From: Karl L. Barrus <barrus@tree.egr.uh.edu>
Date: Tue, 1 Dec 92 09:18:49 PST
To: cypherpunks@toad.com
Subject: Nike says: Just Do It
Message-ID: <9212011718.AA01351@tree.egr.uh.edu>
MIME-Version: 1.0
Content-Type: text/plain



I agree with Mark's letter: we've philosophized enough - let's
implement a digital bank.

The only thing I have to go on is a recently posted summary of a
digital cash scenario - apologizes to the poster, I don't have the
article handy and I can't remember who nicely summarized the protocol.
If that summary isn't good enough, I'll go look for more of Chaum's
stuff.

I'm willing to help pull such a feat off - even if to start I (and
anybody else who wants to help!) take Email and respond by hand.  

The only high precision math routines I have are the rsa programs that
are available @ghost.unimi.dsi.it (see Dave Vincenzetti's (sp?) recent
message on sci.crypt) OR what I used during my crypto class for
working through various protocols: Mathematica.  Automating the
procedure can come later, and I'm willing to work with anybody in
doing that as well.  If initially the transactions are handled by hand
and Mathematica, don't expect a rapid turnaround! :-) The nice thing
about Mathematica is that I can run it remotely and don't need to be
in front of the NeXT to do it - especially since the command line
version is good enough for out purposes: high precision integer math,
etc.

I remember the protocol involved a hash (MD5 is suggested) of an input
to produce a number.  Suggestions?  Would turning the MD5 output into
a hex number be good enough?

Anyway, I'm going to work through the protocol a few times by hand and
post again with more info in a bit...

/-----------------------------------\
| Karl L. Barrus                    |
| barrus@tree.egr.uh.edu (NeXTMail) |
| elee9sf@menudo.uh.edu             |
\-----------------------------------/