From: Jay Holovacs <holovacs@styx.ios.com>
Date: Wed, 17 Apr 1996 22:17:23 +0800
To: cypherpunks <cypherpunks@toad.com>
Subject: Re: math patents
In-Reply-To: <m0u8Ysr-0008yhC@pacifier.com>
Message-ID: <Pine.3.89.9604150603.A29831-0100000@styx.ios.com>
MIME-Version: 1.0
Content-Type: text/plain


On Sun, 14 Apr 1996, jim bell wrote:

> . . .  However, I  don't see the 
> basis for patenting what is just about pure mathematics, and RSA is very 
> close to pure math.  The fact that there is a practical use for it is 
> almost  a secondary consideration.   
...
> Had mathematics always been patentable, the patent on that math would have 
> expired at least decades, and possibly centuries ago.
> 
> In any case, I don't think it's unrealistic to suspect that the government 
> was playing games with the patent system due to RSA. 
> That's right, the patent system was actuallly 
> denying the public this system.
> 

This falls into the same category as software patents. According to an ATT
patent attorney, the ALGORITHM is not patented, however building a virtual
machine to implement it is covered by the patent. This is the same as
chemical processes which have been patented historically, you use public
domain chemicals, and normal laboratory procedures in appropriate
sequence, apply an algorithm to create something new. The patent covers
use of this process (algorithm). As such it is not new. 

Crypto is a very TINY part of the picture. General software patents 
(where the big money is) had been submitted for years, and these 
financial considerations apparently drove the picture.

Not that this makes much sense, but then the whole concept of 
intellectual property law is littered with absurdities.

> 


-----------------------------------------------------------------------
Jay Holovacs <holovacs@ios.com>
-----------------------------------------------------------------------
PGP Key fingerprint =  AC 29 C8 7A E4 2D 07 27  AE CA 99 4A F6 59 87 90