Iva Kopraleva is project coordinator at Sofia Platform.
When it comes to digital innovation, software patents are one of the hot-button issues that are rarely touched upon. Indeed, much of the controversy, particularly in Europe, revolves around the question whether software patents are a reasonable approach to protect intellectual property.
Should software be patented?
Broadly defined, software patents are granted for inventions that provide a useful, novel, and non-obvious solution to a problem in the field of technology. The rational behind patents is to incentivise inventors to expend their time and resources, in exchange for guaranteeing them the exclusive rights to their invention for a limited period of time.
In comparison to other fields of technology, the software industry is extremely fast-paced. In “Engineering maintenance” B. S. Dhillon estimates that the average software development life cycle ranges between one to three years. In contrast, the actual usage of a software product can be as low as a merely five years. In some instances, a new invention can turn mainstream within six months or it could become already obsolete before the patent application process is completed. Compared to the 20 years of exclusive rights that a non-software patent grants, the serious mismatch in its applicability is glaringly obvious. Further complicating the issue is that software is a building block technology which relies in its development on previously implemented solutions.
Software patents have also given rise to the phenomenon of ’strategic patenting.’ Strategic patenting essentially occurs when a company accumulates a substantial portfolio consisting of patents that are aimed at securing the company’s market position vis-à-vis its competitors. Google, for example, purchased Motorola in 2012 for a staggering $12.5 billion, only to keep a fair amount of Motorola’s patents while reselling most of the company to Lenovo two years later for a discounted $2.91 billion. Strategic patenting is also obviously a very expensive undertaking that is practically unmatchable for the smaller players in the market.
Ironically, not all software patents a company holds are necessarily useful for the products it actually produces. Big software companies accumulate large patent portfolios sometimes even just to defend themselves against potential future litigation. This is evidenced by Google’s 2013 Open Patent Non-Assertion Pledge in which the tech giant asserts that it uses its patents for defensive purposes only, and essentially promises not to sue open source software companies if they infringe upon certain Google patents. Currently, the OPN Pledge contains 246 patents, ranging from encryption and prefetching, to XML parsing and middleware.
Strategic patenting is also utilized to guard against so called ’patent trolls.’ Patent trolls are companies that accumulate large patent portfolios for the sole purpose of licensing and suing other companies for patent infringements, which can result in immense litigation costs for businesses. Apple for instance fought and lost several cases against patent troll VirnetX, and has been ordered in October 2016 to pay more than $302.4 million in damages. Equally, Microsoft settled out of court with VirnetX, paying $200 million and $23 million for patent infringements related to Skype.
Patent trolls are a particular problem of the tech industry, because software patents can be easily infringed upon unintentionally. Indeed, due to its idiosyncratic nature, laying down a precise description of a software product is difficult, which is why software patents are often times subject to interpretations that expand their actual scope.
Additionally, acquiring patents in practice is not as straightforward as it may seem. A nasty side-effect of strategic patenting is the fragmentation of patent rights and their distribution among two or more patent holders. Those so called ‘patent thickets’ have resulted in a network of dependencies in which all parties need a license from one another to produce their distinct products.
Negotiating licensing agreements with other patent holders is a difficult task altogether.
Although the details from broad licensing deals are rarely disclosed to the public, a court order in the now infamous Samsung v. Apple lawsuit series allowed us to take a glimpse at the heavily redacted 2012 cross-licence settlement agreement between Apple and HTC; Which, unsurprisingly, was the result of settling a previous lawsuit. While the 140-page document does not reveal the specific details of the deal, it does confirm that both parties maintain non-exclusive, non-transferrable and non-sub-licensable licenses to certain patents pertaining to the other. The deal is clearly the product of intense and very expensive legal negotiations.
Patent fragmentation also creates a major problem for many small businesses and start-ups alike, who are facing the difficult task of navigating a labyrinth of patent rights and holders.
Should software be patentable?
The principle of patenting software is regularly being questioned. In general, patents are granted for distinct solutions to specific technological problems, but not for abstract ideas or even algorithms. This is precisely why patents for computer programmes ‘as such’ were excluded from the 1973 European Patent Convention (Article 52(2)(c)). However, the phrase ‘as such’ has been used to create a legal loophole. Thus, if a computer programme provide a technical contribution to the prior art, a patent may be granted according to the European Patent Organisation (EPO), which despite its name is not an official EU body.
There are several issues concerning the ‘technical contribution’ requirements of software patents: First, the scope of the patent could be overly-extensive; second, protection could be granted for a relatively trivial invention; and finally, given that there are different interpretations of what constitutes a technical contribution, patents that are granted by the EPO could be challenged in national courts, which creates an additional layer of uncertainty and ambiguity for inventors.
Despite these obstacles, the EPO is actively promoting the exception to Art. 52 concerning computer-implemented inventions, and even provides online courses on the topic. Indeed many analysts fear that the successful push for a Unified Patent Court in Europe will also result in a more lenient stance towards software patents.
An attempt by the European Commission in 2002 to formalise the software ‘as such’ exception at the EU level resulted in a serious backlash from software engineers and non-governmental organisations. Three years later the European Parliament attempted to amend the proposal in a way that significantly restricted the possibility of acquiring a software patent. Ultimately, however, the proposal was never adopted.
Compared to the United States, the European software patenting eco-system has historically been limited and remains contained to the exceptions outlined by the EPO. Recently, however, several US court rulings, and most notably the 2014 US Supreme Court decision in Alice v. CLS Bank, have dramatically changed the climate on the other side of the Atlantic. US analysts are already proclaiming the end of software patents in the United States. A recent opinion by Judge Haldane Mayer succinctly explains why software should not be patentable in the US:
“Software is a form of language—in essence, a set of instructions […] It is inherently abstract because it is merely ‘an idea without physical embodiment’ […] Software development has flourished despite—not because of—the availability of expansive patent protection. […] [T]he existence of a vibrant open source community suggests that innovation can flourish in software absent patent protection.”
So if patents are not a suitable option to protect the intellectual property behind software, what is the alternative?
From the moment a piece of software code is written down, it is immediately protected by copyright much in the same way as a book or a piece of art. This protection is qualitatively different from a software patent as it protects the actual code from being copied but keeps its functionality open. As a consequence, the open source community has the ability to compete and offer alternative products to the public.
Open source has proven to be a driving force of software development. Crudely described, it is a collaborative effort in creating software products whose code is available to anyone who wants to build upon it. Those products are also often free for the public to use. Blocking the ability of this community to use certain functionalities by payable licensing fees or through software patents is an undesirable disincentive.
In fact, stimulating the development and propagation of open source, by limiting the use of software patents and streamlining intellectual property rules at the EU level, might just be the push Europe desperately needs to become a global digital leader.
edited by Mark Leonard
Stefan Soesanto & Fosca D'Incau
Stefan Soesanto & Fosca D'Incau