Computer implemented business method patents – Where do we stand?

Greg Potter

By Greg Potter

The patentability of computer related business methods in Australia was discussed in two recent Federal Court decisions: Research Affiliates[1] and RPL Central.[2]  These decisions have created uncertainty in that the Research Affiliates decision appears to provide a very difficult subject matter test for patentability while the RPL Central decision provides an easier subject matter test where most business methods that are anchored to a physical technology would be considered patentable.

The Patent Office has been reluctant to acknowledge the RPL Central decision and is currently relying on the Research Affiliates decision.   However, we have had some success in arguing that the RPL Central decision should take precedence in cases where the facts support relying on the RPL Central decision instead of the Research Affiliates decision.

Contact Ernest Graf for more information

Contact Ernest Graf for more information

This article considers the law regarding the patentability of business methods as it stood prior to the RPL Central and Research Affiliates decisions, and the impact of these decisions on the current law. The Federal court last assessed the patentability of business methods in detail in the Full Federal Court decision of Grant in 2006.[3]  In Grant, the Full Federal Court set out basic requirements for a method to be patentable.  These requirements included that a patentable method belong to a useful art as distinct from a fine art.  Thus a patentable method must have some “industrial or commercial or trading character”, and there must be some “useful product”, physical phenomenon or effect resulting from the working of the method.

The Full Federal Court reaffirmed the view of previous cases that “‘intellectual information’, a mathematical algorithm, mere working directions and a scheme without effect are not patentable”.

With regard to methods implemented in computers or other physical devices, the Full Federal Court considered that a method may still be patentable, in the absence of a physically observable end result in the sense of a tangible product, if “an artificial state of affairs”, in the sense of “a concrete, tangible, physical, or observable effect”, is produced by the application and operation of the method in the physical device.

Grant’s invention was an asset protection method involving establishing a trust, an owner making a gift to the trust, the trustee making a loan from the trust, and the trustee securing the loan by taking a monetary charge over the asset.  The Full Federal Court considered Grant’s invention not to be patentable, because it was a mere scheme, an abstract idea or mere intellectual information that had no physical consequence.

In the recent decision of Research Affiliates, a computer implemented method for generating an index for use in securities trading was claimed.  The method included accessing data, processing data to identify a selection to include in an index, accessing a weighing function configured to weight the selections, and applying the weighting function to generate the index.

Emmett J echoed the view of the Full Federal Court in Grant, stating that “a mere scheme, abstract idea, or mere information, is not, of itself, patentable.  Some physical effect is required.”  Emmett J considered that the only physical result generated by the method in Research Affiliates was a computer file containing an index, and that the index was simply information.  Emmett J added that there was nothing in the claim or the specification to describe how the computer implemented steps, such as accessing data and applying the weighting function, were performed.  In view of this, Emmett J considered there to be no “economically valuable artificially created state of affairs” resulting from the use of the computer in the claimed invention.

Emmett J additionally noted that “the steps could readily have been carried out manually” and that implementation of the method by means of a computer was “no more than the modern equivalent of writing down the index on pieces of paper”.

In RPL Central Middleton J appears to have taken a contrary view.  This case concerned a claimed method of gathering evidence for assessing an individual’s competency relative to a recognised standard.  The method included retrieving criteria via the internet, processing the criteria to automatically generate corresponding questions, presenting the questions via the internet to an individual, and receiving a series of responses to the questions from the individual via their computer.

It appears that steps similar to this method could be carried out manually without a computer.  However, Middleton J disagreed that this was a requirement of the law, stating that “one should not subtract from the invention any aspect of computer implementation, and then determine whether what remains is the proper subject matter of letters patent”.  Middleton J also considered there to be no requirement under Australian law that the physical effect be substantial or central to the invention.

Middleton J ruled that the invention was patentable because “the generation of questions for and presentation of questions to the user creates an artificial state of affairs in their computer”, in that there is “a retrieval and transformation of data into questions, and a corresponding change in state or information in a part of a machine”.  Middleton J considered each of these events to be “a physical phenomenon in which the effect of the invention may be observed”.

Middleton J considered the decision in RPL Central to be distinguishable from the decision in Research Affiliates, because the specification and claims in issue in RPL Central provide significant information about how the invention is to be implemented by means of a computer and the computer is integral to the claimed invention, such that a specific effect is generated by the computer.

Both Research Affiliates and RPL Central are under appeal before the Full Federal Court and we expect decisions in these cases by the middle of next year.  In our opinion, the reasoning of these two cases represents polar extremes regarding likely rulings of the Full Federal Court.  The Research Affiliates reasoning would create a very difficult subject matter test for business method patents in Australia, and the RPL Central reasoning would allow a finding of patentable subject matter in almost any business method patent claim having some tangential anchor to a physical technology.  The Full Federal Court is thus likely to define some middle ground between these extremes.

As previously reported by Fraser Smith, it appears that in the meantime the Patent Office is reluctant to acknowledge the ruling of the RPL Central decision and is currently relying on the Research Affiliates decision.  This practice is particularly noteworthy because Middleton J specifically referred to the Research Affiliates decision and set out reasons why Research Affiliates can be distinguished from his RPL Central decision.

We have had some success in arguing that the RPL Central decision should take precedence in cases where the facts support relying on the RPL Central decision instead of the Research Affiliates decision on the distinguishing basis provided by Middleton J.  However, not every application can be distinguished in that manner, and so the content of each application will need to be considered carefully.

Therefore, if a patent applicant has subject matter which they consider may not be distinguished from the Research Affiliates decision, we suggest delaying examination as long as possible considering that the Full Court of the Federal Court may soon put an end to the uncertainty.


[1] Research Affiliates LLC v Commissioner of Patents [2013] FCA 71.

[2] RPL Central Pty Ltd v Commissioner of Patents [2013] FCA 871.

[3] Grant v Commissioner of Patents [2006] FCAFC 120.