Re: what does a patent restrict?

On Thu, 21 Oct 1999, Mike Oliver <mikeoliver[_at_]home.com> wrote:
>
> As an admitted non-patent prosecution lawyer (trying to learn from
> your explanations), I just find it interesting that from this
> relatively basic proposition, we end up with the 'legal fiction' that
> a software program is a 'machine' even when it is embodied solely in
> the medium in which it is stored for distribution (and I distinguish
> culpability of a software publisher from that of a software distributor,
> so the observation that a number of publishers have been liable for
> direct patent infringement is not surprising)

I don't think that Mr. Lemley's "legal fiction" remark was entirely accurate. Certainly, legal fictions were not the theory on which statutory claims were found in State Street Bank or AT&T v. Excel -- both cases addressed claims directed to a process.

Further, the statutory predicate for the Beauregard claims are not apparatus "machine" claims, but rather articles of manufacture. The structure and utility is all that is needed to support patentability.

For statutory authority, review 35 U.S.C. s. 101.

> > The test for the validity of a claim is not whether the inventor
> > would have related the claim or framework of claims at a preliminary
> > interview without prompting, but whether the inventor's disclosure
> > would support the claims in the sense of Section 112. It inheres
> > in the idea of a computer program in most contexts today that the
> > representation of the program might be stored in magnetic media, so
> > that such a disclosure would likely support such a claim.
>
> I took a look at the MPEP and it says in part:
>
> "Claims that define a computer-related invention as a
> specific machine or specific article of manufacture must
> define the physical structure of the machine or manufacture
> in terms of its hardware or hardware and "specific software."
> ("Specific software" is defined as a set of instructions
> implemented in a specific program code segment.
> * * * *
> Generally a claim drawn to a particular programmed computer
> should identify the elements of the computer and indicate how
> those elements are configured in either hardware or a combination
> of hardware and specific software."

I didn't make up the things I posted. Had he read a bit further, Mike would have discovered that MPEP Section 21 expressly sanctions the Beauregard claims as articles of manufacture, not as an apparatus:

     [A] claimed computer-readable medium encoded with a
     computer program defines structural and functional
     interrelationships between the computer program and
     the medium which permit the computer program's
     functionality to be realized, and is thus statutory.
     Accordingly, it is important to distinguish claims that
     define descriptive material per se from claims that
     define statutory inventions.

Had he read the Examiner's guidelines, he would see express and actual examples of article of manufacture claim and propagated signal claims. Finally, the MPEP language he quotes, written prior to State Street and AT&T, can no longer be regarded as the lower bounds for patentability. While they may provide "safe harbors" to assure Section 101 status, it is highly unlikely that the office, will press 101 rejections beyond an initial argument unless they choose to ignore those cases.

> In my (admittedly to this point failed) effort to understand this
> issue, I read the above statements to provide that a software
> invention can be claimed as an article of manufacture if it is
> defined in terms of the hardware, or the hardware *and* 'specific
> software'.

RTFM, R-ALL OF-TFM. Then read the case law. It doesn't say what you think it says. This is not even a close question these days. Read, at least, the cases I cited and you'll see this is so.

> This appears to explain your comment that it is the specific novel,
> non-obvious software code plus the disk (hardware) that can be
> claimed as an article of manufacture.

Mike appears to misapprehend the difference between an article of manufacture claim and an apparatus claim. This is discussed elaborately in the very reference he cites. I suggest he read the authorities suggested, and then re-read the MPEP in light of them.

> > > To preserve it and actually use it, *he had to* embody the
> > > program in a fixed medium so it could be 1. copied,
> > > 2. distributed, 3. loaded into ram and 4. executed.
> >
> > Which is precisely why a claim directed to a magnetic medium configured
> > to store the program is supported by the disclosure of the software
> > invention.

Not really, although I suppose that one might argue as a backup that a Beauregard claim is allowable as a "component" of an apparatus, it has been expressly recognized as an article of manufacture.

> So (if I have this right) in the case of a pure software invention,
> in non legal-speak, U.S. law provides that the making, use, sale,
> offer for sale or importing of a disk containing the specific
> software embodiment of the invention infringes a properly drawn
> article of manufacture claim.

Yes, if the word "pure" is removed. I have no idea what that means.

As I have noted countless times in this thread, the claim is infringed if, but only if, the accused apparatus, article of manufacture or acts entail each and every element and limitation of one claim of the patent. If no claims are directed to the article of manufacture, direct infringement can be avoided. (This doesn't mean that indirect infringement wouldn't occur.)

> > > The person who without authority engages in the first 3 activities
> > > infringes the copyright in the particular expression (assuming MAI
> > > is the law as to 3.). I just don't think engaging in any of the
> > > first 3 activities infringe a (valid) patent claim.
> >
> > This is simply not the way courts determine whether a patent claim is
> > infringed.
>

*snip*
>
> I realize (now) that this is not the law, and that this has not been
> the law. I also appreciate that an inventor of a true software
> invention has to have some way of protecting the article of
> manufacture of his invention. I just think that the proper analysis
> is under indirect infringement principles -- which all require someone
> to have committed direct infringement as a predicate.

Well, I suppose what I am awaiting is a reason why Mike thinks this should be the case.

> I also stand by what I have said before that the open source community
> (which I favor because it is a freedom-of-choice and contract based
> rights management model) is at risk from some of these patents.

I agree. I have some solutions in mind, however, which I have repeatedly noted on slashdot and elsewhere. Karsten Self has been an active proponent of "Latent" (Patent versions of Copyleft) and patent pooling as a solution. I support those efforts fully.

> And its this article of manufacture claim that could kill the people
> distributing the code. So far so good though, as I have not heard
> of a distributor of open source code being held liable for direct
> patent infringement.
>
> Thankx for taking the time to help me through this.

You're almost there, but not quite -- the distinction between article of manufacture and apparatus claims is important, particularly for subtle 101 questions. Read State Street and AT&T cases and then reread MPEP, but all of Section 21. Examiner's guidelines would be a help as well.

And, given the pending Priceline and Amazon cases, hold your breath -- there's much more to come! :-)

Andrew C. Greenberg
<werdna[_at_]gate.net> Received on Sat Oct 23 1999 - 00:05:15 GMT