The Java Wars: Oracle vs Google

The Java Wars: Oracle vs Google The Supreme Court Finally Decides in Favor of Google

Meta Description: The legal dispute between Oracle and Google began in August 2010 when Oracle filed an $8 billion lawsuit in the U.S. District Court for Northern California. Now that it has come to an end, here’s our analysis of the issues. 

In 2007, Google released the Android operating system which it had created by writing millions of lines of new computer code. However, it also copied 11,330 lines of code from Oracle’s Java platform. These lines of code were a part of a tool called an Application Programming Interface (API).

Oracle asserted that Google’s use of the Java APIs required a software license from Oracle under copyright law. Google, on the other hand, maintained that the APIs were in the public domain just like the Java programming language. As such, they were free and open for use.

The legal dispute between Oracle and Google began in August 2010 when Oracle filed a $9 billion lawsuit in the U.S. District Court for Northern California. They alleged that Google knowingly and actively violated seven Java patents and other copyrighted material connected to the Java platform in its development of the Android operating system.

This lawsuit came only seven months after Oracle acquired the Java technology and its related copyrights and patents when it finalized its $7.3 billion purchase of Sun Microsystems in January 2010. The Java software code was considered one of the prized jewels of the acquisition.

From the start of the lawsuit, most tech companies sided with Google. In fact, industry heavyweights like Microsoft and IBM filed briefs backing Google in the case. They warned that ruling in favor of Oracle could have severe consequences. According to them, it would stifle innovation and upend software development.

On the other hand, movie and recording industries, as well as publishers, supported Oracle. This wasn’t surprising as they tend to favor expansive copyright protections to shield their profits from movies, TV shows, books, articles, and music. Oracle also had the backing of the Trump administration.

The case went on for over a decade before its final resolution in April 2021. The Supreme Court finally ruled in favor of Google stating that its use of Oracle's API amounted to fair use and was not a violation of the copyright law.

With many varying views, there were two questions before the Court. The first was whether Java’s API was copyrightable, and the second was whether Google’s use of the API actually amounted to fair use. 

Was Java’s API copyrightable?

According to Google, using Java’s API was good for technical progress and was a long-settled, common practice in the industry. They argued that Java’s API was a purely functional, non-creative computer code that could not be written any other way. As such, it had no copyright protection. However, Oracle asserted that Google had committed a terrible act of plagiarism. 

Under Art. I, §8, cl. 8 of the U.S. Constitution, the purpose of copyrights and patents is to promote the advancement of science and useful arts. It does this by securing the exclusive right of authors and inventors to their respective writings and discoveries for limited periods. For a work to be copyrightable, it must meet the following qualifications:

• It must be original. The work must have at least some degree of creativity. According to the Supreme Court in the case of Feist v. Rural, the required degree of creativity is very low and even a slight amount will suffice. Additionally, the work must be the independent creation of the author. This means that the work would not be eligible for copyright if the author copied the work from elsewhere.
• It must be a work of authorship. Works of authorship include musical works, literary works, pictorial, sculptural, architectural, and graphic works, sound recordings, audiovisual works, and various other types of creative works. As of 1980, works of authorship were expanded to also include computer programs.
• It must be fixed in any tangible medium of expression. Writing a work on a computer hard drive or paper, sculpting a work out of marble, and recording a work on tape all satisfy this requirement. 

However, copyright protection, by 17 U.S. Code § 102, does not extend to any procedure, process, idea, system, method of operation, principle, concept, or discovery. This provision asserts that copyright protection extends only to the author’s expression of an idea, but not the idea itself.

The Court, therefore, in determining whether the Java APIs were copyrightable had to examine these statutory provisions. The one permitting the copyrighting of computer programs and the other forbidding the copyrighting of systems, processes, and methods of operations.

Google argued that the API’s declaring code and organization are methods of operation and are excluded from copyright protection. They based this argument on the functionality of the declaring codes and its distinction from implementing code. Although, they agreed that implementing code was copyrightable.

The Court disagreed with this argument stating that declaring code and implementing code are indistinguishably bound together. As such, if one is copyrightable, the other would be as well.

The Court went on to agree that “for argument’s sake”, the API was copyrightable. The second legal issue then was, if the codes were copyright protected, did Google violate the copyright law by using the code? Or was the use of the API “fair use”?

What exactly constitutes fair use?

Google asserted that its use of Java’s API was fair use. Under 17 U.S. Code § 107, fair use of a copyrighted work does not constitute an infringement on the copyright. In determining fair use, the law sets out four factors that the court must consider. They include:

• The objective and character of the use. Was the purpose commercial in nature or was it for nonprofit educational purposes? If the use was for commercial purposes, it may not fall under the exception of fair use.
• The nature of the copyrighted work. Was the work informative or entertaining? As seen in Sony Corp. of America v. Universal City Studios, Inc., the courts are more likely to find a determination of fair use if the material was copied from informational works. This is because it encourages the creation of new educational or scientific works, all of which benefit the public.
• The amount of copyrighted work used. How much of the work did the infringer copy?
• The effect of the use on the value of or the potential market for the copyrighted work. Under this factor, the Court tries to strike a balance between the benefit the public will gain if the use is permitted and the personal profit that the copyright owner will obtain if the use is denied.

After considering these factors, the Supreme Court ruled 6-2 that the copying of the Java API by Google constituted “fair use” and did not violate copyright law.

The Court, in siding with Google, asserted that copied lines of code, while functional in nature, were inherently bound together with ideas that were not copyrightable. It emphasized the benefits the public would derive from the use of the code, including greater creativity and competition.

The Supreme Court considered Oracle’s claim to the declaring codes an attempt to monopolize the market which makes it impossible for others to compete. Giving Oracle an exclusive right to the codes, according to them, would limit the future creativity of new programs.

The Court also pointed out that when looking at the declaring code that Google copied, the amount seems large in isolation (approximately 11,500 lines of code). However, in comparison to the entire set of the ava API computer code, which is about 2.86 million lines, the copied lines are little, only 0.4 percent.

The Court's decision is definitely a turning point in the world of software development. It enables developers to create applications that work across platforms with no fear that companies would use copyright law to prevent interoperability.

However, the Court failed to directly deal with the question of whether APIs can be a subject of copyright. This oversight will likely invite future similar lawsuits.