OSEC

Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com
 
Subject: Re: I would like to see some big integer examples of GMP code
From: Peter D. Junger (jungersamsara.LAW.CWRU.Edu)
Date: Tue Jul 25 2000 - 07:59:18 CDT

Bill Stewart writes:

: >On Sat, 22 Jul 2000, Peter D. Junger wrote:
: >> Right. And I have thought of trying it in Lisp. But at the moment for
: >> a legal argument about the crypto export regulations, I need to produce
: >> a strong crypto program---but not necessarily a usable one---in the form
: >> of compiled object code for which the source is not available. I'm lookin
: g
: >> for the object code equivalent of Adam Back's RSA in 2 lines of Perl.
:
: I'm confused. Are you trying to show them an algorithm for which
: you know the source code but aren't telling them? Or are you trying to
: show them
: an example of what a commercial software developer might do,
: where they have the source in their development shop but their
: customers only get the compiled binaries?

I'm sorry. I did not make myself clear. I want this example for use
in connection with legal arguments to be made in an article and perhaps
in litigation, not for demonstration to my students. The issue has to
do with the distinctions that lots of lawyers try to make between
source code and object code, which I don't think makes any sense.

So what I am after is a very short chunk of object code that I can
contrast with source code that does the same thing. For these puposes
Perl is considered source code not object code. So the idea of
converting Perl to C source code to C object code is quite attractive.

: Depending on your copyright, RSA patent compliance, and writing-your-own
: needs,
: the RSAREF source code is widely available, and there are probably
: example programs showing how to use it. There's also Bruce Schneier's
: Applied Crypto book, which had lots of software implementations
: available on a separate disk, which Phil Karn used with his court cases.
: If it's not too long, the PGP 2.6 source code could be another good example.
: The old Unix crypt program also works, if you don't need a program you
: really wrote yourself - source code should be publicly available.
: It's very weak crypto, but that's ok.
: There are also DES programs in the popular FTP archives, like ftp.ox.ac.uk.

I have most of these items, and could well make use of them, but I do
also need examples for which source is not available, and, because of
copyright issues, and because I like to know what is going on with
the examples that I use, it would help if the example were something that
I wrote or adapted.

:
: For a really ugly evil thing to do, there's a perl2c program that
: transforms PERL code into C. Feed it the 2-lines-of-perl :-)
: That's probably a bad idea, since that version of the RSA program
: formats its output to feed to the "dc" multiple-precision calculator program,
: but there's also a pure-perl RSA implementation that's not much longer
: and is a lot cleaner.

I did write a Perl implementation that only seems to work with very small
large integers. I could try to see if that will convert into C. But if
I try that I assume that the converter will have to know about some C
big integer program.

The legal issue that I am now concerned with is the fact that in my
suit on the constitutionality of the export regulations, the court of
appeals for the 6th circuit has held that source code is protected as
speech by the first amendment and I need ammunition for the argument
that the holding should also apply to object code.

Part of the problem is that most lawyers who are concerned with these
matters, whether academics or not, seem to think that object code
is in binary one's and zero's, while source code is not. So I want
to come up with binary representations of both a small chunk of source
code and a small chunk of object code that do the same thing but that
are different programs.

That is probably a nonsensical project, but if it is, then one can say
that I am trying to fight nonsense with nonsense. 

--
Peter D. Junger--Case Western Reserve University Law School--Cleveland, OH
 EMAIL: jungersamsara.law.cwru.edu    URL:  http://samsara.law.cwru.edu   
        NOTE: jungerpdj2-ra.f-remote.cwru.edu no longer exists