Vin McLelland wrote:
> Nice article in USAToday, Will!
>
> You might find it useful to note -- and I'm open for correction on
>this from anyone -- that the US Government's Bernstein brief is, I believe,
>the first time the Govt has openly acknowledged that the export control
>issue is all about sigint -- listening to the legal communications of
>citizens and officials of other national, allied and friendly.
There's more brewing on this with PECSENC, if not PEC. Recall
that PECSENC has been directed by the President to come up
with recommendations for a more publicly acceptable crypto policy
by September under the rubric "Liberalization 2000:"
Quote from a proposed Federal Register notice by PECSENC:
The PECSENC has designated an encryption export control
experts' group to evaluate and propose an agenda of plausible,
incremental reforms as early as next year. The experts' group
will consider proposals from the PECSENC, from industry, and
from the public. It will recommend proposals it finds worthy of the
PECSENC's consideration. The proposals will be considered
independently by the PECSENC and modified, adopted, or
rejected as the PECSENC chooses.
This is from a report on the May 14 PECSENC meeting:
http://jya.com/pecsenc051499.htm
There may be more news of this from the PECSENC meeting
today on when the public is to be engaged in this "liberalization"
policy.
What's intriguing is whether PECSENC, now headed by an ex-NSA
honcho, is going to bite NSA's sigint bullet, and recommend that
strong encryption is better for the public interest than natsec snooping,
what with the world now getting its hands on means of strong protection
for conventional telecommunications of text and to a lesser extent
voice.
This would correspond with the CRISIS report of 1996, which recommended
liberalization on strong crypto and the development of other (unnamed)
technologies for snooping and law enforcement.
The rapid advance of technologies for identification, interception and
surveillance other than those for text and voice transmissions could
replace the need for weak crypto.
There are some pretty amazing things being done with Hidden Markov
Modeling to track patterns for identification, based on a survey of
some 300 patents utilizing the invention in a wide host of applications.
