You Can’t Just Patent An Idea — You Actually Have To Make A Thing, Supreme Court Rules

The Supreme Court today issued rulings on a handful of cases. One was about two companies nobody’s ever really heard of, arguing over patents for software to manage banking transactions. The details of the patents themselves, and the transactions they deal with, are kind of complicated and insidery — but they’re also not necessarily that important. The broader implications of the ruling, and the legal precedent the Court set with it, though, will have an impact for years to come.

What was the actual case before the Court?
The case is Alice Corp. v. CLS Bank International. It all started over ten years ago: Alice owned patents on certain computer programs for managing risk in financial trades. CLS started to use a similar technology to solve the same risk problem. Alice said: you can’t do that, we have a patent and we will sue your pants off. CLS said: you can’t patent an idea, we’ll sue you. Lower courts sided with CLS about the validity of Alice’s patents. After many years of suits and appeals, the case came to the Supreme Court and they heard arguments on it in March.

Specifically, the legal question the Court was looking at was “[w]hether claims to computer-implemented inventions—including claims to systems and machines, processes, and items of manufacture—are directed to patent-eligible subject matter within the meaning of 35 U.S.C. § 101.”

Back in English, that question more or less is: can software and computer processes be patented under this section of patent law as currently written and understood?

What was the Court’s ruling?
The Court ruled 9-0 (PDF) to affirm the lower court’s finding that Alice’s patents on the financial software were not valid. Justice Clarence Thomas wrote the opinion, with Justices Ginsburg, Breyer, and Sotomayor concurring. They found that the particular kind of risk mitigation that Alice held the patents for was an abstract idea, and as such, not eligible for patent.

The justices differentiated ideas from execution, writing:

The Court has long held that §101 [of patent law] … contains an implicit exception for ‘laws of nature, natural phenomena, and abstract ideas. … In applying the exception, this Court must distinguish patents that claim the ‘building blocks’ of human ingenuity, which are ineligible for patent protection, from those that integrate the building blocks into something more.

So did this undo patents on software?
No. At least, not entirely. Some software patents, on review, may not hold up to the new legal standard. However, patents that still show actual human legwork are still fine. Basically, it’s a narrowing of the category of “invention.” Just saying, “Let’s do something we’ve always done, but use a computer for it” doesn’t clear that bar. Showing how your idea, plan, code, or product is creative and inventive — something you actually made or thought up — keeps it eligible for patent.

Who does this ruling affect?
Most of the largest companies in the country, for starters. Amazon, Facebook, Google, Microsoft, and a ton of others filed briefs in the case, as did the EFF and the ACLU. For once, though, the advocacy groups and the corporations were more or less on the same side. Though they had different legal arguments and different personal priorities, they all basically agreed that software patents should be limited.

For companies, narrowing the field of patentable software allows them to keep making stuff, and prevents them from having to spend time dealing with lawsuits from patent trolls. For advocates, narrowing the field of patentable software allows more developers and potential innovators to, well, keep making stuff, and to prevent getting sued by larger corporations. And patent trolls find themselves without a way to keep being a nuisance to the world at large. Kind of a win/win all around.

But the devil’s in the details, right?
As always. This case exists not in isolation but in a pattern of other cases. The Court has been gradually narrowing the scope of what can be patented since about 2005, as Vox points out, and the Alice finding is right in line there. But it also doesn’t draw a definitive, clear line in the sand.

As Gigaom explains, instead of creating a bright line, the Court proposes a new way of looking at patents: a relatively simple two-part test. First: does the patent cover an abstract idea that isn’t patentable? And second: if it does, are there new, inventive steps involved making the idea unique enough to patent?

The trick, for the Court, was to look at a way to limit patent trolls while also protecting legitimate inventors. But it’s a huge uphill battle; over 40,000 software patents are issued every year.

Tech and legal experts agree that patentability and software patents will almost certainly wind up before the Supreme Court again in the future at some point, because of the sheer number of them and because the Court has issued fairly narrow rulings each time the matter comes up, leaving the window open for stricter readings in the future.

Got it! But I don’t work in computers or IT. Why does this matter to me?
Because in 2014, pretty much everything you use is somehow driven by software. Sure, there are actual computer programs and mobile apps, but it’s more than that. Traffic lights and the cars that drive through them are software-driven. Consumer products from TVs to iPhone power adapters to wristwatches and coffeemakers have code controlling them. So, too, does physical infrastructure like the electric grid and virtual infrastructure like the entire financial system.

The ability of individuals and companies to create new software innovations is crucial to almost every field. Patent trolls have been claiming ownership of abstract, general ideas to game the system and turn a buck. That wastes time and resources and increases costs for everyone up and down the line — including, in the end, consumers.