ietf-openpgp
[Top] [All Lists]

Re: Subject: Re: Legal issues in implementing OpenPGP

1998-05-05 09:55:35
From: Paulo Barreto <paulo(_dot_)barreto(_at_)unitools(_dot_)com(_dot_)br>
To: Jim Gillogly <jim(_at_)acm(_dot_)org>; ietf-open-pgp(_at_)imc(_dot_)org
Subject: Re: Subject: Re: Legal issues in implementing OpenPGP
Date: Tuesday, May 05, 1998 7:59 AM

At 12:55 PM 98/05/04 -0700, Jim Gillogly wrote:

SHA1 was designed by NSA and has no intellectual property restrictions.

Hitachi claims to have a patent covering SHA1.

This seems rather odd.  Is it on a particular hardware implementation 
of SHA1?  Do you have a patent number for this?

HAVAL's reference implementation is copyrighted; no charge for use, but
 the algorithm author would like to be notified.

My corrected public domain implementation is now available at
<http://www.nw.com.br/users/pbarreto/haval.zip>, but I think it's a good
idea to notify Dr. Zheng of uses of Haval.  On the other hand, there's no
published analysis of this hashing function (yet), so I'd be somewhat
reluctant in using it.

DSA may probably be used freely, but [Schnorr claim??].

This issue seems to have been already settled in court; DSA may be freely
used.

I don't think it ever went to court, but no one seems to be getting sued
for using DSA...


Elliptic Curve [??]
ECDSA [??]

There are endless patents regarding elliptic curves (sigh); take a look
at
the P1363 site <http://grouper.ieee.org/groups/1363/index.html>


ECDSA in particular is not patented.  Certicom has patented various
hardware tricks to implement EC over GF(2^m).  Certicom also has
a patent on the MQV variant of Diffie-Hellman which can be used over
EC, but has granted a royalty-free license for its use.

Regards,
Rich