"Newsweek" Article on Clipper and Encryption

[Eric Hughes <hughes@soda.berkeley.edu>]

I got a call one week ago today (Tuesday May 25th) from Josh Ramo at
the science desk at Newsweek.  I spoke to him for about an hour on the
technicalities and politics of encryption.  He was to my pleasant
surprise quite able to follow a telephone description of how
Diffie-Hellman key exchange works (!) and was quite conducive to my
explanation of some of the less public aspects of the clipper project.

I think we got extremely good coverage in this article.  Here are some
of the aspects involved.

-- Josh mentioned that he had Dorothy Denning on his list of people to
call.  She did not get quoted; I did.  There's significance to that.

-- The pro-crypto quote came first.  Kammer's quote, on technical
matters, not political ones, came in the middle.  The scary ominous
'mandatory standard' quote, from NIST, came last.

-- They did not replay the White House line that skipjack is so much
harder to crack than DES.  I convinced Josh that by iterating DES, the
pracatical security of the underlying ciphers was the same, i.e.
impenetrable.  Thus, no propagation of half-truths.

-- The sub-headline is against false cryptography.

-- The phrase "civil libertarians and corporations" was used, implying
a united front across liberal/conservative lines against this
proposal.  This phrase was extremely clever on their behalf to avoid
specifically mentioning partisan politics.

-- The NSA is protrayed as demanding and coercive.  First they'll deny
government contracts and export licenses, and if that doesn't work,
they'll outlaw it.

-- Cellular phone are touted as insecure, implying that something
ought to be done about that.

-- The sidebar has an example of cryptography four millenia old;
that's respectable.

-- The article does not play up the escrow aspects of the wiretap
chip.  Their simplification, that the government has your key, attains
the root issue without confusion.

-- They mention that the keys wil have to be stored on computers, and
are thus vulnerable.  This a point I made specifically to Josh, and
they took my example of foreign intelligence and *expanded* on it.

--They mention that NIST worked on the algorithm with the NSA.

All in all, I don't think we could have hoped for better.  There's
just about nothing flattering said about the wiretap chip, and plenty
of things against it.  The article is about as anti-Clipper as you
might expect given that Newsweek does not want to appear too partisan
one way or another.

Eric