From: David Honig <honig@otc.net>
Date: Thu, 25 Dec 1997 01:39:44 +0800
To: cypherpunks@toad.com
Subject: computer modelling of cryptanalytic thought processes (references)
Message-ID: <3.0.5.32.19971224092430.007dc490@otc.net>
MIME-Version: 1.0
Content-Type: text/plain




I was reading the Military Cryptanalysis Field Manual 
at ftp://ftp.ox.ac.uk/pub/crypto/cryptanalysis/ basic_cryptanalysis.ps.tar.gz
and was reminded of a book I had partly read a while ago, Doug Hofstadter's
_Fluid Concepts & Creative Analogies : Computer Models of the Fundamental
Mechanisms of Thought_
(http://www.amazon.com/exec/obidos/ISBN=0465024750/r/0509-6726066-321937).

'Hofstadter and his colleagues created a variety of computer programs that
extrapolate sequences, 
apply pattern-matching strategies, make analogies, and even act "creative."
'  

For instance, he developes a 
model that solves a type of newspaper-puzzle cryptogram.

The interplay between levels of abstraction ---recognizing words in the
plaintext language, 
and reconstructing the key together from statistical and other constraints
(Ch. 4)--- is what's 
similar between Hofstadter's model and what the field manual describes.

Caveats: Hofstadter's book is very dry.  The Field Manual only mentions
that computers may be useful,
and gives advice that may have been useful 70 years ago.





------------------------------------------------------------
      David Honig                   Orbit Technology
     honig@otc.net                  Intaanetto Jigyoubu

"Windows 95 is a technologically complex product that is best left alone by
the government..."
 ---MSFT Atty B. Smith