Supreme Court to Hearing Business Method Patent Case

This is big. Basically, the Supreme Court is reviewing a patent on a business method, specifically a way to hedge against inclement weather (I sh$# you not, someone patented betting on a cloudy day), and it could effect the future of much genetic and software algorithm patents, which, after all, are more discoveries than inventions.

I am with the anti-patent side, whose basic argument is here:

Eben Moglen, director of the Software Freedom Law Centre is emphatic that business process patents should never have been allowed in the first place. Patent law, he says, cannot award ownership of facts of nature, or mere mental activities, or algorithms because the Supreme Court has been unambiguous on that point for more than 150 years. However, for the last 20 years, the USPTO and its supervising appellate court have been liberal with patents for inventions consisting of software or business methods enabled by software.

But I would actually go further: While I understand the need to update patent law to apply to new technologies, I believe that the standard should be a clear showing that a lack of significant innovation is resulting from the lack of protection.

After all, the basic reason for IP, Patent and Copyright specifically is to encourage innovation by limiting the rights of other people to use that expression or invention^*, as it says in Article 1, Section 8 of the Constitution of the United States:

The Congress shall have power to .....

To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries;

(emphasis mine)

It's about public benefit, not property rights: a temporary exclusive license is granted to an individual in order to help society as a whole.

While some solutions have been offered to deal with this problem, most notably crowd sourcing patent review, the real solution is to go back to where we were in 1985, when neither genes, species, nor software algorithms were patentable. We got innovations in those areas without those protections.

It should be noted that the Supreme Court only takes the cases that it wants to, and lately when it takes up patent cases, it does so to slap down the USPTO and/or the Federal Patent Court, both of whom tend to be like a man with only a hammer, and see everything like a nail.

In arguments, the court, except for Clarence Thomas, who never talks, appeared to be somewhat disparaging of the arguments of the plaintiffs:

Huge legal expenses and 13 years later, the two men behind the case, Bernard Bilski and Rand Warsaw, had their day in the U.S. Supreme Court on Nov. 9. Most legal experts though, agreed that the duo had no chance of victory. "I don't think anyone other than Bilski thinks that Bilski deserves a patent," says Mark Lemley, a professor of law at Stanford University. (See the 50 best inventions of 2009.)

The bench seemed to reflect this view, and several Justices suggested somewhat humorously that if the Bilski argument were to proceed, a number of other ludicrous patents could be issued. Justice Antonin Scalia asked if under Bilski's argument, methods of horse-training could be patented, while the court's newest member, Justice Sonia Sotomayor, asked if a "method of speed-dating" was patentable.

The interesting thing here is that most of the business community, excluding patent trolls and their close relatives, realize that the current system is completely out of control, which is obvious when the Wall Street Journal has an OP/ED that describes the case as, "The Supreme Court v. Patent Absurdity".

^*Trademark protection is really about protecting the consumer by ensuring that what they buy is what they thought that they were buying.

Well, I Missed a Major Decision……

Bilski v. Kappos, a business patent case that made it to the Supreme Court was decided 3 months ago. Basically, Bilski had patented a method of hedging energy based on the weather.

The Supreme Court unanimously invalidated the patent, it was after the unique idea that one should bet on the weather, but by a 5-4 majority, they kept the business method patent, albeit with a tightening of standards:

While all nine justices agreed that the “invention” at issue in the case—a method for hedging weather-related risk in energy trading developed by Bernard Bilski and Rand Warsaw—was too abstract to merit patent protection, only four signed on to Kennedy’s opinion.

That opinion held that the "machine-or-transformation" test for patentability--created by the U.S. Court of Appeals for the Federal Circuit in its Bilski decision--was a "useful clue" when gauging a subject’s patentability but shouldn't be considered the only applicable test.

I think that this was generally a loss for patent sanity, though it does make getting a business patent more restrictive, though, unsurprisingly, SCOTUS didn't say how much more restrictive the standards should be.

My earlier posts on the matter.

I Hope So

Ars Technica asks, "Will the Supreme Court end human gene patents after three decades?"

I think that it likely that they role back patent protections.

These days, they only seem to take patent cases when the United States Court of Customs and Patent Appeals goes too far with patents.  (Which it does with mind-numbing regularity):

Since the 1980s, patent lawyers have been claiming pieces of humanity's genetic code. The United States Patent and Trademark Office has granted thousands of gene patents. The Federal Circuit, the court that hears all patent appeals, has consistently ruled such patents are legal.

But the judicial winds have been shifting. The Supreme Court has never ruled on the legality of gene patents. And recently, the Supreme Court has grown increasingly skeptical of the Federal Circuit's patent-friendly jurisprudence.

Meanwhile, a growing number of researchers, health care providers, and public interest groups have raised concerns about the harms of gene patents. The American Civil Liberties Union estimates that more than 40 percent of genes are now patented. Those patents have created "patent thickets" that make it difficult for scientists to do genetic research and commercialize their results. Monopolies on genetic testing have raised prices and reduced patient options.

On Monday, the high court will hear arguments about whether to invalidate a Utah company's patents on two genes associated with breast cancer. But the legal challenge, spearheaded by the American Civil Liberties Union and the Public Patent Foundation, could have much broader implications. A decision could invalidate thousands of patents and free medical researchers and clinicians to practice medicine without interference from the patent system.

It's very clear that a gene is a discovery, not an invention, but the patent court believes that you can patent a rainy day (I mean this literally: They approved a patent on weather derivatives in Bilski v. Kappos, which was later overturned by the Supreme Court. This court also allowed for patenting of tax deductions)

Court Rules Gene Patents Illegal

It is generally true that bad people push the law until it breaks, and the folks at Myriad Genetics are a truly nasty bit of work.

Basically, they discovered the genes BRCA1 and BRCA2, which predispose one to breast cancer, and had received a patent on the genes, and offered a not particularly good and very expensive test for the genes, and using their government granted monopoly, they were prohibiting any and all competing tests or research on better tests for these genes.

The judge has now ruled that the genes are a discovery, and not an invention, which makes them non-patentable, except, perhaps in Germany.^*

The VC's are wringing their hands, because patent protection helps with their pump and dump schemes, but development has been continuing apace on genes that are not covered by these patents:

Some biotechnology investors and executives say that lack of patent protection for DNA could diminish investment and remove incentives to develop tests. That could slow the move toward so-called personalized medicine, in which genetic tests are used to determine which drugs are best for which patients.

James P. Evans, a professor of genetics at the University of North Carolina, said that would not necessarily be the case. There is thriving competition in areas like testing for mutations that cause cystic fibrosis or Huntington’s disease, even though no company has exclusivity.

“It’s quite demonstrable that in the diagnostic area, one does not need gene patents in order to see robust development of these tests,” he said.

Note that this does not prohibit patenting the tests, it just prohibits patenting the genes.

This is a good ruling: the patenting of genes, or for that matter the patents of hedging techniques, as is the case in Bilski, which the Supreme Court will heard arguments for in November.

IP law is, at its core, public interest law (it's in the constitution), and patents on business methods, species, and genes, do not serve the public interest

Earlier posts on the subject.

^*This is what got us that Mercedes ad where they say that they have a patent on crumple zones, but "Never enforced the Patent". They never enforced the patent, because it is not recognized anywhere else in the world.^†
†I believe Germany changed their patent laws at some point in the 1970s.^‡
‡I offer the caveat that these comments in the footnotes regarding the Mercedes patent are recollections of a conversation over a decade ago vague 20+ year old memories though, so YMMV, though a Google search does have people who recall the ad.

Today in Bad Ideas

There are bills in the Senate Judiciary Committee that would restore patents for software and genes, because ……… a f%$# tonne of lobbyists and corrupt campaign contributions I guess?

I don't think that this Supreme Court has done much good, but their work in reining in excesses of our current IP regime is good.

The current regime is a license for parasite rent seekers.

It creates massive inequality and it destroys innovation when some patent troll can shake you down at a moment's notice:

The Senate Judiciary Committee is scheduled to consider two bills Thursday that would effectively nullify the Supreme Court's rulings against patents on broad software processes and human genes. Open source and Internet freedom advocates are mobilizing and pushing back.

The Patent Eligibility Restoration Act (or PERA, S. 2140), sponsored by Sens. Thom Tillis (R-NC) and Chris Coons (D-Del.), would amend US Code such that "all judicial exceptions to patent eligibility are eliminated." That would include the 2014 ruling in which the Supreme Court held, with Justice Clarence Thomas writing, that simply performing an existing process on a computer does not make it a new, patentable invention. "The relevant question is whether the claims here do more than simply instruct the practitioner to implement the abstract idea of intermediated settlement on a generic computer," Thomas wrote. "They do not."

Yeah, Clarence f%$#ing Thomas got this right. ¯\_(ツ)_/¯  

That case also drew on Bilski v. Kappos, a case in which a patent was proposed based solely on the concept of hedging against price fluctuations in commodity markets.

………

Software and Internet advocates have taken notice. This week, the Linux Foundation, working with the Cloud Native Computing Foundation (CNCF), announced an expanded partnership with Unified Patents, intended to defend open source software against what it gamely calls "non-practicing entities" (NPEs), but most people would term patent trolls. "As the risk and volume of frivolous litigation against open source projects grows, the need to provide accessible protection from NPEs has become crucial," the Linux Foundation writes.

In interviews with The Register, leaders at CNCF and Unified Patents described patent trolls as actively chasing any widespread technology, aiming for settlements over the cost of trials. Nearly 98 percent of NPE claims are settled, according to Unified Patents, but NPE claims challenged at the US Patent and Trademark Appeals Board lose 67 percent of the time.

Challenging patent claims, however valid, could get tougher under the PREVAIL Act, the other bill being considered by the Senate Judiciary Committee this week. PREVAIL would, among other changes, limit patent challenge petitions to 14,000 words, hampering attempts to debunk complex patents. The Act would also eliminate clearance patents, which companies can use to clear any infringement claims prior to their own products' release.

………

Another wrinkle in the PERA bill involves genetic patents. The Supreme Court ruled in June 2013 that pieces of DNA that occur naturally in the genomes of humans or other organisms cannot, themselves, be patented. Myriad Genetics had previously been granted patents on genes associated with breast and ovarian cancer, BRCA1 and BRCA2, which were targeted in a lawsuit led by the American Civil Liberties Union (ACLU). The resulting Supreme Court decision—this one also written by Thomas—found that information that naturally occurs in the human genome could not be the subject to a patent, even if the patent covered the process of isolating that information from the rest of the genome. As with broad software patents, PERA would seemingly allow for the patenting of isolated human genes and connections between those genes and diseases like cancer.

This is a horrendously bad idea. 

Shut it down

Supreme Court Sides With Google on API Copyright

It should also be noted that the Supreme Court also slapped down the United States Court of Appeals for the Federal Circuit (CAFC, AKA, the "Patent Court") once again.

This time, the Patent Court big-footed the normal appellate process, it does not normally handle copyright claims, and made an egregiously bad decision, which it pretty much always does, to boot.

Once the Supreme Court noticed, it was over but for the fat lady singing.

The purpose of patents is two fold:  To encourage innovation, and to make the information contained in patents available to the public.

The patent troll friendly CAFC is doing exactly the opposite.

This time the Supreme Court ruled that using the application programming interface (API) is fair use.

An API is a series of rules which describe how a computer program is programmed, or how it interfaces with hardware, and for program compatibility, programs running on the same hardware, or the same operating system, or on a similar language.

It's not a creative expression (copyright) but a functional description of the system, like the manual for a toaster.  The specific words in the manual are covered by copyright, but the idea that you slide a lever from left to right to set the doneness of your toast is not.

For example, we can look at a "Hello World" program in Java, where Oracle was asserting infringement: (This is also an example of why I don't want to program in Java, it's way to obscure and verbose)

import javax.swing.JFrame;  //Importing class JFrame
import javax.swing.JLabel;  //Importing class JLabel
public class HelloWorld {
    public static void main(String[] args) {
        JFrame frame = new JFrame();           //Creating frame
        frame.setTitle("Hi!");                 //Setting title frame
        frame.add(new JLabel("Hello, world!"));//Adding text to frame
        frame.pack();                          //Setting size to smallest
        frame.setLocationRelativeTo(null);     //Centering frame
        frame.setVisible(true);                //Showing frame
    }
}

If you want to create an application, or an operating system, which is what Google did, and allow for Java programmers to work in this without learning a completely new way of doing the same thing, then you have to use the same code words, things like "Jlabel", "Jframe", "setTitle", etc. 

That's what Google did, and what the makers of IBM clones did in the 1980s, and what third party spare parts manufacturers do when they make cheaper parts for your (increasingly computerized) car.

The Supreme Court said that it's fair use, which is nice, but not as nice as if they had said that it was not subject to copyright:

It's taken over a decade, but the Supreme Court has finally said that Google's copying of the Sun Java API for Android was clearly fair use in a 6-2 ruling (Barrett did not take part since she was not on the Court when the case was heard). The background of this case is actually kind of important to understanding what just happened, so here's a quick(ish) recap.

As you'll probably know, this case began in 2010 when Oracle, which had just taken over Sun, sued Google for patent infringement. The patent parts were quickly tossed out and what remained was what referred to as a "sideshow" to the main event: a question about whether APIs could be covered by copyright. Pretty much all historical evidence, including an important Supreme Court ruling from the 1990s, said no, interfaces are not covered by copyright.

Oracle and friends then spent the next decade deliberately gaslighting basically everyone who doesn't understand what an API is, and insisting that it's the same as executable code. The district court, under Judge William Alsup, who somewhat famously taught himself Java to better understand the issues in the case (he already knew how to code and was a hobbyist programmer), correctly found that APIs are not subject to copyright as they represent a "system or method of operation" which is explicitly not covered by copyright, as per Section 102(b) of the Copyright Act.

Rather than go to the 9th Circuit (as it should have) the case went to the Federal Circuit, which hears all patent appeals. That was because the case began as a patent case, even though it no longer was. CAFC judges are somewhat infamous for never finding a patent issue they couldn't screw up, and decided to extend that to copyright. In the ruling overturning the lower court, they made it clear that because they were code illiterate they could not understand the difference between executing code and an API, even though it's like the difference between a novel and a dictionary.

The CAFC bigfooted the case, which is not surprising.  They also do things like change determinations of fact by the lower courts, something that is not supposed to be done by appellate courts, which are only supposed to adjudicate matters of law.

The case went back to the district court, where the jury this time sided with Google, this time saying that the use of the code was covered under fair use. That whole trial was a little weird, because reading between the lines, you could see that nearly all the arguments for why copying an API was fair use were really about why an API shouldn't be covered by copyright at all (as per Alsup's original ruling) and then squeezing that square peg into the round hole of fair use to make it work. Once again, however, CAFC got flummoxed by an API looking like code and overturned -- which is quite crazy because CAFC had, in its first ruling, insisted that the jury should decide this issue (as a matter of fact) and then when the jury said "fair use" CAFC suddenly decided that it was a matter of law that it could overrule.

It's taken over a decade, but the Supreme Court has finally said that Google's copying of the Sun Java API for Android was clearly fair use in a 6-2 ruling (Barrett did not take part since she was not on the Court when the case was heard). The background of this case is actually kind of important to understanding what just happened, so here's a quick(ish) recap.

So, basically, we have another set of incoherent, and weird rulings from the CAFC, and once it gets on the SCOTUS radar, they shoot it down.

The patent court needs to be abolished.  The argument for it, that there needs to be a court with a specialized focus on patents is simply not justified by the results.

Instead, we have an institution that is completely captured by the patent bar, and has literally allowed for the patenting of a rainy day (Bilski v. Kappos,).

Just shut them down, and reassign the current judges to traffic court.