
It is essential to a particular class of spam...
EAH> There are a lot of tools that only stop spam on an incidental basis, but
EAH> are really only useful because of that incidental benefit. Think RFC2821
EAH> syntax checks, greylisting, callback systems, etc., all of which are
EAH> designed to probe implementation weaknesses instead of "block spam", but
EAH> which happen to achieve that goal anyway. MARID would be the similar.
There is an important difference between local algorithms that have
transient utility against a problem, versus global standards that have
no enduring benefit.
The difference is adoption cost in terms of time and effort, notably
when it requires global cooperation.
Perhaps you know of a global standard that was/is designed for
targeting such a narrow bit of behavior and has been beneficial, but I
don't.
Especially when removing the targetted behavior will do nothing to
reduce the core problem. That is, none of this will reduce spam.
As I say, the Internet community is going to be a bit disappointed
that we spent this much time and effort and did nothing to reduce
spam.
d/
--
Dave Crocker <mailto:dcrocker(_at_)brandenburg(_dot_)com>
Brandenburg InternetWorking <http://www.brandenburg.com>
Sunnyvale, CA USA <tel:+1.408.246.8253>, <fax:+1.866.358.5301>