One of the most important copyright cases was recently decided by the Court of Appeals for the Federal Circuit, addressing Google’s use of Java code owned by Oracle. On March 27, 2018, in Oracle America, Inc. v. Google LLC,[1] the Fed Circuit held that Google’s use of Oracle’s Java code was not fair use. This ruling could have far-reaching ramifications on multiple fronts.
Because this is only the latest version of the case, here is a recap of the procedural case history. Oracle originally filed its copyright and patent infringement suit against Google alleging unauthorized use of 37 packages of the Java application programming interface (APIs). In the first district court action, the jury found Google infringed Oracle’s copyrights in the Java Standard Edition code, but deadlocked on the question of fair use. Upon the verdict, the district court entered JMOL on the copyright issue in favor of Google. In other words, Oracle’s API packages were not copyrightable as a matter of law.[2] Upon appeal, the Fed Circuit reversed, holding that the structure, sequence, and organization (SSO) of the Java APIs were eligible for copyright protection.[3] The Fed Circuit further remanded to enter the jury verdict of infringement, and for further proceedings on fair use and damages. Google then sought certiorari to the U.S. Supreme Court on the copyright issue, but it was denied.[4] At the second district court proceeding, the court bifurcated the fair use and damages issues. The current trial dealt with the fair use defense. This second jury found in favor of Google’s fair use defense. The court denied Oracle’s JMOL and entered final judgment for Google.[5] Oracle then appealed the JMOL and final judgment decisions to the Fed Circuit.
APIs are pre-written Java source code programs that act as short-cuts for creating more detailed and sophisticated code. APIs are extremely valuable for software engineers because the pre-packaged declarations, which are the basis of any programming code, allow for efficiency in writing programs without having to rewrite the entire code each time from scratch.[6] APIs play an integral and functional role in communicating between applications, and are used in hundreds of apps and programs created throughout the tech industry. The original Java SE was developed by Sun Microsystems, Inc. in the 1990s. Oracle acquired Sun in 2010. The Java SE contains 166 API packages with 3,000 classes containing over 30,000 methods. The number at issue in this case is 37 APIs. Oracle makes the Java programming language free for use without permissions for app development, but requires licensing for commercializing efforts aimed at competing platforms or embedding in devices. Strict compatibility is required of all Oracle licensees.
Google acquired Android, Inc. in 2005 as part of its mobile software development strategy. During licensing negotiations between Google and Oracle over use of Java APIs in the new Android system, a key sticking point was Google’s insistence on using the APIs free in Android with no modification limitations. As a result, negotiations broke down, and Google sought to create its own APIs, but was unable to do so. It was undisputed during trial that Google “copied verbatim the declaring code of the 37 Java API packages – or 11,500 lines of Oracle’s copyrighted code” and the “SSO of the Java API packages,” which were used as the basis for Google’s own implementing code for Android.[7] While Google allows the Android platform free to its developers under an open source license, as well as to smartphone manufacturers, it generated $42 billion in ad revenue. Android is now a widely-used mobile OS, spanning TVs, cars, wearable devices, gaming, and cameras.
The Fed Circuit took jurisdiction since it has exclusive jurisdiction over patent appeals. Although the patent claims are no longer at issue, it retained jurisdiction of the case. However, it analyzes the case using the regional circuit in which the district court sits, namely, the Ninth Circuit.[8]
Because the prior Fed Circuit ruling already held that the Java APIs were eligible for copyright protection, this appeal was limited to Google’s fair use defense.
The Copyright Act states:
[T]he fair use of a copyright work, including such use by reproduction in copies or phonorecords or by any other means specified by that section, for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright. In determining whether use made of a work in any particular case is a fair use the factors to be considered shall include –
(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.[9]
Each factor is analyzed separately, then the results are further analyzed together in light of the infringed copyright and its purposes.[10] Because fair use is an affirmative defense, the alleged infringer has the burden to prove the statutory factors are in its favor.[11] The panel consisted of Judges O’Malley, Plager, and Taranto, with Judge O’Malley writing for the court. It observed that as a procedural notation, the analytical standard is similar to an obviousness case in patents. That is, like obviousness, fair use is a mixed question of law and fact, where there is a presumption that the jury resolved all factual disputes in favor of the verdict. As such, any factual findings are kept intact unless a finding of substantial evidence to the contrary. Further, any legal conclusion is examined de novo to assess whether it is correct in light of the presumed jury findings.[12]
The panel identified these historical facts concerning Google’s use of the Java APIs, and specifically the following were agreed upon by both Google and Oracle:
- history and origin of the copyright work, including what declaring code is (i.e., Oracle created the Java APIs, and those APIs were the declaring code which was foundational for both the Java SE and Android programs);
- how much of the copyright work was copied (i.e., 37 APIs and 11,300 lines of code were copied);
- whether there were other ways to write the API packages (i.e., yes, there were other ways in which Google could have written its API packages); and
- whether the copied material was used for the same purpose as in the original work (i.e., Google’s use of the Java APIs were the same use for Android as they were originally used for Java SE).
The parties could not agree on these historical facts:
- whether the use was commercial in nature;
- whether Google acted in bad faith in copying the work;
- whether there are functional aspects to the copyrighted work that make it less deserving of protection; and
- whether there was harm to the actual or potential markets for the copyrighted work.
More on these last four facts will be discussed when analyzing the four factors.
Factor 1: Purpose and Character of the Use
The first factor analyzes the purpose and character of the use, including any commercial or nonprofit purposes. Two components must be analyzed: whether the use was commercial in nature, and whether the use was transformative.
As to commercial use, although not dispositive, use of a copyrighted work in a commercial setting will tend to weight against fair use.[13] Since it was undisputed that Google copied the declaring code and SSO from 37 Java APIs owned by Oracle for a commercial purpose, the issue turned on how much non-commercial use based on Google’s decision to make Android open source and free-of-charge would counter the commercial use of its copying of the APIs. The panel flatly stated giving something away for free that which would have ordinarily been something that customers would have had to buy constitutes commercial use.[14]
Further, the panel observed:
[T]he repeated and exploitative copying of copyrighted works, even if the copies are not offered for sale, may constitute commercial use.[15]
Commercial use is not dependent on how Google made its money, which it claims is not directly related to the use of the Android system, but rather from ad sales. However, the panel demurred, stating “direct economic benefit is not required to demonstrate commercial use.”[16] As a result, the panel found that Google’s commercial use of the APIs weighed against fair use.
As to transformative use, transformation of the alleged copied work “lies at the heart of the fair use doctrine’s guarantee of breathing space within the confines of copyright, and the more transformative the new work, the less will be the significance of other factors.”[17] Use is deemed transformative where the new work “adds something new with a further purpose or different character, altering the first with a new expression, meaning, or message.”[18]
After considering both the district court’s and Google’s argument that the copying was transformative because Google’s use of the APIs to create the Android operating system was deemed a new expression in a new platform, namely, smartphones. The panel disregarded this argument, because 1) Google’s use was not one of the listed uses in the preamble of §107 (i.e., not criticism, comment, news reporting, teaching, scholarship or research);[19] 2) the purpose of the APIs in the Android system is the same as in the Java system (namely, for a commercial purpose); 3) there was no alteration of the expressive content of the APIs when the copied APIs were used in the Android system; and 4) smartphones are not a new context (since Oracle was already using Java SE for mobile phones, Google’s use of the APIs for Android smartphones was not transformative into a new context).
The panel summed up its transformative analysis with:
[While] there is no bright line identifying when a use becomes transformative[,] [b]ut where, as here, the copying is verbatim, for an identical function and purpose, and there are no changes to the expressive content or message, a mere change in the format is insufficient as a matter of law to qualify as a transformative use.[20]
To hold otherwise, the panel noted in dicta, is to allow encroachment upon a copyright owner’s right of derivation.
An additional component which the panel analyzed was Google’s bad faith intent. Fair use presupposes “good faith” and “fair dealing” between the two parties involved.[21] The panel noted there was a sharp divergence in evidence introduced by Oracle and Google as to bad faith intent. Oracle introduced evidence that Google needed to accelerate its Android program by copying the Java APIs. On the contrary, Google introduced evidence that Sun promoted Java as a free programming language and believed the declaring code and SSO were free to use. The panel, using Ninth Circuit case law as support, found that while bad faith can weigh heavily against fair use, a copyist’s good faith cannot weigh for fair use; an “innocent intent of the defendant constitutes no defense to liability.”[22] So, even assuming the jury found good faith intent on the part of Google, it does not negate the non-transformative and commercial uses of the Java APIs.
Consequently, the panel found the first factor weighs heavily against fair use.
Factor 2: Nature of the Copyrighted Work
The second factor, nature of the copyrighted work, looks at whether a work is informational or creative in nature. While the “law recognizes a greater need to disseminate factual works than works of fiction or fantasy,”[23] creative expressions “falls within the core of copyright’s protective purposes.”[24] While not strictly creative endeavors, software is protected copyrightable material. The panel observed that while the APIs did, in fact, involve some creativity in the creation process, it found that the jurors could also reasonable conclude that the APIs’ functional aspects were more important. As a result, the panel found that this factor favored a fair use finding.
Factor 3: Amount and Substantiality Used
The third factor, the amount and substantiality of the work used, the panel quoted Harper & Row:
[T]he fact that a substantial portion of the infringing work was copied verbatim [from the original work] is evidence of the qualitative value of the copied material, both to the originator and to the plagiarist who seeks to profit from marketing someone else’s copyrighted expression.[25]
Copying an entire work generally will work against a finding of fair use. The parties during trial had stipulated that only 170 lines of actual code was necessary to write in Java. However, and it was undisputed fact at trial, Google copied 11,500 lines of code – or, as the panel noted, 11,330 more lines than was necessary to write Java. The panel interpreted this to mean Google copied substantially more than was necessary to find fair use. Quizzically, in laying out a persuasive argument against fair use, the panel concluded that the third factor was merely neutral, and weighed only slightly against fair use.
Factor 4: Market Harm
The fourth factor, market harm, looks at the potential market harm created by the infringing use. The panel noted that the most recent Ninth Circuit case law held that market harm could be presumed “where use is commercial and not transformative.”[26] However, the U.S. Supreme Court’s case law addressed the fourth factor to be balanced with the other three factors.[27] The panel opted to use the high court’s precedence rather than the most recent precedent of the relevant regional circuit in analyzing this last factor.
Evidence was introduced that Java SE had been in use in mobile devices before Google’s copying of the APIs, and Android was directly competing against Java SE. Further, other evidence showed that Oracle had licensed Java SE to Amazon for use in its Kindle tablet readers. However, after Android’s release, Amazon was able to use Android’s free status to negotiate a massive discount in its licensing fee. Further damaging, Amazon eventually opted to move the Kindle platform to the Android system. All of this evidence directly undercut Google’s argument that there was not market harm to the copying. (Google specifically argued that some Java SE was for free on the OpenJDK platform, and in doing so Oracle limited its market harm, but the panel adroitly noted that Amazon moved its business from Java to Android, not OpenJDK). As a result, the panel concluded this last factor weighed heavily against a fair use finding.
In sum, factors 1 and 4 were heavily against fair use, while factor 2 was in favor of fair use, and factor 4 was neutral, although it favored ever-so-slightly against fair use. The panel concluded that Google’s use of the 37 Java APIs did not constitute fair use as a matter of law.
Google’s Cross-Appeal
The panel last addressed Google’s cross-appeal alleging that the declaring code and SSO of the Java APIs were 1) unprotected methods of operation under patent law (17 U.S.C. §102(b)),[28] and 2) subject to the merger doctrine. The merger doctrine is a copyright principle that where expression and idea have merged together, the resulting expression cannot be eligible for copyright protection. Subject matter entitled to copyright registration is an original work of authorship fixed in a tangible medium of expression. Methods of operation or functional expressions do not qualify as “creative expressions” that is required for copyright registration.[29] The panel perfunctorily dismissed the arguments, observing that the U.S. Supreme Court already denied certiorari on the question of whether the Java APIs were copyrightable (they were) prior to the current appeal.
This case still has several more iterations left before it is completed, but as it is now, it is a devastating result for Google. The case will then be remanded back to the district court for a third time, to delve through the issue of damages, which according to Ars Technica, could potentially amount to $8.8 billion.
The effect of this case is profound, although not necessarily for Google. Android, as of the Nougat version, runs on OpenJDK, or an open source environment. If Google must pay Oracle a service bill going back eleven years, it would end roughly around Marshmallow in 2015, although the exact cutoff would have to be determined at the damages trial. Wired noted that because of the numerous Android developers, this case could force major rewrites of their product lines resulting in less interoperability and higher R&D costs. Further, it could have ramifications on cloud service providers which utilize the largest cloud provider, Amazon Web Services, and its APIs for writing their product code. What the Oracle v. Google case may do is realign the tech industry’s open and collaborative nature with that of the legal and IP industry’s focus on structure and adherence to legal standards across all media and industries. Google will probably petition for en banc hearing at the Fed Circuit, and if it is denied or loses en banc (both are high likelihoods), Google will file a petition for certiorari before the U.S. Supreme Court. Since SCOTUS already heard one iteration of this case, it is 50/50 that it might get involved again.
[1] ___F.3d___ (Fed. Cir. 2018), rev’g and remanding on damages, No. C 10-03561, 2016 WL 3181206 (N.D. Cal. June 8, 2016).
[2] See Oracle Am., Inc. v. Google Inc., 872 F. Supp. 2d 974 (N.D. Cal. 2012).
[3] See Oracle Am., Inc. v. Google Inc., 750 F.3d 1339, 1348 (Fed. Cir. 2014).
[4] See Google Inc. v. Oracle Am., Inc., 135 S. Ct. 2887 (2015).
[5] See Oracle Am., supra (slip op. at 6).
[6] Id. (slip op. at 7-8).
[7] Id. (slip op. at 10).
[8] See Atari Games Corp. v. Nintendo of Am., Inc., 975 F.2d 832, 837 (Fed. Cir. 1992).
[10] See Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 577-78 (1994).
[11] see Wall Data Inc. v. L.A. Cty. Sheriff’s Dep’t, 447 F.3d 769, 778 (9th Cir. 2006).
[12] See Kinetic Concepts, Inc. v. Smith & Nephew, Inc., 688 F.3d 1342, 1356-57 (Fed. Cir. 2012).
[13] See Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. 539, 562 (1985).
[14] See A&M Records, Inc. v. Napster, Inc., 239 F.3d 1004, 1015 (9th Cir. 2001).
[15] Id.
[16] Id.
[17] See Campbell, supra, 510 U.S. at 579.
[18] Id.
[19] See 17 U.S.C. §107.
[20] Oracle, supra (slip op. at 38-39).
[21] See Harper & Row, supra, 471 U.S. at 562-63.
[22] See Monge v. Maya Magazines, Inc., 688 F.3d 1164, 1170 (9th Cir. 2012).
[23] See Harper & Row, supra, 471 U.S. at 563.
[24] See Campbell, supra, 510 U.S. at 586.
[25] See Harper & Row, supra, 471 U.S. at 565.
[26] See Disney Enters., Inc. v. VidAngel, Inc., 869 F.3d 848, 861 (9th Cir. 2017).
[27] See Campbell, supra, 510 U.S. at 590 n.21.
[28] “In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work.”
[29] See Baker v. Selden, 101 U.S. 99, 103 (1879): “the copyright of a work on mathematical science cannot give to the author an exclusive right to the methods of operation which he propounds, or to the diagrams which he employs to explain them, so as to prevent an engineer from using them whenever occasion requires. The very object of publishing a book on science or the useful arts is to communicate to the world the useful knowledge which it contains. But this object would be frustrated if the knowledge could not be used without incurring the guilt of piracy of the book.”