Is software patentable? This simple question, despite over 40 years of judicial wrestling, remains unanswered. Many in the business world are hostile to software patents. After all, much of the vibrant software startup culture seems to operate just fine in an atmosphere of uncertainty. For example, just last month, Facebook acquired software startup WhatsApp for $19 billion, despite WhatsApp owning zero issued patents. Further, patent litigation reform is a hot topic, and software cases can, and frequently do, carry very high stakes.

For example, this coming Monday, the Supreme Court will hear arguments in the much-anticipated case Alice Corp. v. CLS Bank International, a case that pits an NPE against a worldwide banking consortium.  The patent in the case covers currency trading, and specifically CLS Bank’s currency settlement platform, responsible for $5 trillion in daily trades. CLS Bank convinced a district court, and a divided panel of the Federal Circuit that Alice’s patents were invalid under § 101. This past summer, the district court was again affirmed by a highly-fractured en banc opinion, with seven separate opinions running over a hundred pages.

So why all the confusion? Why are software patents such a thorny issue? Much of it has to do with what software is. At the end of the day, software is no more than a list of instructions, written in a highly-structured language that a computer can understand. And patent law has consistently drawn a line between inventions and abstract ideas, including lists of mental steps. Inventions are patentable, mere “abstract ideas” are not. So is software really just a list of steps, and hence an abstract idea, rather than an invention? And what happens if a patent applicant appends to his claim that the steps are performed “on a computer?” Is that enough?

I would argue that it is. As Judge Rader said in CLS Bank, “a computer without software collects dust, not data.” And if the software can be run on a machine to accomplish a useful task, that software should be patent-eligible. That said, they may be still be unpatentable either as not novel, obvious, or insufficiently enabled. After all, U.S. patent law is supposedly technologically neutral, and covers “anything under the sun made by man.” Software included.

In this month’s issue of JETLaw, I argue that courts should step in to fine-tune the standards for software patents. Specifically, the Federal Circuit requires a grossly inadequate level of disclosure to support very broad claims. If the Patent Office is going to be granting broad software patents, the public is entitled to a complete understanding of how to implement the full scope of those claims. And § 112, not § 101, is the tool it should use to regulate software patents. By requiring more disclosure, software patents will be more useful for follow-on innovators, and issue with narrower, less troublesome claims.

The patent system is in need of reform, but that doesn’t mean abolishing software patents. Instead, the Supreme Court should reverse the Federal Circuit, and hold that Alice’s patents are patent-eligible. On remand, the Federal Circuit and the District Court should start the process of more diligently enforcing the other conditions of patentability, specifically requiring more detailed specifications of how the invention works under § 112, and limiting the claims to that description.

Parker Hancock

Image Source

Comments are closed.