On 26 May 2016, a jury in the US District Court for Northern California handed down a unanimous verdict that Google’s use of Oracle’s Java Application Programming Interfaces (APIs) in its Android Operating System for its smartphones was a fair use of - and therefore did not infringe Oracle’s copyrights in - those APIs. Computer programmers momentarily breathed a collective sigh of relief.

However, the underlying decision of the US Court of Appeals for the Federal Circuit remains: Oracle’s Java APIs are protectable by copyright, with an attempt to have this aspect of the decision reviewed by the US Supreme Court having already been denied.  So in the US, APIs may still be copyrightable but any copyright claim may be subject to a fair use defence (an inherently fact dependent defence). Oracle is reported to be intending to appeal against the finding of fair use.  This is all, quite frankly, a bit of a mess, and a mess of copyright’s making.

An API is a type of interface: it is a set of functions and protocols designed to facilitate interaction between computer programs. In Oracle v Google, noting that the Java API was comprised of packages, classes and methods, the judge described it rather more elegantly as follows “An API is like a library. Each package is like a bookshelf in the library. Each class is like a book on the shelf. Each method is like a how-to-do-it chapter in a book. Go to the right shelf, select the right book, and open it to the chapter that covers the work you need.”  An example would be the Google Maps API that allows a website developer to embed Google Maps on their website. It’s important to note here that the manner in which an interface such as an API is implemented (i.e., the coding which is written to complete the functions and protocols) as distinct from the API  may also be protected by copyright.

Historically, courts have drawn the lines of copyright protection between ideas and the expression of those ideas, with only the latter (rather than the former) being  copyrightable. In practice, it is often difficult to drawn the line between these two points.  At what point does an idea morph into an protectable expression of an idea?  Can you protect the plot of story, the format of a TV show or the API of a computer program?

The value of what the likes of Oracle offer is inextricably tied up in the functionality of their products and services. If they can ‘monopolise’ that functionality they can go some way to preserving its value.  And copyright is often the weapon of choice deployed to try and do so.  As such the idea/expression analysis is being tested to its limits.  There is great concern about the US Court of Appeals’ decision and enough to have prompted 77 eminent computer scientists to petition the US Supreme Court to review it (albeit without success). There is also a risk that stretching copyright too far will have anticompetitive consequences by foreclosing development and start-ups and providing de facto patent protection to functional elements of computer programs.

Within the EU similar challenges arise. Under the Computer Programs Directive protection applies to the expression (in any form) of a computer program but not ideas or principles including those which underlie its ‘interfaces’ such as APIs.   Some of the provisions of the Computer Programs Directive also reflect a policy choice favouring interoperability – for example, use of the underlying source code of a program is permitted where necessary to achieve interoperability.   As such, one would expect the courts in the EU to adopt a fairly permissive approach to the use of APIs, at least to the extent that they define the rules and methods with which another program must conform to interoperate with an existing program. That remains to be seen. Conversely, the courts in the EU will be more inclined to consider a specific ‘implementation’ of an API as copyrightable.