----- Original Message -----
From: "J.D. Falk" <jdfalk(_at_)yahoo-inc(_dot_)com>
To: <ietf-dkim(_at_)mipassoc(_dot_)org>
This means that it is foolish to assume that the mere presence of a
DKIM signature or any other technical detail means anything good about
a message. This is true whether it is people or software making the
assumption.
If anyone can turn this into a short (but still accurate) slogan,
I'll arrange for a small run of t-shirts.
Yeah, about 2-3 t-shirts is all that is needed. <g>
Having experience in AI, diagnostic and expert systems for a large range of
applications from DoD, ARPA, Robotics, Nuclear, and in the business world,
it is often best applied in monitoring and detecting failure; deviations
from the physics, emperical, including expert knowledge.
If you expect something to behave a certain way, the deviation from that
expectation has significant value. Most control systems work this way.
But with the low tech world of email, DKIM and SSP, we are mixing apples and
oranges.
Mail Content analysis or the true intent of the mail author has nothing to
with the mechanical protocols in question. The proposed DKIM-BASE protocol
has an expected mode of operation. The deviation from that expectation is
what is being studied. That is what SSP is about.
It is certainly feasible for the negative results and process entries from
SSP and/or DKIM-BASE validation to be feed into a neural net model for
example.
The arguments here by some to be that the deviations from expected behavior
has no significance.
--
Hector Santos, Santronics Software, Inc.
http://www.santronics.com
_______________________________________________
NOTE WELL: This list operates according to
http://mipassoc.org/dkim/ietf-list-rules.html