Supreme Court smashes “do it on a computer” patents in 9-0 opinion

The most-anticipated patent decision from this Supreme Court term was published today. The decision involves finance-related software patents that were being used against CLS Bank, a key part of the global financial infrastructure.

The court ruled unanimously that all of patent-holder Alice Corp.'s claims were invalid, because they simply added computer language to a basic idea: using a financial intermediary to create trust in transactions. The 9-0 opinion [PDF], written by Justice Clarence Thomas, is the clearest statement yet from the Supreme Court that adding technological-sounding language to existing ideas isn't enough to get a patent.

Some advocates were hoping the case would go so far as to eliminate software patents altogether. If that were to happen, this would have likely been the case to do it. But the court didn't go that far, instead suggesting that software patents could still be allowed when they "improve the functioning of the computer itself" or "improve an existing technological process."

Today's decision clarifies a fractured decision in the US Court of Appeals for the Federal Circuit that couldn't reach a majority about whether the patents were valid. That left intact the lower court decision, which had ruled the patents invalid.

The Supreme Court addressed this area of patent law in the 2010 Bilski decision, which invalidated a patent on a financial hedging scheme. Today's ruling essentially found that the Alice patents were a kind of "Bilski on a computer." Thomas writes:

On their face, the claims before us are drawn to the concept of intermediated settlement, i.e., the use of a third party to mitigate settlement risk. Like the risk hedging in Bilski, the concept of intermediated settlement is “‘a fundamental economic practice long prevalent in our system of commerce.’”

The court rejected the argument of Alice's lawyer that the abstract-ideas exception in patent law should only apply to "preexisting, fundamental truths."

A final section of the opinion blasts the Federal Circuit's idea that the "system" and "media" claims should be treated any differently than Alice's "method" claims. A group of four judges, including recently departed Chief Judge Randall Rader, banned method and media claims but would have allowed system claims, which they said involved hardware "specifically programmed to solve a complex problem."

The high court dismisses these terms as being much gamesmanship by patent lawyers. To allow one type of claim but not another would be to interpret patent law "in ways that make patent eligibility ‘depend simply on the draftsman’s art.’”

So what's abstract?

Alice's patent claims did nothing more than "simply instruct the practitioner to implement the abstract idea of intermediated settlement on a generic computer," the court held.

That makes the patents abstract, but the justices specifically declined to lay out a definition of what "abstract" is.

"Both the Bilski invention and Alice's method are squarely within the realm of 'abstract ideas' as we have used that term," Thomas writes. And the court says, that's enough for now. "We need not labor to delimit the precise contours of the 'abstract ideas' category in this case."

Here, as in Bilski, the Supreme Court is navigating between three early computer-oriented cases that arose between 1972 and 1981. Gottschalk v. Benson and Parker v. Flook both rejected algorithm-based patents, while Diamond v. Diehr allowed a patent on a computer program that described a method of curing rubber. These three early cases all relate to Section 101 of patent law, like today's case, and they don't exist in comfortable harmony.

Today, the Supreme Court emphasized that it was the non-computing elements of the Diehr invention that made it patentable. "The invention in Diehr used a 'thermocouple' to record constant temperature measurements inside the rubber mold"—something “the industry ha[d] not been able to obtain.'"

Future court battles are likely to circle around the concept of just what is abstract. In one sense, it seems that the Supreme Court is taking technology that's actually just quite old—like using hedging and intermediaries in business—and calling it "abstract."

The effect of that will be to put more patent battles in the area of Section 101, which is where frequent patent defendants want it. Getting a ruling that software or Internet patents are "abstract" means they can be thrown out of court relatively quickly. Proving patents are invalid because they were anticipated by earlier inventions is much more likely to lead to expensive court battles that involve discovery and the hiring of expensive experts.

Patent owners who go to court will certainly always argue their inventions aren't abstract. Still, today's opinion has guideposts that will make it easier for judges to rule that they are.

"The Federal Circuit had decisions that basically said, if you had a physical thing—even a generic thing, like a 'data processing unit'—that was enough to make it patent eligible," said Matthew Levy, a lawyer who works on patent reform issues for the Computer and Communications Industry Association. "District court judges would hang their hat on that. Now, in a lot of cases, they'll say [to a patent owner], you just have generic terms here, like a CPU—that's exactly what the Supreme Court said doesn't matter."

The future of software patents

The big software companies that wrote amicus briefs that essentially were long paeans to software patents, like Microsoft and IBM, can rest assured that their nightmare scenario—the mass-invalidation of the thousands of software patents they own—will not come to fruition. The idea of throwing out software patents as a class, via the courts, looks more pie-in-the-sky than ever before.

As for the many software and Internet companies that emphasized the harm being done by software patents, they'll have a stronger basis to stand their ground. This decision seems to explicitly encourage decisions like the one written by US District Judge Denise Cote, who eviscerated a patent troll's claim on "matchmaking," then awarded attorney's fees to defendant FindTheBest. We're likely to see more of those in the future.

Still, it's hard to see this decision wiping out many lawsuits filed in a patent-friendly area like the Eastern District of Texas. It's really up to individual judges whether they see a patent as being broad or brazen enough to be claiming an "abstract idea." And even in the best case scenario, a defendant who wins a speedy decision will have to spend a few hundred thousand dollars, so "bottom feeder" patent trolls asking for $25,000 or $50,000 will still be able to score settlements. Sadly, as pro-defense decisions make it harder for trolls to compete in the "mid-market" asking for six-figure settlements, smaller companies may become more frequent targets.

That was clearly the concern for the Main Street Patent Coalition, which sent out a press release this morning lauding the decision but insisting that Congress must take action to "stop extortionist claims," even though this year's patent reform bill was killed in the Senate. "Most American businesses do not have the time or money to fight these frivolous lawsuits that can take years and millions of dollars to resolve in the courts," the group wrote.