Judges split on software patents and computer transubstantiation | Opensource.com

Judges split on software patents and computer transubstantiation

Posted 20 May 2013 by 

Rob Tiller (Red Hat)
Rating: 
(12 votes)
Software patents: a note on the new CLS Bank case
Image by : 

opensource.com

submit to reddit

The law of software patents took an interesting, and ultimately encouraging, turn a little more than a week ago. In the CLS Bank case, ten judges of the Federal Circuit issued five separate opinions, without any single legal theory gaining a majority. Their debate showed that the scope of the subject matter requirement for patenting software is far from settled. It also makes it more likely that the Supreme Court will speak to the issue, and get it right.

First, a little background. The case concerned a group of patents relating to settlement risk in financial transactions. Generally in financial transactions, there is a risk that a party to the transaction may not pay or perform as promised. The patents address that risk by relying on a trusted third party to assure a simultaneous exchange. This is done using a computerized trading platform. (If this sounds sort of abstract or blindingly obvious to you, read on—you will soon feel vindicated.)

CLS Bank filed suit in federal court in the District of Columbia against the patent holder, Alice Corp., seeking to invalidate the patents. Alice counterclaimed alleging infringement. CLS filed a motion for summary judgment, which argued that the patents were invalid under 35 U.S.C. Section 101 based on the Supreme Court’s decision in the Bilski case. The trial court agreed, finding the patents invalid on the ground that they were directed to an abstract idea.

A panel of the Federal Circuit reversed, but the en banc Federal Circuit (meaning all the judges sitting together) agreed to rehear the case. (Full disclosure: at this point Red Hat joined in an amicus brief with Google, Dell, Facebook, Homeaway, Intuit, Rackspace, and Zynga arguing that the patents were invalid.)  The en banc decision affirmed the lower court’s finding of invalidity, but the judges were sharply divided on the grounds for their decision.

Although the opinions are lengthy (all told, 135 pages) and address several significant points, there is one major issue that both divided the judges and partially illuminates their world views:  the significance of describing a computer as part of the invention. The judges mostly agreed that the idea of managing settlement risk with a third party was abstract such that by itself it could not be patented. They differed, though, on whether using a general purpose computer for managing settlement risk meant that the patents avoided invalidity based on abstraction.

One camp of five judges led by Judge Lourie supported invalidating the patents based on the view that the computer limitation as merely "insignificant post-solution activity relative to the abstract idea..." Judge Lourie wrote that "simply appending generic computer functionality to lend speed or efficiency to the performance of an otherwise abstract concept does not meaningfully limit claim scope for purposes of patent eligibility."

At the same time, a group of four judges strongly disagreed on the significance of the computer limitation. Chief Judge Rader, writing for this group, said that "a computer programmed to perform a specific function is a new machine with individualized circuitry created and used by the operation of the software." Similarly Judge Moore wrote for this same group of four judges that "a general purpose computer in effect becomes a special purpose computer once it is programmed to perform particular functions pursuant to instructions from program software," on the theory that software "effectively rewires a computer."

The Supreme Court’s Bilski case earlier established that a tie to a machine is a clue to patent eligibility. Viewing the software and hardware together as a new machine was, in these judges’ view, sufficient to overcome the objection that the patent was too abstract.

It seems strained and unconvincing to think of computers as becoming "new machines" each time they run a different program. It makes almost as little sense as thinking of the radio as a new machine each time it plays a different song, or the telephone as a new machine each time a different person calls.  My laptop can do a lot of different things, but it doesn't ever strike me as transforming its identity.

Perhaps the "new machine" argument would seem justifiable if one thought it would prevent the end of the world as we know it. In fact, this may explain some of this debate.

Judge Moore’s tone was in places apocalyptic. She worried that Judge Lourie’s approach would mean "the death of hundreds of thousands of patents" and would "decimate the electronics and software industries." She said, "There has never been a case which could do more damage to the patent system than this one." Chief Judge Rader, in an unusual second opinion titled "Additional Reflections," stated that without patent protection "investors would quickly opt to put their resources" into some field safer than technology.

In short, four of the ten Federal Circuit judges seem to think that highly abstract software patents serve the accepted purposes of the patent system in encouraging innovation and economic progress. They seem to fear that eliminating such patents would harm the economy. On the other hand, five of the ten judges, led by Judge Lourie, did not buy this view.

The Judge Lourie group is correct. While eliminating abstract software patents could temporarily affect some corporate balance sheets, most of the effects would be positive. It’s highly unlikely that the software business would be decimated, since, after all, it did quite well before the age of software patents began in the 1990s.

As the amicus brief noted above put it, "low quality patents in computer-related industries have become a scourge that raises costs and places a drag on innovation. One study found that patents in these industries have produced net litigation costs far in excess of the net profits derived from the patents themselves." Low quality patents "add little, if anything, to human knowledge while broadly foreclosing future development by others."

It isn’t really surprising that Federal Circuit judges who have devoted their working lives to the patent system tend to think most patents, including software patents, are valuable and useful.  What is surprising is that, in spite of this likely tendency, the en banc Federal Circuit invalidated the patents here. Half of the en banc judges took the sensible view that running a program on a computer does not transform the computer or render software less abstract.

Both sides of this debate have now been carefully articulated, and the battle lines are drawn. Alice is entitled to seek discretionary Supreme Court review. The Court could well take this case in the next few months. If it does, it seems doubtful it would find that a generic computer is continually becoming a new machine.

submit to reddit

6 Comments

stites
Open Source Champion

Judge Moore’s tone was in places apocalyptic. She worried that Judge Lourie’s approach would mean "the death of hundreds of thousands of patents" and would "decimate the electronics and software industries." She said, "There has never been a case which could do more damage to the patent system than this one."

To which I reply:

hip, hip, HURRAY
hip, hip, HURRAY
hip, hip, HURRAY

-----------------------------
Steve Stites

P.S. I submitted an amicus curiae in this case and if it is appealed to the Supreme Court I will submit an amicus curiae to the Supreme Court.

Vote up!
6
Vote down!
0
Ragu Bhat

The very argument that we have a level playing field w.r.t software patents is flawed. Erick Robinson of Qualcomm has argued for this elsewhere. Patents on software has to be done away with for real innovation to happen, more software programs to be written and for people to enjoy computing rather than dread it!

Vote up!
4
Vote down!
0
Richard

I suspect that the apocalypse that Judge Moore fears is not for the technology industry, but for the legal industry.

Vote up!
4
Vote down!
0
Greg Aharonian

The CLS Bank decision are mostly semantic nonsense, because they repeatedly use the undefined "abstract". You can't say some piece of software is "abstract", unless you define what you mean by "abstract". And for over 40 years, the CAFC and Supreme Court have refused to define "abstract" (and Judge Lourie defines it circularly in CLS Bank).

And it is a violation of Equal Protection to require software patents to show "significant post-solution activity", but not to require hardware patents to do the same. Look at signal processing patents, both hardware and software - they all claim the implementation of the new SP technique, based on inputs and outputs. But no one cares how to apply such new inventions - that's what engineers are trained for. And that is why hardware patents never claim post-solution activity, yet no one complains. It is only a fake-engineering principle invented by the engineering ignorant Supreme Court that applies this test to software.

CLS Bank is one of a long line of 101 court decisions that resolves nothing, because it engages in semantic sloppiness and nonsensical engineering concepts.

Vote up!
2
Vote down!
0
Rubberman
Open Enthusiast

Generally speaking, I am not in favor of software patents. That said, I do think there are some that are worthy of such protections. I am the sole inventor of a patent for adaptive systems related to binary compiled languages/programs. The solution to a major problem that allows users to modify/create classes and methods in compiled programs without actually changing existing code or recompiling/linking new code, was, IMO, novel, non-obvious, and transformative - hence suitable for patent rights. There are related technologies, such as reflection patterns in Java, .net, et al, but they still require compiling (if even at runtime) code to machine or byte code before it can be used by a program. My solution does not require this step.

So, are software patents bad? I'd say that in 99.9% of the time, they are technologically worthless and should not be granted such status. However, there is that 0.1% that are.

So, in the final analysis, I would prefer that software was made unpatentable. Let such "innovations" be trade secrets in that duplication of function does not mean infringement!

Vote up!
2
Vote down!
0
Larry

Three brief comments:

1. It is certainly possible to have software machines. Every program that runs might be regarded as one. It is infrequent that realization in hardware, firmware, or software need make a difference in the intrinsic value of the idea, although the commercial value might differ by platform employed.

2. Not every machine is patentable, nor should be. Further, there are many patents that, well understood, should never have been issued.

3. The public interest that needs to be served is simply this: someone who has paid the time, effort, and other cost of development of something new and useful ought to be allowed to profit from this investment rather than have the benefits usurped by others, while at the same time, the benefits should be available to others on reasonable terms. If the creator cannot profit, why should he create? And if the useful idea is not made available so as to be useful, why should the rest of us expend resources to protect it? Patents and copyrights are intended to serve this purpose. Perhaps a third category of protection is worthy of consideration. But in any case, the lawmakers need to keep it simple and to the purpose. (If I invent a superior legal paradigm, can I patent it? :-) )

Vote up!
2
Vote down!
0

Rob Tiller is vice president and assistant general counsel for Red Hat, where he manages patent, trademark, and copyright matters. He is a frequent speaker and writer on open source legal issues. Before coming to Red Hat, he was a partner with the law firm of Helms, Mulliss & Wicker, PLLC, where he specialized in commercial and IP litigation. He is a graduate of the University of Virginia School of Law, and a former clerk for Justice Antonin Scalia of the U.S. Supreme Court, and Judge

Should I use a permissive license? Copyleft? Or something in the middle?