Intangible Inventions: A History of Software Patenting in the United States, 1945–1985

On March 31, 2014, at the United States Supreme Court, Justice Stephen G. Breyer asked a lawyer named Carter G. Phillips to imagine King Tutankhamun, the ancient Egyptian pharaoh, sitting on top of a pyramid containing all his gold.¹ Breyer instructed Philips to suppose that King Tut, who had the habit of handing out vouchers for free gold, had hired a man with an abacus to keep tabs on his wealth. This ancient accountant was responsible for telling the king to stop handing out vouchers as soon as the total amount of gold given away surpassed whatever amount was initially stored in the pyramid. Essentially, his job was to say "Stop" as soon as the difference between the two amounts became zero or less. In this situation, Breyer explained, King Tut had simply used a human being to implement a very simple abstract idea: to say a word when a value reaches a limit.

Phillips immediately realized that thinking about King Tut was essential to his success. The lawyer stood before the justices in representation of the Alice Corporation, an Australian firm dedicated to the acquisition of software patents for financial transactions.² Alice had sued CLS Bank International, a New York-based bank, for the infringement of four patents aimed at so-called electronic escrow services. These are transaction methods wherein a computer manages payments between two parties in order to minimize the risk that one of the parties will fail to uphold its part of the deal. In its defense, CLS had argued that Alice's patents were invalid because the inventions that they described were not patent-eligible.³ Mark A. Perry, representing the bank, argued that merely requiring a computer implementation did not render abstract ideas such as those at the core of the Alice patents eligible for patent protection.

All the major players in the computer industry were watching. Earlier that year, firms such as Google, Amazon, Facebook, and Netflix had written amicus briefs arguing that Alice's patents were improperly broad; that is, that software patents should be limited to a "specific way of implementing an abstract idea."⁴ Microsoft, Adobe, and Hewlett-Packard each argued that software implementations of processes and algorithms are patent-eligible, but that Alice had failed to disclose an implementation of this kind.⁵ Advocacy groups, including the Software Freedom Law Center, the Free Software Foundation, and the Open Source Initiative—all distinguished by their strong opposition to intellectual property protections for software—filed briefs arguing that software comprised nothing but algorithms written "in human-readable terms," and that without specialized machinery, they were altogether ineligible for a patent.⁶

These groups had submitted forty-two briefs in total, and Breyer had read all of them.⁷ He had chosen the caricature of King Tut because it helped the justice to ponder the role that a computer plays in a financial transaction. If King Tut somehow managed to use a programmed electronic computer instead of a man with an abacus, would the resulting financial management system be patent-eligible? Breyer believed that it would not, and the other eight justices agreed. On June 14, 2014, the Supreme Court handed down a unanimous decision in favor of CLS. The decision, delivered by Justice Clarence Thomas, favored CLS by ruling that requiring a generic computer implementation did not render an abstract idea eligible for patent protection.⁸

Since then, industry and legal commentators have noted that the country has entered a new era in the history of software patenting—one in which, as one blogger for the American Bar Association put it, it is now "open season on these patents."⁹ Hundreds of writers have taken on the task of speculating what the future of software patenting in the United States will look like in the aftermath of Alice. Software firms might now turn to copyright law and trade secrecy more eagerly than ever before, or they might entirely abandon traditional intellectual property (IP) protections such as patents and copyrights. Perhaps patent agents will need to develop...