From: "Peter Trei" <trei@process.com>
Date: Thu, 5 Jun 1997 00:33:07 +0800
To: cypherpunks@toad.com
Subject: DES [Was: Re: Webpage picketing?...]
Message-ID: <199706041540.IAA14829@toad.com>
MIME-Version: 1.0
Content-Type: text/plain



(maybe) TruthMonger (maybe not) wrote:
>[lots of editing]
>   I used Peter Trei and others' software for the DES Challenge,
> but I posted a suggestion to the list that perhaps their programs
> contained subterfuge designed to thwart others in their attempts
> to find the secret keys. Was I spreading FUD? The responses to
> my post gave me much more information about the processes behind
> the software than any direct inquiries I had made about their
> workings. (And they reinforced the fact that "blind trust" in
> their programs or anyone else's is foolish.)
 
>   I could put a pointer to Peter's software which opens an
> adjoining window which warns that it may be compromised as a
> result of ulterior motives. The fact is, however, that Peter
> might himself put a disclaimer on his page saying, "If you
> can't read code, then you can't be certain of what my software
> is doing." (As a matter of fact, I believe his documentation
> mentions that there is no guarantee his logic is not in error.)

TM:
    Yes, you were spreading FUD, of a particularly stupid kind.

    I've always distributed my DES cracking software with full 
source code, as well as a compilied Wintel version. Anyone 
who wants to can examine the source and compile it independently 
if they worry about a mismatch between the source and the 
included binary. 

    Sven Mikkelsen (http://inet.uni-c.dk/~svolaf/des.htm) has 
now published full source for BrydDES was well. It runs faster 
than my code, but is only for x86 (mine includes generic 'C').

    If you want to participate in a coordinated attack, and don't
mind using a binary-only client, see. 
http://www.frii.com/~rcv/deschall.htm. This effort claims to have
already searched 15% of the keyspace.

    For overall info on the DES challenge, see
http://www.eskimo.com/~joelm/des.html

Peter Trei
trei@process.com