1994-03-28 - Re: Ames/clipper compromised?

Header Data

From: Jim Gillogly <jim@rand.org>
To: cypherpunks@toad.com
Message Hash: affc9fa88da047e80dc0ae569579e5d16c1c045caec8840f209fa8c7220ad3b8
Message ID: <9403282159.AA10013@mycroft.rand.org>
Reply To: <199403282050.MAA03159@well.sf.ca.us>
UTC Datetime: 1994-03-28 21:59:39 UTC
Raw Date: Mon, 28 Mar 94 13:59:39 PST

Raw message

From: Jim Gillogly <jim@rand.org>
Date: Mon, 28 Mar 94 13:59:39 PST
To: cypherpunks@toad.com
Subject: Re: Ames/clipper compromised?
In-Reply-To: <199403282050.MAA03159@well.sf.ca.us>
Message-ID: <9403282159.AA10013@mycroft.rand.org>
MIME-Version: 1.0
Content-Type: text/plain



> Brian D Williams <talon57@well.sf.ca.us> writes:
> I remember awhile back someone posted some clipper documents that
> were released under FOIA as I recall. The thing that struck me was
> that the NSA was refering internally to clipper as "The Trapdoor
> chip." Why refer to it as such if there is no back door?

Those letters made it clear the "trapdoor" was the escrow, and the internal
debate was over whether the existence of the escrow would be made public.
So far it's been NSA's consistent public position that the escrow is the
only way in... and from the FOIA, that's evidently what they're telling the
President also.

Most days I'm pretty sure I believe that there aren't any known gotchas in
the Skipjack algorithm.  If they can really get the escrow, it's ever so
much cheaper than doing real cryptanalysis.  As Carl Ellison and others
point out, that's really one of the big dangers -- if LE doesn't have to
break Skipjack to read the traffic, neither do the attackers... and
breaking the escrow is probably much cheaper than breaking the algorithm.

My position is that Clipper is iniquitous whether or not there's a[nother]
trap door.

	Jim Gillogly
	Mersday, 6 Astron S.R. 1994, 21:58





Thread