[1293] in cryptography@c2.net mail archive
Re: ciphers - patented then 'released'
daemon@ATHENA.MIT.EDU (Steven Bellovin)
Thu Jul 31 12:12:25 1997
To: Rodney Thayer <rodney@sabletech.com>
cc: cryptography@c2.net
Date: Thu, 31 Jul 1997 11:01:23 -0400
From: Steven Bellovin <smb@research.att.com>
I gather DES was originally patented (by IBM?). Now some
people are looking at CAST, which is apparently patented but
being freely shared by Entrust. Does anyone know where there
are references to the DES patent, and/or discussion of it's
being released for use without a fee?
You're quite right -- IBM did indeed patent DES, but made it freely
available. The latter was, if nothing else, a NIST-imposed condition --
they were only going to give their blessing to a freely-available cipher.
I'm not certain of the patent number -- a quick rummage through my
files shows two candidates, 3,958,081 and 3,962,539, and I don't feel
like wading through them to see if either is it.
What I'm looking for is documentation to show CAST is freely
usable, and as a comparison I am wondering what was done to
make DES freely usable.
What IBM did was to publish a note in the official Patent and Trademark
Office Gazette that gave everyone a royalty-free, non-exclusive license
to use DES. Technically, there was a catch -- as I read it (in conjunction
with a patent lawyer), DES was only free *if* you used one of the four
official modes of operation! As far as I know, they never tried to
enforce that...
Anyway -- given the legal situation (that NIST required that the
standard cipher be free to all comers), IBM needed a nice, formal,
legal mechanism to make this grant irrevocable. The Gazette publication
was used. Any patent lawyer can show you how to find the statement
I'm referring to.