🚀 go-pugleaf

John Kasdan and Carl Oppedahl both commented on my summary of case law
cribbed from the 5 September 1989 Official Gazette. To save time I'll
respond to both of them at once.

First of all, John, you keep bringing up some cases where the USPTO
obviously didn't follow its own rules and where the CAFC didn't object.
Once again: *I admit that it's possible to pull the wool over the CAFC's
eyes.* So I'm not arguing with you there!

All I'm saying is that current law prohibits patents on mathematical
algorithms, even when disguised in various ways. The Supreme Court, the
old CCPA, whichever USPTO lawyer wrote that article for the Official
Gazette, all agree with what I've said, and I think you'll be very
hard-pressed to find a CAFC judgment which explicitly disagrees, *even
if the CAFC's actions do not always conform to their words*.

Finally, you write:

> And finally, note that the PTO is giving tons of algorithm patents
> (look at all of the cryptographic stuff).  They may eventually be
> overturned, but I can't imagine why you are sure they will be.

Huh? When did I say that I was sure? I hope they will be, and I'm doing
what I can to see that they are, but I can't be sure.

Now, Carl: I object to two things in your summary of current USPTO
practice wrt mathematical algorithms. One is that you seem to emphasize
``patentable form,'' which not only has nothing to do with legal
terminology, but in fact is the least important part of the question of
whether a computer-related patent will be granted. Any competent
draftsman can replace ``I claim a computer program'' with ``I claim a
system'' and make other changes necessary so that the USPTO is satisfied
with the *form* of a patent. But just because something is in patentable
*form* does not mean that it should, according to the USPTO, be granted.
Why don't you instead emphasize the issues of *content*?

My second objection is that none of the broad classes of examples you've
given would seem to be patentable according to the above-mentioned case
law, though as I said before your summary is probably a good
approximation to reality (who says the USPTO executes the law?).

You write:

> Indeed I did not see Dan having cited any law from the CAFC in his posting. 

So what? The case law is still valid. Now if I had referred to ``point
of novelty'' and explained its application, you could reasonably have
mentioned that the CAFC has overturned ``point of novelty'' for various
reasons; but I'm not treading on any such dangerous areas.

> I have [reformed] Dan's legal terminology

Huh? You seemed to spend quite some time adding redundant information to
my citations; as I'm not a lawyer, I really don't care. Anyone who
bothers to look has enough information to find what I'm referring to.
You also spent some time pointing out things that I didn't quote, and
adding your own interpretations to the case law; as I know lawyers can
always come up with an infinite supply of things to quote as well as an
infinite supply of interpretations, I once again don't care.

All I was doing was justifying my claims about what can't be done with
mathematical algorithms. I did that, and the case law I quoted stands as
is. If you want to argue interpretations go find the lawyer who wrote
that article for the Gazette.

> And the footnote to that sentence:  " ...
> unless form is to control over substance, the fact that a process result is
> stated in computer-compatible form should not be the basis for holding an
> invention to be non-statutory." 

One of my points to the USPTO was that the above is consistent with the
following: Just because a mental process is performed with the aid of a
computer is no reason to declare the process *statutory*. This will, I
hope, eventually make it into the law.
 
> [mod- Hmm.. I think things could get a bit warm....

Okay, I've refrained from responding to Carl's invective.

---Dan