Shane R. Thielen is a patent attorney at Suiter Swantz IP and teaches Copyright Law at Creighton University School of Law. His course focuses primarily on the Copyright Act of 1976 and considers the types of works protected, the requirements for protection, and the scope of protection. Shane is registered to practice before the USPTO and is also a member of the Patent Bar, and the Nebraska State Bar.
After a decade-long legal battle between Google and Oracle, the Supreme Court has ruled in favor of Google – changing the landscape of software protection.
Computer software is a relatively recent addition to copyright law. Beginning in 1976, and with more recent amendments, software was explicitly added to the definition of a “writing:” a constitutional requirement for copyright protection. There has been tension ever since.
Even after the 1976 amendment, significant legal cases centered around whether compiled code (code converted to machine language) was protectable or just a functional part of a computer; whether a user interface was protectable; and whether protection extended beyond the literal written code. At every turn, the scope of protection for software expanded. Ultimately, copyright protection for software extended all the way to flow-charts, and the organization of classes and modules.
And so we come to Google LLC v. Oracle Am. Inc. The Android operating system, developed by Google, has an application programming interface (API) similar to Java SE, developed by Sun Microsystems, which Oracle later acquired. When developing the Android operating system, Google wanted to leverage the large group of existing Java developers to produce software for the Android platform. Google tried to enter into an agreement to license Java. When no deal could be reached, Google’s engineers produced an original operating system including class names, module names, and parameter organizations that were identical to corresponding classes and modules in Java; none of the functionality (that is to say the code that makes the computer do work) was copied. Oracle sued Google for violating its copyright in Java, including the high-level structure of the language.
In a series of high-profile legal decisions spanning more than a decade, courts, including the Supreme Court, handed Google several defeats. First, the overall structure of the Java language was held to be an original, creative expression, and therefore subject to copyright protection. Second, Google argued the Java structure had become standard, stock elements that everyone needed to be able to use (the doctrine that standard elements or necessary elements should be free to everyone is called “scenes a faire” or “merger”), but that argument was unpersuasive.
Finally, in what seemed to be its last chance, Google argued that its use of the Java structure was a “fair use.” Fair use is probably the most challenging part of copyright law; copying a work is not an infringement (i.e., a fair use) if the use is transformative, limited to what is necessary, and not intended to coopt the economic value of the original. You could spend the rest of your life trying to pin down all the factors of a fair use analysis.
In a ruling issued on April 5, 2021, the Supreme Court ruled that Google’s use of the Java structure was a fair use: expanding the structure to mobile devices was transformative, Google limited the amount copied to only class and module names necessary for mobile devices (a relatively small percentage of the total structure), and Google was not displacing any Java product for mobile devices. Perhaps most critically, the Court noted that the class and module structures at issue performed an “organization function” and therefore were more open to fair use.
Congress and courts have been very pragmatic about balancing the public’s interest in free access to ideas against the economic necessity of protecting software developers’ products. For several decades, that balance has taken the form of allowing copyright protection for aspects of software that may be more functional than artistic expression (and therefore more properly protected by patents than copyrights). In a nutshell: patents are intended to protect functional embodiments of an idea; copyrights are intended to protect artistic expressions of an idea. Software has always occupied a nearly unique position because it takes the form of a written that includes creative expression, but it also causes a machine to perform a function. The line for copyright protection has always been functionality; the machine’s actual functionality is not protectable by copyright, but all of the code that facilitated that functionality was protectable.
With the Supreme Court’s April 5 ruling, the scope of software copyright protection appears to be scaled back. In a dissenting opinion, Justices Clarence Thomas and Samuel Alito thought the decision “makes it difficult to imagine any circumstance in which declaring code will remain protected by copyright.” That may be a substantially accurate prognostication.
Going forward, software developers will have to give more serious thought to the mechanisms they employ to protect their software. Copyright can be relied upon for the lowest level of code (the code that implements the modules’ functionality), but probably not for the most abstract and highest level of code (flowcharts and class and module definitions). Software developers will have to give renewed thought to patent protection.
Software developers have been split on patent protection; some prefer an open-source model; others prefer every available protection mechanism. The economic reality is that copyrights are cheap and easy to obtain; patents are costly and require substantial examination to be issued. As long as copyright protection seemed to cover everything, patent protection was easy to discount. That can no longer be the case.
Software patents have had historic reshuffling over the last fifteen years. Business owners that produce and sell (or license) software based on unique functionality will have to be familiar with the patent system, and how it applies to software. Software that improves the operation of a computer and software integrated into a novel process is patent-eligible; software that is a computerized version of standard number crunching is not patent-eligible but is still copyrightable.
It will be the business owner’s responsibility to assess early in product development whether the functionality embodied in the software is sufficiently new to seek advice about patentability, and recognize that your competitors will be doing the same thing.
Suiter Swantz IP is a full-service intellectual property law firm providing client-centric patent, trademark, and copyright services. If you need assistance with an intellectual property matter and would like to speak with one of our attorneys, please contact us at info@suiter.com.