SFS2000

[Padgett 0sirius]

>Any key that would enable the decrpytion of one ciphertext into two
>distinct but plausible plaintexts would itself be of length comparable
>to the texts (or else the ciphertext would be suspiciously long compared
>with the "plaintext" provided, which would be a dead giveaway). Any such
>key (and OTP fits this nicely) of such a length would not be suitable
>for use as a session key in the accepted sense, because the whole reason
>for session keys is that they are short enough to be en(de)crypted with
>the asymetric keys on a reasonable timescale.

People used to believe that ASCII text could not be executable also - until
my 1994 Christmas Card that is (still think the first 64 bytes of the final
product is the most elegant code I have ever written 8*).

What I am trying to get across is that Internet encryption has evolved the
way it has because people believed their private keys were safe from
disclosure. Change that perception and you will also change the path of
evolution.


      	A. Padgett Peterson, P.E., CISSP: Cybernetic Psychophysicist
 Anti-Virus, Cryptographics, & Antique Radio Researcher
http://www.freivald.org/~padgett/index.html
 mailto:padgett@gdi.net     PGP 6.5 Key on request