Copyright in Software: Perspectives on Oracle v. Google

This is the second report in the 7-part series from The 20th Annual BCLT/BTLJ Symposium. The reports were written by Kevin Hickey, Microsoft Research Fellow, Berkeley Center for Law & Technology at Berkeley Law School.

A Report from the 20th Annual BCLT/BTLJ Symposium, Part 2

Oracle v. Google, 750 F.3d 1339 (Fed. Cir. 2014) is the most significant decision in many years in the field of copyright protection for software. In Oracle, the Federal Circuit held that the declaring code of Java application programming interfaces (APIs) was copyrightable, and that Google infringed Oracle’s copyright by copying the names and structure of 37 Java API packages and using them in the Android mobile operating system. The decision in Oracle, which represents a significant increase in the availability of copyright protection for non-literal aspects of software, animated many of the presentations at this year’s BCLT/BTLJ Symposium.

Professor Peter Menell spoke to how Oracle v. Google has revived a number of issues about the significance of interoperability and the scope of copyright protection in software. Until Oracle, many believed that cases like Sega v. Accolade, 977 F.2d 1510 (9th Cir. 1992), and Lotus v. Borland, 49 F.3d 807 (1st Cir. 1995), had settled these issues against copyright owners, holding that functional elements or elements necessary for interoperability were simply not protectable by copyright. Oracle has upset that equilibrium, and raised the possibility of a significantly increased scope for copyright in computer programs. Professor Menell argued that whether elements like APIs ought to be copyrightable depends on a balance between incentives for platform developers, permitting platform competition, network effects, and other economic considerations.

This led to a spirited debate between Jonathan Band and Professor Randal Picker as to whether the Federal Circuit got that balance right in Oracle. Mr. Band maintained that the Federal Circuit fundamentally misapplied Ninth Circuit precedent in Oracle, pointing in particular to language in Sega excluding elements necessary for interoperability from copyright. In his view, this exception was approved by Congress when it passed 17 U.S.C. § 1201(f) in the Digital Millennium Copyright Act. The API declaring code can be seen as a kind of “developer interoperability”: the main reason why Google copied the APIs was to make it easier for developers familiar with Java to write programs for Android. In Mr. Band’s view, API structure and declaring code is therefore ineligible for copyright under 17 U.S.C. § 102(b) as a “system [or] method of operation.”

Professor Picker took a contrary view, arguing that the result in Oracle is correct, albeit for a different set of reasons than those offered by the Federal Circuit. Professor Picker reconciles the leading software copyright decisions as being about software elements that restrict technological access (like the Sega decision), or attempts to protect functional aspects of software (like the Lotus decision). In Professor Picker’s view, Sega was wrongly decided. If it wishes, Sega should be allowed to restrict access its platform with a “technological lock,” closing it off to independent developers like Accolade in order to protect its business model: selling video game consoles at a loss and recouping the loss with game sales. Lotus, on the other hand, was about an attempt to monopolize the functional interface of a spreadsheet program, and the First Circuit got it right in allowing Borland to compete in this market on equal terms. Although he admits the case is quite complex, Professor Picker argued that Java should be viewed as a kind of platform, and that the Oracle dispute is closer to Sega than to Lotus.

Professors Joseph Farrell and Randall Davis took a step back from this close reading of the case law to provide some perspective on the economic and technical aspects of software. Professor Farrell, an economist, posited that compatibility is generally socially desirable, allowing more competition and greater individual efficiency. The tough issues are whether compatibility is an outcome that rational actors will negotiate to own their own, and if not, how society should change initial right allocations to reach that socially desirable outcome. Professor Davis of MIT offered a perspective on how intellectual property issues are seen by computer science researchers. In his view, hard problems are inevitable as software is simply not a good fit for either patent or copyright. Moreover, these problems are exacerbated when technical terms that have precise meanings in computer science—“abstraction”, “processes”, “procedures”, “systems”—are used in a very different way as terms of art in IP law.

Ronald Johnston of Arnold & Porter LLP surveyed the landscape of software copyright and reached the somewhat depressing conclusion that, despite all the ferment in the case law, most of the fundamental questions remained open and unanswered. For example, the basic legal test for software copyright infringement—the “abstraction, filtration, comparison” test—is really only an approach, i.e., a starting point and not an actual test. We still do not know at what level of abstraction copyright protection ceases to attach. As a result, it is not clear when—if ever—user interfaces, program organization, sequencing, or the external structure of a program are protected by copyright or instead filtered out of the infringement analysis. The role of interoperability and the scope of fair use are also not clearly established, as the dispute in Oracle demonstrates. In Mr. Johnston’s view, the most that decades of judicial development have clearly established is that software is treated as a literary work, drawing the line between protectable expression and unprotectable elements is ad hoc and contextual, and software elements truly dictated by functional or efficiency constraints will not be protected.

Related Resources

Jonathan Band – Oracle v. Google and Interoperability

Randall Picker – Access Versus Functionality

Ronald Johnston – Will Lightning Strike Twice? (pp. 3–14)

Randall Davis – A View from the Technical Side (pp. 15–26)

Read more on the BCLT/BTLJ Software IP Symposium

• Intellectual Property Protection for Computer Programs: Past, Present & Future – Introduction
• A Brief History of Software IP – Part 1
• The Elusive Meaning and Impact of Alice – Part 3
• Functional Claiming Under Williamson v. Citrix – Part 4
• Trade Secret Protection for Software – Part 5
• Empirical Studies of IP Protection for Software – Part 6

This report is written by Kevin Hickey, the Microsoft Research Fellow at the Berkeley Center for Law & Technology, UC Berkeley School of Law. Mr. Hickey’s scholarship on intellectual property law and innovation policy is available online at SSRN.