Tuesday, May 10, 2016
Software Patents Will Soon Rise Again
By Joshua Van Hoven
(originally published in Daily Journal Newswire)
Software patents have had rough run over the last few years. Courts have declared many software-based inventions invalid and the U.S. Patent and Trademark Office has denied patentability for large numbers of software-based applications. This trend has reached its zenith. Whether through the courts or legislative action, a gradual loosening of restrictions on patentability of software is likely in the near future.
Patent protection for software patents has varied wildly over the last 20 years. The meaning of the term "software patent" is itself the subject of debate and can cover different drastically different inventions in different contexts. For purposes of this article, a software patent will be thought of as a patent where the novel aspects of the patent claims are performed by software operating on a device. Under this definition, a software patent includes highly technical inventions, such as the communication protocols and procedures that enable massive amounts of data to be transmitted wirelessly and the code that controls how a robot navigates through a complex physical environment.
In each of these cases it is the code - not the hardware - that is the source of the new and useful functionality. This definition of a software patent also encompasses many other software applications such as business methods that operate on a computer.
Courts and the PTO struggled with software patents for decades. In addition to the typical requirements that a patent claim be novel and non-obvious, software patents were often challenged as lacking patentable subject matter. Courts and practitioners had difficulty distinguishing between patentable software inventions and those that merely claimed abstract ideas, such as mathematical algorithms. This changed with State Street Bank & Trust v. Signature Financial Group, 149 F.3d 1368 (Fed. Cir. 1998). The State Street court read the existing caselaw in a manner that broadly permitted patents on software having a practical application, including for business methods implemented on a computer.
As a practical matter, patentable subject matter ceased to be an effective limitation to software patents, so long as the patent claims included a device such as a computer. The PTO saw an explosion in the number of applications for software inventions, and those patents were soon asserted in litigation.
The early 2000s also saw the rise of the so-called "patent troll." Their popular weapon were patents that implicated many potential defendants, including some questionable new patents with broad application to the internet or to business practices. The tactic of patent trolls was to assert often questionable patents against many defendants in a single lawsuit, while seeking settlements for less than the cost of litigation from each defendant.
A number of legislative fixes for the patent troll issue were enacted, such as the creation of new administrative procedures to invalidate patents under the America Invents Act. The U.S. Supreme Court also took action. In a series of cases (Bilski, Mayo and Alice), the Supreme Court reinvigorated the proposition that many software patents may lack patentable subject matter. In Bilski and Alice, the underlying methods of hedging risk and exchanging financial obligations through an intermediary were found unpatentable. Mayo was not a software case, but laid out the test that was eventually adopted for software by Alice.
After Alice, patentable subject matter is determined by a two-part test in which courts first determine whether the gist of a patent claim is directed to an "abstract idea," and if so, whether there is an "inventive concept" beyond the abstract idea. These steps are intended to prevent a patentee from preempting an entire field with an overbroad patent on an abstract idea.
Importantly, many of these Alice decisions to invalidate patents were decided on motions to dismiss, motions on the pleadings, or early summary judgment motions. This possibility of early invalidation of an asserted patent has modified the economic incentives for patent trolls. However, some courts and patent examiners have taken the position that virtually any claim to software that runs on a computer or other pre-existing hardware is invalid under Alice.
This affects not only patent trolls with weak patents, but also technology companies attempting to patent highly technical inventions. New patents on innovative software are expensive if not impossible to obtain in some instances. Existing patent assets have been significantly devalued.
The courts should have ample opportunity to clarify the proper approach to software patents as technology cases replace troll cases on their dockets. For example, highly technical software inventions should likely be patentable, so long as the software and related hardware are claimed with sufficient specificity. Where patent claims include concrete claim elements that are particular to a technical environment, the fact that those claim elements are implemented in software should not bar patentability. This approach is consistent with the principal U.S. Court of Appeals for the Federal Circuit case that upheld claims under Alice - DDR Holdings LLC v. Hotels.com L.P., 773 F.3d 1245 (Fed. Cir. 2014). Claims with sufficient specificity within a particular technical environment are not directed to abstract ideas, and do not preempt entire fields.
When faced with concrete claims directed to technical applications, the courts are likely to find that many of these patents pass muster under Alice. Some complex business-related patents may also satisfy Alice. This claim-based approach will maintain some of the most useful aspects of the Alice decision such as earlier invalidation of troll-type patents, without broadly precluding patents for a broad swath of the high-tech economy. The guidance from court decisions will improve decision making at the PTO.
By adopting an incremental turn towards patentability of some software patents, courts may also head off more drastic action such as legislation. A general consensus is beginning to form among business leaders, commentators and scholars that questions the present framework due to its lack of clear standards and potential to harm to economic growth. For example, the United States is now viewed has having the weakest patent protection for software inventions among developed economies.
Proposals for legislation to address Alice have been proposed, and a former director of the PTO has even suggested doing away with the patentable subject matter requirement in its entirety. If courts are unable to distinguish between overbroad troll patents and concrete technology patents, perhaps Congress will.
Joshua Van Hoven is a patent law partner in the San Francisco office of Maynard & Gale. The opinions stated are those of the author only, and not reflect the opinions of other attorneys at the firm, the firm as a whole or its clients.