I am continually amazed that for over a decade the legal, technical and academic communities have debated the issue of software patent eligibility. Many trees have been killed in the form of articles, blogs and law review articles devoted to both sides of the debate, not to mention the lobbying dollars spent. Further, the Supreme Court’s most recent ruling on the issue, Alice Corp. v. CLS Bank has left the IP bar without a clear, reliable test to determine when exactly a software (or computer-implemented) claim is patentable versus being simply an abstract idea “free to all men and reserved exclusively to none.”
This has left us with a myriad of USPTO Section 101 guidelines, flowcharts, presentation slides, reference sheets, and case law charts interpreting Alice and its progeny of district court and Federal Circuit cases—the totality of which merely confuse and distract the USPTO, courts, and practitioners from focusing properly on Sections 102 (novelty) and 103 (obviousness). The net effects are increased cost, lower patent quality, and in some instances even the denial of patent protection for software inventions.
I have always argued that software patent eligibility is a must in a country where patent rights are guaranteed by the Constitution. That is, all fields of innovation should be treated equally under the law such that one field of endeavor (e.g., pharmaceuticals or electronics) is not deemed more “patent worthy” than other fields (i.e., computer science and information technology). This is especially true when one considers how important software is to the U.S. economy as evidenced by the following facts and figures:
- The U.S. software market was valued at approx. $200B in 2016. [CompTIA, IT Industry Outlook 2017]
- A PriceWaterhouseCoopers report pegs the cumulative value of technology-related M&A activity for 2016 at US$323.5B, with software representing the largest portion of that total volume with $88.5B worth of deals, representing over 40% of the deal volume. [PwC, US Technology Deals Insights Year-end 2016 (2017)]
- In 2016, over 5.2M U.S. workers were employed in core IT occupations (g., software developers, database administrators, web developers, systems analysts, user support specialists, Information Security Analysts, and the like). Id.
- PayScale – the online salary, benefits, and compensation information company – reports that Computer Science & Engineering college major ranks in the top five for salary potential, with a median mid-career (e., 10+ years of experience) salary of $116,000. [PayScale, 2016-107 College Salary Report]
- In 2015, there were 59,281 Bachelor’s degrees, 31,474 Master’s degrees, and 1,998 Ph.D.’s awarded in the U.S. in computer and information sciences. [Institute of Education Sciences at U.S. Department of Education, National Center of Educational Statistics, Table 325.35]
- “Employment of software developers is projected to grow 17 percent from 2014 to 2024, much faster than the average for all occupations. … The main reason for the rapid growth in both applications developers and systems developers is a large increase in the demand for computer software.” [U.S. Bureau of Labor Statistics, Occupational Outlook Handbook, Software Developers]
Add to the above the fact that we are at the beginning of the Fourth Industrial Revolution – the age of the Internet of Things (IoT) – where many companies are undergoing “digital transformations.” This involves not only implementing digital processes (i.e., replacing labor with software) but also using software to analyze the big data sets collected from hardware (and people) equipped with sensors that are wirelessly connected to the Internet.
In sum, a substantial amount of U.S. commerce is software-dependent and the associated innovation in the field – when novel and non-obvious – deserves stable and predictable patent law protection! As further evidence that software must be patent eligible – on both policy and legal grounds – patent activity at the USPTO tells a compelling story. Using data from the USPTO, I sought to determine what percentage of USPTO issued utility patents are “software related.” To define what patents are “software related,” I utilized the same methodology employed by the United States Government Accountability Office in a 2013 IP litigation report to Congress. That report relied on certain classes and subclasses of applications most likely to include patents with software-related claims as selected by expert advisors to the USPTO. With respect to the selection of classes and subclasses that comprise “software patents,” these experts acknowledged:
“[I]dentifying patents with software elements can be a tricky business. … While this definition will certainly be both over-inclusive and under-inclusive, the method is calibrated to help us identify classes in which patents with software claims are most likely to be found. As shorthand, we refer to those applications or patents which falling to these classes and subclasses as “software” applications or patents, and to those which fall outside as “non-software,” with the understanding that this nomenclature is one of convenience, and will not be accurate in all cases.”
After settling on this methodology, I enlisted the assistance of patent analysis firm GreyB Services to tabulate the issuance rates of those classes and subclasses [Id. at n.7] for the last quarter-century to produce the following two figures:
While I will leave to the reader the bulk of the interpretation of the data – including year-over-year trend observations – some conclusions stand out:
- Since 2012 – the same year Mayo Collaborative Svcs. v. Prometheus Labs was decided by the U.S. Supreme Court, and upon which the Alice two-step framework rests – over half of the annual total of issued U.S. utility patents have been software related.
- The percentage of software-related issued U.S. utility patents has averaged an annual growth rate of 2.98% over the past quarter-century.
- The largest single-year percentage increase of software-related U.S. utility patents issued occurred from 1997 to 1998 – the same year State Street Bank & Trust Co. v. Signature Fin. Group was decided by the Federal Circuit.
- Since a year after the dot-com bubble bursting (e., 2001) to 2016, the percentage of software-related U.S. utility patents issued has grown 18%.
Thus, I think that Congress, the courts, and the USPTO must end their assault on Section 101 and software patents. The latter two must return to a proper focus on Sections 102 and 103. Further, the USPTO must continue to prioritize improving patent quality and pendency times. Anything less is simply counterproductive to the digital age in which we live.
This article reflects my current personal views and should not be necessarily attributed to my current or former employers, or their respective clients or customers.