Re: [dvd-discuss] Does software really satisfy the requirements for Copyright?

[microlenz(at)earthlink.net]

On 17 May 2002 at 14:25, John Schulien wrote:

Date sent:      	Fri, 17 May 2002 14:25:05 -0500
From:           	John Schulien <jms@uic.edu>
To:             	dvd-discuss@eon.law.harvard.edu
Subject:        	[dvd-discuss] Does software really satisfy the requirements 
for Copyright?
Send reply to:  	dvd-discuss@eon.law.harvard.edu

> > The only softare patent I come close to agreeing with is the RSA
> patent on
> > encrypting with trap door functions. That was clearly non-obvious to
> somebody
> > competent in the field. It took a PHD Math Professor to discover it.
> And it
> > was truly novel and unique.
> 
> I disagree.   A patent is not supposed to be a prize for cleverness.

While Novelty is a necessary but not sufficient condition for a patent. Utility 
is another. 
So is nonobviousness. It takes more than that to get a patent (or should...or 
did)


> The purpose
> of patents is to promote progress by encouraging disclosure of
> inventions. that
> would otherwise not be disclosed.  The societal value of a patent is the
> placement
> of a valuable trade secret into the public domain, albeit in a delayed
> fashion.
> 
> In the world of cryptography, secret algorithms are considered useless.
> Only
> peer-reviewed algorithms are ever given any credibility by the
> cryptographic
> community.   It is very unlikely that RSA could have ever been
> commercially
> exploited as a trade secret algorithm.  The algorithm would have never
> been
> accepted.
> 

Yes and your argument proves the value of the patent system and the wisdom of 
granting RSA a patent. It forced public disclosure and peer review. It 
encouraged others to work on different schemes.

> It is also unlikely that the inventors of  RSA, being members of the
> academic
> community, would have concealed their breakthrough if patent protection
> were unavailable.   Most likely they would have simply done what
> academics
> have always done -- publish their successful work.
> 

Maybe and Maybe not...but it's only speculative and not really relevant. That 
they were able to take advantage of it. They benefited and so did the 
community. 

> Contrast this with, for instance, a patent granted on an industrial
> chemical
> process -- say an improved method of removing impurities from gasoline.
> Such a process would be extremely valuable as a trade secret -- the
> company
> that invented it could very well set up a purifying plant, and purify
> gasoline
> as an ongoing business, never disclosing the purification procedure.  In
> 
> such a case, granting a patent would benefit society by placing in the
> public
> domain a valuable trade secret that would have otherwise remained
> concealed.
> 
> 
> 

ANd so did RSA do the same. The question here is not if RSA fits the 
requirements for patents that existed at that time (which were more stringent 
than today) but whether an encryption ALGORITHM is patentable since it is an 
algorithm and not a mechanical device. Previous to the invention of the 
computer all encryption devices were mechanical in nature (e.g., Jefferson's 
coding wheels, Hagelin, etc). Now all algorithms are implemented in software, 
gate arrays (close to software) or ASICs (look at VHDL...looks like Ada but it 
isn't).