From: wichita@cyberstation.net
Date: Wed, 11 Dec 1996 01:30:14 -0800 (PST)
To: "Igor Chudov @ home" <ichudov@algebra.com>
Subject: Re: Ignoramus Chewed-Off on IPG algorithm
In-Reply-To: <199611301953.NAA14436@manifold.algebra.com>
Message-ID: <Pine.BSI.3.95.961211030715.11832F-100000@citrine.cyberstation.net>
MIME-Version: 1.0
Content-Type: text/plain




On Sat, 30 Nov 1996, Igor Chudov @ home wrote:

> Igor Chudov @ home wrote:
> > 
> > Let's go on, to the description of the "scrambling tables" and 
> > actual encryption.
> > 
> > He uses three tables, DIFF, DISP, DETR, each containing 4096 elements.
> > DISP is randomly generated (or so I understand his term "prescrambled"),
> > DIFF is a random transposition of DISP (same values as in DISP, but in
> > another order), and DETR, again, is filled with some random data.
> > 
> 
> Correction: by "scrambling" Don means transposing elements of the 
> table containing 4096 numbers 1-4096.

Yes, but using the algorithms set out at the web site and our own
8192
byte keys, using the timing of keystrokes. Thus, we have randomized
them, in a manner similar, but far more complex,  to what Dr. Rivest did
in his systems, and what I have done previously at NSA. 
 > 
Only the DIFF and DISP tables are random transpositions of the numbers 0
- 4095, the DETR table is a random transposition of 16 sets of the numbers
0 - 255, the ASCII values.

> 
	- Igor.
> 
Yes, but if you read the web site, you will find that those are only the
initial values. The user generated key, the most important element, and  
the time and the message number, both of which are transmitted, are used
to further randomize the three tables.

Also, an user can customize their own initial values so that they are
unlike any other set of values.

In these respects, the technique is similar to aspects RC4/RC5, except far
more complex, three tables instead of 1, and 4096 values instead of 256. I
might add that the table lookup techniques are in effect similar to
prime number cipher wheel systems employed by NSA over the years for very
secure encryption systems, except that instead of the clear text providing
an additional variable, CFB, the PRNG stream, the ABC equations, does
that.

Again, the best way to analyze the system is to get a copy and analyze the
results using your own keys. As indicated, we even provide a test
version where you can look at all the intermediate tables, the As, Bs, and
Cs actually used and everything.

With kindest regards,

Don Wood