The Supreme Court Copyright Battle Between Google and Oracle Will Confront “Fair Use” of Software
Android did, however, reuse aspects of the Java platform. Android replicated the syntax and structure of declaration codes associated with 37 Java API libraries. Reusing these Java APIs allowed third-party developers to use familiar Java declarations and commands to write applications that Android would recognize.
In total, Google copied 11,330 lines of Java code including the structure and arrangement of the copied code. Despite reusing the Java code, it took almost 100 Google engineers over three years to build Android. The copied Java APIs represents less than 0.1% of the over 15 million relevant lines of code in Android.
Using Java code, Android developers have created millions of applications used by more than a billion people. Between 2007 and 2016, Android generated over $42 billion for Google. However, Android applications are incompatible with Oracle’s Java platform and Java applications are incompatible Android devices.
In 2010, Oracle sued Google for copyright infringement. Oracle estimates its damages to be $9 billion. The jury concluded that Google’s use of the copied Java APIs was “fair” and therefore immune from copyright infringement and damages. The Court of Appeals reversed the jury verdict and ruled that Google’s use was not fair, wiping out Google’s defense to copyright infringement. Google appealed the infringement and fair use holdings to the Supreme Court.
This case will turn on whether the Java APIs copied by Google are eligible for copyright protection and if they are, was Google’s use of the Java APIs considered to be fair and therefore immune from infringement of Oracle’s copyrights.
The copyright eligibility question asks whether the API code that Google copied can be copyrighted in the first place. Copyright is meant to protect “creative expression,” not functions or methods of operation.
Google argues the Java APIs it reused in Android were functional – namely, allowing its operating system to recognize commands written in Java, a computer language that many developers know (and created by Oracle). Oracle, on the other hand, claims Google made use of Java’s creative expressions and elegant organizational framework. Oracle claims Google could have taken a commercial license to reuse the Java APIs or created its own organizational framework for implementing the API functions. Instead, Google copied the Java APIs, including their copyrightable expression and organization.
U.S. Federal law provides in 17 U.S.C. § 102 that:
Copyright protection subsists . . . in original works of authorship fixed in any tangible medium of expression . . . from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device.
However, § 102 also states that “in no case does copyright protection . . . extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery.”
Oracle argues that the expression and organization of Java APIs are eligible for copyright protection. The Java APIs copied by Google are “original” because the copied code is meant to be readable by humans. The copied code explains to app developers a function of the code, how the computer will use the code, and how the code relates to other parts of the Java platform. The copied code could have been written in countless other ways and Oracle’s creative choices represent an expressive, copyrightable work.
Oracle also argues that aside from the expressiveness of the copied Java APIs, its copyright covers “structure, sequence and organization” (“SSO”) of the code. A computer would run perfectly fine if thousands of lines of code were all stored in single file. However, such an arrangement would be difficult for humans to use. Java’s authors determined relationships and built interdependencies between different Java modules. Oracle argues that the SSO itself reflects creative choices that were critical to Java’s widespread adoption and are therefore independently copyright eligible.
Google counters that the copied Java APIs and its SSO are not eligible for copyright protection because of the “merger doctrine.” The law states that “in no case does copyright protection . . . extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery.” Under the merger doctrine, when there are only a few options to express a particular function, any potentially copyrightable expression in such content “merges” with the functionality of the content, obliterating copyright protection.
To utilize the Java APIs, code must strictly adhere to the Java syntax and SSO. No other code can perform the Java API functionality. Google utilized the copied the Java APIs only to the extent necessary to achieve functionality of the APIs. Therefore, the merger doctrine dictates that any creative expression of the copied Java APIs “merges” with its functionality, obliterating copyright protection. Without applying the merger doctrine, Oracle’s copyrights would convey an exclusive right to the functionality of the copied Java APIs, a right that should be properly secured by the patent laws – not the copyright laws.
Note – Google cursorily argues that even without the merger doctrine, the copied Java APIs are rote, de minimis instructions that do not include enough creative expression to be eligible for copyright protection.
If the Court determines the Java APIs at issue are eligible for copyright protection (as Oracle argues), it will then evaluate whether Google made “fair use” of the code. Copyright law allows people to make limited use, or “fair use,” of copyrighted material without infringing the copyright. Google argues that even if the copied Java APIs are eligible for copyright protection, the jury correctly found that its use of the code was a non-infringing “fair use.”
The copyright statute, in 17 U.S.C. § 107, provides that:
[F]air use of a copyrighted work . . . for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright.
The statute lists the following four factors that should be considered to determine use of copyrighted material is fair:
(1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;
(2) the nature of the copyrighted work;
(3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and
(4) the effect of the use upon the potential market for or value of the copyrighted work.
The Court will weigh the four factors to determine if they favor Google (that the use of the Java APIs is fair) or Oracle (that Google infringed Oracle’s copyright).
With respect to factor #1, Google argues it “transformed” use of the copied Java APIs by implementing them on the constrained operating environment of mobile devices. This new implementation gave new expression and meaning to the copied code. Although creation of Android was a commercial endeavor, any commercial advantage was due to the underlying code implemented by the copied Java APIs – which Google wrote on its own.
Oracle counters that to be transformative, a use must change the expression or meaning of the copied material. Every line of code copied by Google has the same meaning and serves the same purpose in Android as it did in Java. Factor #1 should focus on whether the copied material itself was transformed. Exemplary transformative uses of the copied Java APIs would include copying to critique the code or for research on how to make a program that does not infringe.
Under factor #2, a more creative work is accorded stronger copyright protection. An informational or functional work is more likely to be entitled to a fair use defense. Google argues that substantial evidence indicates that the copied Java APIs were primarily functional, not creative. Therefore, the jury reasonably concluded that the copied Java APIs were entitled to minimal copyright protection and reusing the code was a “fair use.”
Oracle argues that even minimally creative works are entitled to copyright protection against copying of their expressive elements. The copied APIs were designed to be read by humans and its expressive elements were copied by Google.
Factor #3 looks at (1) the amount of material that was copied relative to the entire copyrighted work and (2) accounts for the importance of the copied material relative to the entire copyrighted work. Google argues the copied Java APIs comprise less than 0.5% of the code in the Java SE libraries, which are themselves only a subset of the entire Java SE platform. Evidence was presented to the jury that although the copied Java APIs were necessary to allow code written in Java to work on Android, the copied code had no value independent of the underlying implementing code which Google independently wrote.
Oracle argues that “statistics cannot trump quality.” The copied code included “central” and “important” Java packages.
Courts consider factor #4 to be the most important in the fair use analysis.
Google argues that use of the copied code (including their SSO) caused no harm to the market for the Oracle’s copyrighted works. Evidence was presented that Android did not supplant or supersede the market for Java. Rather, Java was designed for servers and desktop computers and is not suitable for the smartphone market.
Oracle contends that its commercial customers leveraged the availability of Android for steep discounts on Java licenses. Furthermore, Google’s copying harmed the potential market for Java. For example, the 2005 licensing negotiations with Google show that Oracle was attempting to license Java for the smartphones market.
Oracle filed its brief last week, and Google will have an opportunity to file a reply brief in March.
The numerous amici briefs (25+) filed in this case raise various policy and legal concerns. Most of the amici briefs support Google’s position and argue that a ruling for Oracle will have an adverse impact on innovation. In a future report, we expect to provide an overview of key amici positions and Oracle’s responses.