When the Supreme Court decided Alice v. CLS Bank in 2014, it marked a new era in software patent law. By ruling that abstract ideas could not be patented simply because they were executed on a computer, the Alice decision significantly narrowed down the possible universe of software patents. Since then, more and more software patents have been invalidated by the courts—particularly in the Federal Circuit, the appeals court for patent cases, which is still grappling with how exactly to apply Alice.
Intellectual Ventures v. Symantec is the most recent attempt at applying Alice. This last week, Symantec and Trend Micro got handed a big win in the Federal Circuit, with the appeals court invalidating three patents on antivirus technology owned by Intellectual Ventures, the biggest patent troll in the world.
The lower court already invalidated two of three patents, but in that first phase of the case, Symantec lost out on the third patent, resulting in an $8 million judgment against the antivirus company. (The third patent didn’t affect Trend Micro). With the new appellate decision, which invalidates all three, Symantec is now off the hook.
A concurring decision gives patent reform advocates a lot to be excited about
Symantec is likely very pleased with Intellectual Ventures v. Symantec, but a concurring decision also gives patent reform advocates a lot to be excited about, potentially making patent trolling a first amendment issue.
In a post-Alice world, it’s tempting to think that the time of absurd software patents—like the patent on internet shopping carts—is ending. But just because more are being invalidated, it doesn’t mean software patents are over. Alice marked a big change, but the Federal Circuit is still trying to decide how to apply this new Supreme Court precedent to a vast range of cases.
The oversimplistic, but very useful way of thinking about Alice is that you can no longer patent “[Thing that already exists] but with a computer.” For example, you can no longer patent computer bingo, no more than you can patent regular paper bingo. Just because a computer gets involved doesn’t suddenly make an existing invention patentable.
You can no longer patent computer bingo
It seems like a sensible approach, but patents very rudely don’t do you the favor of naming themselves in understandable ways. It’s not like the famous internet shopping cart patent was called “Shopping Carts, But With A Computer.” The Federal Circuit is having a hard time figuring out how to apply Alice, because, well, it’s hard.
In Intellectual Ventures v. Symantec, Intellectual Ventures asserted patents on technology that looks for, identifies, and filters out computer viruses.
The court found that the first patent was really just the same as sorting through your own physical mail—but with a computer. “Here, it was long-prevalent practice for people receiving paper mail to look at an envelope and discard certain letters, without opening them, from sources from which they did not wish to receive mail based on characteristics of the mail,” Judge Dyk wrote in an opinion for a panel of three judges. Doing it with a computer wasn’t anything new.
The second patent was no better—it was a corporate mailroom but with a computer, where the computer, rather than mailroom clerks, would follow rules to release, delete, return, or forward messages.
It was a corporate mailroom but with a computer
The important part of the decision comes with the third patent, which isn’t just a “but-with-a-computer” patent. The patent covered a method of antivirus protection in which communications inside a “telephone network” (e.g., in this era, the Internet) are routed, viruses are detected, and communications are partially inhibited to prevent the spread of the viruses.
But the court ruled that this idea was patent-ineligible, because abstract ideas can’t be patented. After all, “Performing virus screening was a long prevalent practice in the field of computers, and, as the patent admits, performed by many computer users. . . By itself, virus screening is well-known and constitutes an abstract idea.”
The reasoning here, however, is just one way of interpreting Alice. Different panels of different judges have gone different ways, said Vera Ranieri, a staff attorney at the Electronic Frontier Foundation. Ranieri pointed out that cases like BASCOM and Enfish, decided earlier this year, went in completely different directions. “It is unclear to me how they can be read to form a consistent body of law,” she said.
The test applied in Intellectual Ventures is a “technical effect” test, said Charles Duan, director of the Patent Reform Project at Public Knowledge. In the technical effect test (also described here by patent attorney John Roglitz) “a software invention is not patent-eligible unless it has a technical effect on the computer system—making it faster, more efficient, etc.—as opposed to just using the computer system to achieve some other end.”
It doesn’t mean that software patents are over
This test is stricter on software patents. But if this way of interpreting Alice wins out, it doesn’t mean that software patents are over—just that there’s fewer of them. We know this, says Duan, because the technical effect test is basically what’s used in Europe—and software patents still exist over there.
But in a concurring decision, Circuit Judge Mayer made a much more drastic argument, saying that patents that constrain “essential channels of online communication” are antithetical to free speech. The implication is that for constitutional reasons, patents on common email antivirus software should be invalidated.
It’s a radical position, one that was advanced by the American Civil Liberties Union in an amicus brief in Alice when that case went up to the Supreme Court. It’s only a concurring opinion so far, but ever since Alice came down, the Federal Circuit has been in flux about what the law around software patents actually is. And now the ACLU’s stance on software patents has entered the fray.
Mayer’s approach potentially tosses out an entire category of software patents
Mayer’s approach potentially tosses out an entire category of software patents because of their effect on the Internet. A less extreme way of interpreting Mayer’s opinion is he only meant to make a policy argument, pointing out the importance of judging patents harshly. If courts aren’t strict with software patents related to the Internet, free speech is imperiled.
The upshot is that Intellectual Ventures v. Symantec settles very little, but it does show that there’s a wide range of possible futures for software patent law—futures in which software patents continue to flourish, and futures in which patents get tossed out just for their effect on the Internet. Which way we end up going, only time will tell.