Re: Software Patents

[Bruce Lewis]

hollaar@antitrust.cs.utah.edu (Lee Hollaar) writes:

> In article <nm9d4zr3jgf.fsf@grumpy-fuzzball.mit.edu> Bruce Lewis 
> <brlewis@users.sourceforge.net> writes:
> >Yes, but if Congress declines this invitation, then one should assume
> >that the Supreme Court's opinion that software is not patentable is
> >good-enough law.
> 
> The Supreme Court *never* said that.  Even in _Benson_, the Court said:

You say "Even in Benson."  Are you implying that Benson was the most
definitive ruling against software patents, and that subsequent rulings
left more loopholes?  I do not think that is the case.  I think Benson,
Flook and Diehr all carry one consistent message.

>     It is said that the decision precludes a patent for any program
>     servicing a computer. We do not so hold.
> 409 U.S. at 71.

They did not so hold because that was not the question before them.  The
question was whether the particular BCD patent before them was a
"process" according to statute, and they held that it wasn't.  Their
reasoning in answering this question has generally led people to believe
that any "generalized formulation for programs to solve mathematical
problems of converting one form of numerical representation to another"
is not patentable, i.e. algorithms for general-purpose digital computers
are not patentable, but programs for special-purpose specific computers
might be.  Every "software patent" example I've seen come up in online
discussions, including misc.int-property, has been of the former type.

In Parker v. Flook the court clarified that you couldn't patent an
algorithm by drafting a claim with non-novel elements that weren't
algorithms would not make it patentable.

In Diamond v. Diehr they explained the law the same way, but said the
patent was good because there were non-algorithm parts of the invention
that were novel, and interactions between the computer and the rest of
the system that were novel.

They didn't always say that you need to look at the non-software parts
of the invention and the ways software and non-software parts are
combined.  More frequently they said you need to look at the invention
"as a whole".  Of course, this latter wording is open to various
interpretations, including ones that turn Benson and Flook on their
ear.  That's the way software patent advocates always read it.  However,
there isn't any indication that the court intended to overturn its
precedents.