Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site cornell.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!mtuxo!mtunh!mtung!mtunf!ariel!vax135!cornell!hal
From: hal@cornell.UUCP (Hal Perkins)
Newsgroups: net.crypt,net.legal
Subject: Re: RSA cryptographic algorithm patented?
Message-ID: <3154@cornell.UUCP>
Date: Mon, 15-Jul-85 16:53:39 EDT
Article-I.D.: cornell.3154
Posted: Mon Jul 15 16:53:39 1985
Date-Received: Wed, 17-Jul-85 07:39:43 EDT
References: <9028@ucbvax.ARPA>
Reply-To: hal@gvax.UUCP (Hal Perkins)
Followup-To: net.legal
Organization: Cornell Univ. CS Dept.
Lines: 32
Xref: watmath net.crypt:409 net.legal:1830
Summary: 

from net.crypt:
>  Unfortunately, he did not reference that this RSA Public Key Cryptosystem
>  was patented by the Massachusetts Institute of Technology in 1983
>  (U.S. Patent 4,405,829).  The worldwide exclusive license to this patent
>  was then purchased from MIT by RSA Security Inc., a company founded by
>  the inventors of the RSA algorithm to develop this technology.
>  
>  Because the RSA algorithm has been published in academic journals, most
>  people assume that it is in the public domain, similar to the DES
>  algorithm.  Unfortunately, some people have developed software and
>  other products based on the RSA algorithm without researching this
>  point.  Nevertheless, the patent exists and, in the opinion of our
>  corporate attorneys, will be easily defended.  As RSA Security Inc.
>  paid a great deal of money for the exclusive patent rights, we plan
>  to actively police the commercial use of the RSA algorithm.

Er, say what?

I'm no lawyer, but from my reading of general articles on patent law,
an algorithm is one of the things that specifically CAN'T be patented.
One can patent a gadget (that's how the Unibus was patented), but one
can't patent an idea unless it's "reduced to practice", i.e., implemented.
That would seem to mean that one could patent a box that encrypted data
using the RSA algorithm, but one couldn't patent the algorithm itself.
What's going on here?

(I've aimed this at net.legal also and I think followups will go there
since that seems to be the appropriate place for this.)


Hal Perkins                         UUCP: {decvax|vax135|...}!cornell!hal
Cornell Computer Science            ARPA: hal@cornell  BITNET: hal@crnlcs