Software patents have been the subject of controversy for some time. According to the most recent Australian court decision Commissioner of Patents v RPL Central Pty Ltd1 (RPL Central), the relevant question is whether the invention is in substance a scheme or plan or whether it can broadly be described as an improvement in computer technology.
Australia has a thriving software economy and a growing number of Australian start-ups innovating in the software space. Many companies in this area use patents as a cornerstone of their commercialisation strategy. Larger companies on one hand use their patent portfolios to maintain their dominance in the market. This includes filing patent applications that form barriers to protect new products or incremental advances in their existing products.
Start-ups on the other hand generally use patent applications as formalised assets in fundraising rounds and most investors require at least pending patent rights to protect their investment in the idea. In the worst case, when the start-up is not successful, the underlying technology covered by patent rights can still be monetised by the shareholders.
While these uses of software patents clearly benefit the economy, software patents have also been the subject of controversy for some time. One reason may be that software technologies offer a relatively low entry barrier to the market when compared to tangible products. A few clicks can allow a start-up to readily bring a new app to market and commercialise it on Google Play or Apple’s App Store.
However, a patent to such an app would often not fulfil the very rationale of having a patent system. That is, patents exist as an incentive to contribute to public knowledge so that everybody in the field can progress their technology. Naturally, this equips competitors with the knowledge to recreate the invention, which is why the patent system also empowers patent owners to stop others from using their invention or request a licence fee during the limited life of a patent.
The easy creation of that app also means that the contribution to the public stock of knowledge is sometimes of limited technical value. In particular, business methods and financial schemes where the invention lies in the way of carrying out a business or in the financial scheme itself constitute little technical progress. This is one reason why patent offices globally do not grant patents to business methods or financial schemes per se.
However, it is often difficult to distinguish inventions that inherently have inventive technical features versus business methods or financial schemes that include technical features as a façade to give the appearance of a technical invention. It has been left to the courts to decide on particular examples and it has led to some frustration by patent owners to see these decisions being applied to their inventions in an interpretation that is often considered as overly restrictive.
The origins of the law on software patentability are surprisingly old, that is, almost 400 years. The Statute of Monopolies of 1624 makes reference to a “manner of new manufacture” and it appears this was meant to include basically every tangible product that is vendible. Under this statute, there was uncertainty as to whether methods can be patentable inventions, which was addressed in 1959 by the High Court in the often cited National Research Development Corp v Commissioner of Patents2(NRDC case).
This is important here because computer software is fundamentally a list of instructions for a processor to perform a sequence of method steps.
The NRDC case sets out that methods need to result in an “artificially created state of affairs”.3 This means that there must be some physically observable effect of the invention. In this particular case, the invention was for applying specific chemicals to crop-bearing land to eradicate weeds and the physical effect was the weed-free condition of the land4 as a result of performing the method.
So at this stage, financial technology (fintech) inventions, for example, that are not performed by a computer would not have been patentable because the exchange of funds did not result in a physical effect. For example, entering into a reverse mortgage scheme at the abstract level has no physically observable effect but instead results in legal and financial obligations.
But what if the mortgage scheme is implemented in a computer? The data stored on memory as a result of transferring funds electronically would definitely be physically observable as otherwise the parties could not rely on the funds being actually transferred. Alas, two recent Australian decisions by the Full Court discuss this question:
which were both implemented as computer software.
Without going into too much detail here, the general gist is that the NRDC case requires a physical effect not just generally, but the invention itself has to result in that physical effect. Despite the computer implementation, the courts found that the invention were to a scheme or business method7 and therefore not patentable.
RPL Central is particularly useful because the judges apply the “substance of the invention” test to a
generation of questions by a computer, which are then presented to the user. The judges commented on the facts of the case that “it is not suggested that the presentation of the questions or the processing of the user’s responses involve ingenuity themselves".8 The judges also observed as a matter of law that:
Simply putting a business method or scheme into a computer is not patentable unless there is an invention in the way in which the computer carries out the scheme or method.9
Consequently, the application was rejected.
These cases also highlight the importance of determining the substance of the claimed invention with consideration to what is described in the specification. This allows a person making the determination (being a judge, hearing officer, or examiner) to consider the alleged invention holistically rather than as a matter of form of the words in the claims.10 Importantly, this allows that person to exclude from patentability alleged inventions that are, in essence, a pure business method or financial scheme that has been dressed up with a façade of technical features by a patent attorney.
Court decisions are, of course, specific to the set of facts that are presented to the judges. Most patent applications, however, present a different set of facts. Before granting a patent, the examiners of IP Australia need to apply the legal principles from the above decisions to the facts of the case before them.
When examiners now assess a computer-implemented invention, they first determine the “substance of the invention”. In general, the claims are the part of the patent application that defines the invention. However, the examiners do not limit their assessment of the substance of the invention to what is defined in the claims. Instead, examiners can consider the specification “as a whole”, which may include the description with potentially little regard to the claims.
In cases where the specification clearly outlines a technical problem, such as processing time, storage requirement or battery consumption, the examiners generally move on to examine the further requirements, namely novelty and inventiveness of the claims.
In the more disputed cases, however, the examiner concludes that the substance of the invention lies in a “mere scheme”. For example, where the application describes a bidding platform, the examiner would conclude that the substance of the invention is a scheme for placing bids and choosing the highest bidder. This conclusion is typically reached irrespective of any technical features of the claims. This also means that amending the claims by adding technical features in response to the examiner’s objection is unlikely to change the conclusion that the specification as a whole is directed to a mere scheme.
Once this conclusion is reached, the examiner considers any technical features in the claims disregarding those that are well-known. For example, a computer performing well-known search and processing functions11 would not contribute to the solution of the technical problem.
This means that once a “mere scheme” conclusion is reached, the claims require technical features that are inventive per se — for example, a new battery with increased capacity or a new processor with increased computing power. Of course, most applications involving business methods or financial schemes do not involve new hardware inventions.
It is noted that under current practice, examiners do not provide evidence showing that the claimed features were well-known. This is a source of frustration and makes it difficult to respond to such objections. In the above decisions, the judges have also not provided guidance on how to assess whether claimed features are well-known.
A result of the current practice is that applications in the area of fintech or business methods require ingenuity in the way the computer is used. When drafting new applications in these fields, it is often possible to focus on technical advantages of a particular implementation, but for applications that were drafted years ago, responding to such objections can be difficult.
The approach by US examiners appears very similar but has fundamental differences. US examiners first determine whether the invention is directed to an “abstract idea”, which is, unlike Australian practice, based on the claims. The second step is to assess whether there is “significantly more” than the abstract idea. “Significant” in this context means that the claimed feature is part of the inventive concept. So again, it is not sufficient to add general computer language but instead, the computer features need to contribute to the inventive concept. For example, software that provides real-time stock market data would be considered an abstract idea but intelligently downloading the data when the device is online for offline use and sending SMS alerts to the user via SMS with links to access the information would be significantly more than the abstract idea.
European patent law also has a requirement for patentable subject matter, but almost all patent applications meet that requirement by formally claiming a “computer-implemented method”, for example. The substantial hurdle in Europe is the inventive step, which can only reside in the technical features of the claims.
Therefore, European examiners determine which claim features are technical. If the only technical feature in the claim is a general purpose computer, the examiner would typically raise a prior art document that discloses a general purpose computer and conclude that the claims are not inventive, irrespective of any inventive but non-technical features.
The Australian practice on computer-implemented inventions remains in a state of flux and there is a need for further cases that proceed to the courts for clarification. Nevertheless, pending patent rights (before examination) are commercially valuable to many software businesses for funding rounds or creating risks for competitors. The next article in this series will provide an opinion on the current practice, a suggestion of best practice and a prediction of where Australian patent law will proceed in the future.
This article originally appeared in the April 2017 LexisNexis Australian Media, Technology and Communications Law Bulletin and has been republished with permission.