From: "Karl A. Siil" <karl@cosmos.cosmos.att.com>
Date: Fri, 26 Jan 1996 21:46:00 +0800
To: cypherpunks@toad.com (Cypherpunks Mailing List)
Subject: Re: Intl. Keysize Limit vs. U.S. Keysize Limit (Was: Re: Hack Lotus?)
Message-ID: <2.2.32.19960126133039.0076f604@cosmos.cosmos.att.com>
MIME-Version: 1.0
Content-Type: text/plain


At 12:50 AM 1/26/96 -0500, Futplex wrote:
>-----BEGIN PGP SIGNED MESSAGE-----
>
It would be just as easy to build the intl. version of Notes with 128-bit+spy
>keys, with 40 bits of truly protected key and 88 bits of espionage-enabled 
>key, and then use straight 128 bits in the domestic version.

Does anyone else find it worrisome that given 24 bits of a 64-bit key, the
encryption may not be as strong as the same algorithm with a 40-bit key
(e.g., for a variable key-length algorithm like RC4)? In other words, I
suspect some algorithms might not derive their protection mechanisms equally
from each key bit. What if guess the remaining 40 bits is as easy as
guessing the rest of xyl******?

Just pondering.

                                        Karl