Will Patent Office Practice on Computer-Implemented Inventions Be ‘Rokt’ By Federal Court Ruling?

A judge of the Federal Court of Australia has reversed a Patent Office rejection of a computer-implemented invention – more precisely, a computer-implemented business process relating to the presentation of online advertising.  In Rokt Pte Ltd v Commissioner of Patents [2018] FCA 1988, Justice Alan Robertson found that a claimed method and system for providing ‘a dynamic, context-based advertising system, introducing a distinction between an engagement offer, without a direct advertising benefit, and an advertisement designed to lead directly to the sale of the product’ is patent-eligible subject matter under the Australian ‘manner of manufacture’ test.

The basic idea of the invention, although the details are not important, is to provide a mechanism whereby a user has an opportunity to engage with an offer while accessing a web site before being presented with targeted advertising.  In this way, advertisements are presented only to those users that are most likely to interact with them, and make a purchase.  Since advertisers must often pay for placement of their advertisements within web pages, the invention provides an improvement in that the costs associated with placing advertisements in front of consumers who do not interact with them may be reduced.

A Patent Office Hearing Officer had found that ‘the substance of the invention in this case amounts to business innovation’, which was not patentable, and therefore refused Rokt’s patent application: Rokt Pte Ltd [2017] APO 34.  Rokt appealed to the Federal Court.

In upholding the appeal, the court rejected an approach to identifying ‘the substance of the invention’ based upon comparing discrete claimed elements with the contents of the prior art, and considering only the contribution made by new or unconventional elements, which has increasingly become standard practice at IP Australia in recent times.  On the face of it, then, the Rokt decision should encourage some change in this practice, although the existence of a number of relevant higher court decisions, a pending judgment from an expanded Full Bench of the Federal Court, and a high likelihood that the Commissioner of Patents will appeal the Rokt decision means that any immediate practice change is unlikely.

Perhaps the most interesting – and, in my view, somewhat alarming – aspect of the Rokt case, however, is the extent to which the parties relied upon expert evidence.  Whether or not a claimed invention comprises patentable subject matter, i.e. is the kind of ‘thing’ for which a patent may be granted, is supposed to be a question of law.  Traditionally, therefore, it has not relied upon an extensive factual background, and particularly not upon evidence as to the state of the prior art, which is more properly the province of enquiries into whether an invention is novel and/or involves an inventive step.  However, largely due to the recent evolution of Patent Office practice, and in anticipation of the arguments the Commissioner was likely to run in the appeal, Rokt clearly felt obliged to back up its case with evidence from an expert witness, causing the Commissioner to follow suit.  Rokt’s expert was even cross-examined in court.

This does not seem right.  I therefore think it important to understand how we got here (spoiler alert – in my view it is IP Australia’s fault) and where we might be headed (hopefully back to more sensible ground, but that is in the hands of the Full Court).

Extensive Evidence

Paragraphs [12] to [171], i.e. three-quarters of the 217 paragraphs making up the complete judgment in Rokt, are devoted to a discussion of the evidence, which included two affidavits of the expert engaged by Rokt (Professor Karin Verspoor), an affidavit of the expert engaged by the Commissioner (Mr Scott Ries), and an account of the cross-examination for Professor Verspoor at the hearing.  I find this vaguely disturbing.  It is a more extensive discussion of evidence than you would often see in decisions relating to traditionally fact-intensive inquiries, such as the assessment of inventive step.  The cost to the parties in adducing this evidence, and to the court in considering it, would have been substantial.  Yet all this effort and expense has been directed to evaluation of something that should be relatively straightforward and coarse-grained ‘threshold’ legal test of patentability.  The situation might be laughable, were it not so serious.  The potential now exists for any case involving questions of patentable subject matter to involve such extensive evidence.  Indeed, it would be foolhardy for anyone litigating on this ground not to adduce expert evidence in support of their arguments.

What is more, this problem is not confined to the Federal Court.  Such evidence could be pertinent in oppositions before the Patent Office, and in the course of patent examination, where applicants may now consider submitting declarations from technical experts in support of the patent-eligibility of their claimed inventions.

So how did this happen?  How did the threshold test for an invention to be a ‘manner of manufacture’ – so succinctly explained by the High Court in National Research Development Corporation v Commissioner of Patents [1959] HCA 67 (a.k.a. ‘NRDC’) as requiring ‘an artificially created state of affairs’ whose value lies ‘in the field of economic endeavour’ – become such a monster?  Unfortunately, I think the answer is that the Patent Office, in seeking to ‘clarify’ what is, and is not, patentable in the field of computer-implemented inventions, is largely responsible for complicating the law in this area.

Early 2000’s – Simpler Times

Following a series of favourable decisions by the Federal Court of Australia in IBM Corporation v Commissioner of Patents [1991] FCA 625, CCOM Pty Ltd  v Jiejing Pty Ltd [1994] FCA 1168 (Full Court), and Welcome Real-Time SA v Catuity Inc [2001] FCA 445, by the early 2000’s the availability of patents in Australia for inventions implemented by computer software, across most fields of endeavour, was considered largely settled.

The 2006 ruling of a Full Bench of the Federal Court of Australia in the case of Grant v Commissioner of Patents [2006] FCAFC 120, refusing a patent to a scheme for asset protection via specified legal and financial transactions, confirmed that ‘pure’ business methods which involve only manipulations of the artificial constructs organising human affairs are ineligible for patent protection.  However, this was not initially considered to have any impact on inventions implemented in software, even when applied in the financial services sector, or more generally directed to improving the efficiency of organisational processes in any sector.  Certainly, for some time following the Grant decision, the Patent Office continued to grant patents in these categories, so long as the claims were clearly limited to computer-implementations, e.g. by expressly reciting elements of computer hardware, such as microprocessors.

Circa 2008 – A Change of Practice?

However, at some point this practice changed.  It is difficult to say exactly when, or why, because there was no public notice issued, and no updates to the published examination guidelines, and there has never been any acknowledgement or clear statement of a policy rationale underlying the change.  With the benefit of hindsight, and some analysis of examination data, there is evidence to suggest that the Patent Office began to take a harder line on computer-implemented subject matter at around the same time that the US Court of Appeals for the Federal Circuit (CAFC) handed down its en banc decision in In re Bilski, in October 2008.

The first public sign that the Patent Office was applying a different approach to computer-implemented inventions was its decision in Invention Pathways Pty Ltd [2010] APO 10.  The Invention Pathways decision was not, in itself, particularly controversial.  However, it proved to be the first in a line of decisions issued by IP Australia rejecting applications directed to inventions which it regarded – rightly or wrongly – as computer-implemented ‘business methods’.  In the course of these decisions, IP Australia came to apply a requirement that a patentable method must involve some physical effect which is significant both in that it is concrete but also that it is central to the purpose or operation of the claimed process or otherwise arises from the combination of steps of the method in a substantial way.  No such principle had ever previously been articulated by the Australian courts.

2014-2015 – Full Federal Court Weighs In

Unsurprisingly, this series of decisions eventually resulted in appeals to the Federal Court, and ultimately in two decisions of a Full Bench of the Court, the first relating to the construction and use of passive portfolios and indexes for securities trading (Research Affiliates LLC v Commissioner of Patents [2014] FCAFC 150), and the second to a computer-implemented method and system for use in assessing the competency or qualifications of individuals with respect to recognised standards (Commissioner of Patents v RPL Central Pty Ltd [2015] FCAFC 177).  These decisions did not directly endorse the Patent Office’s ‘central to the purpose’ test, instead focusing on whether the claimed invention, in substance, resides in a technological innovation (patentable), or a business innovation (not patentable). 

Beyond 2015 – Worse, Not Better!

While one might have expected that the decisions in Research Affiliates and RPL Central would have brought some much-needed clarity and certainty to the law relating to patent-eligibility of computer-implemented inventions, in recent years the trickle of Patent Office decisions resulting from disputed rejections has become a veritable flood.  As of the beginning of September this year, there had been 38 such decisions since the beginning of 2016 (i.e. the month following the RPL Central decision).  There have since been six more, bringing the total to 44. 

Throughout this period, the approach taken by the Patent Office has continued to evolve.  Increasingly, prior art has become a consideration in determining whether or not a computer-implemented invention constitutes patent-eligible subject matter.  It is right here that IP Australia’s approach breaks what I have called the ‘ball point pen principle’ and, as a result, is in direct conflict with authority of the Full Federal Court in CCOM.  It results in the absurd outcome that patent-eligibility of computer-implemented subject matter becomes time-dependent.  But surely particular subject matter cannot suddenly cease to be the kind of thing for which a patent might be granted simply because certain examples of that subject matter are no longer new?

IP Australia’s Current Practice

In a recent series of seminars on its approach to the examination of computer-implemented inventions (a recording of the webinar version is available on YouTube), IP Australia justified this apparent deviation from the orthodox understanding that ‘manner of manufacture’ is assessed on the ‘face of the specification’ using the following (in my view dubious) reasoning:

1. the High Court’s 1959 decision in NRDC, long considered to be the solid foundation on which the Australian law of patent-eligibility is built, may be less pertinent to computer-implemented inventions because the Court’s words ‘do not appear to be drafted with a mind to the advent and proliferation of computer technology’; and
2. the High Court’s recent decision (relating to claimed gene sequences) in D'Arcy v Myriad Genetics Inc [2015] HCA 35 implies that ‘it is necessary to compare the invention as claimed to the state of the art to determine what has been added to the sum of human knowledge by the work of the inventor.’

This second point is particularly suspect, considering that the corresponding ‘authority’ is taken, in part, from the minority judgment of Gageler and Nettle JJ, and that the relevant ‘human knowledge’ to which their Honours were referring is knowledge of the science of genetics, i.e. the fundamental basis for non-patentability was that Myriad’s claims were directed to something that had already existed in nature (even though it needed to be discovered).  This is quite different from comparing a claimed invention with prior art that is a product of human endeavour, i.e. that is itself an ‘artificially created state of affairs’. 

IP Australia also justifies consideration of prior art on the basis of the plurality’s statement in Myriad (at [12]) that ‘a definition of the allegedly patentable invention … depends upon the construction of the impugned claims read in the light of the specification as a whole and the relevant prior art.’  However, it is one thing to construe the claims through the eyes of the skilled person – including their knowledge of the art in question – and another thing altogether to restrict an evaluation of the ‘contribution’ made by the invention to only those claimed elements that are not to be found in the prior art.

The Rokt Decision

The court in Rokt rejected an approach to identifying the contribution made by the invention based on any ‘element-wise’ comparison with prior art (at [202]):

In my opinion, the [Commissioner’s] focus tended to be on issues of separate or indeed discrete differences or distinctions rather than on the claim or claims as a whole and as a matter of substance. This thread also ran through the respondent’s cross-examination of Professor Verspoor. For example, there were, as one would expect, undoubted similarities between what computers could do or what computers were used for in December 2012 and the claimed invention. This focus on similarities does not of itself gainsay that the claimed invention solved a technical problem or involved steps foreign to the normal use of computers as at December 2012. A focus on elements known as at December 2012 in isolation tends to lose sight of the combination of techniques or components in an innovative and previously unknown way.

Furthermore, the court rejected the Commissioner’s submission that the claimed invention addressed only a business problem, and not a technical problem, finding (at [207]) that:

…there was a business problem of attracting the attention of the user and having the user choose to interact with the advertiser, but this problem was translated into the technical problem of how to utilise computer technology to address the business problem. The invention aimed to solve this technical problem through the introduction of the engagement offer and identifying what steps the software needed to execute in order to modify dynamically the website that the user was browsing while they were browsing it to, first, implement in the web browser or device the concept of the engagement offer, second, to implement in the computer system the necessary software for selecting engagement offers and advertisements for the given user based on the previous interactions with the system and the interactions of other similar users and, third, to have that system interact with the widget in the web browser in real time.

The court concluded (at [213]) that:

Taken in isolation, a database, a client-server architecture, the running of the Javascript program on a publisher’s website and the creation of a ranking engine to rank abstract data to achieve an ordered list were each known as at December 2012 but, in combination, the distinction between engagement offers and general advertising, coupled with the algorithms making use of background data for personalisation and ranking was a new combination of new and previously existing components and a new use of computer technology.

Conclusion – Will Rokt Change Patent Office Practice?

Despite the court’s clear rejection in Rokt of the Commissioner’s approach based on ‘separate or indeed discrete differences or distinctions rather than on the claim or claims as a whole and as a matter of substance’, I do not expect that we will see any immediate change in Patent Office practice in relation to patent-eligibility of computer-implemented inventions, for at least three reasons.

Firstly, this is only one decision of a single judge of the Federal Court of Australia.  IP Australia clearly believes that its current practice is justified by the higher authority of the Full Federal Court in Research Affiliates and RPL Central, as well as the High Court in Myriad.  I anticipate that examiners and Hearing Officers will continue to cite these cases and distinguish inventions claimed in new applications from the facts in Rokt.

Secondly, the Office will take comfort from the fact that the court in Rokt found the claimed invention to be a combination of both new and previously-existing components and, as such, is arguably distinguishable from a case in which an invention comprises a new combination made up only of components known in the prior art.

Thirdly, I anticipate that the Commissioner will seek leave to appeal the decision to a Full Bench of the Federal Court.  In normal circumstances, I would not expect an appeal to be successful in a case such as this.  The court’s judgment is heavily fact-intensive, and the primary judge had the benefit of observing the cross-examination of an expert witness.  An appeals court would be reluctant to disturb the judge’s findings of fact, and the Commissioner would have to establish a clear error of law in the decision, such that the judge’s ruling that the claimed invention is a patentable ‘manner of manufacture’ was not open to him on the facts.

However, the current circumstances are not ‘normal’.  Judgment is pending in the appeal from Encompass Corporation Pty Ltd v InfoTrack Pty Ltd [2018] FCA 421 which has been heard by an expanded bench of five judges, and which concerns similar issues to those which arose in Rokt.  The Commissioner of Patents has intervened in the Encompass appeal, and the Institute of Patent and Trade Mark Attorneys of Australia (IPTA) has also sought to intervene, and filed written submissions.  The decision in Encompass is likely to clarify the role (if any) of prior art in assessing patent-eligibility of computer-implemented inventions, and could therefore change the understanding of the law relied upon by the judge in Rokt.  If the Commissioner does not seek to appeal, the opportunity to have the facts in Rokt reconsidered under the law to be explained in Encompass will be lost.

So, while the Rokt decision may be a welcome development for innovators seeking protection for computer-implemented inventions, any substantive change in Patent Office practice will most likely have to await a decision in the Encompass appeal.  My hope is that the court in that case will put an end to the idea that assessing whether or not a computer-implemented invention is a patentable ‘manner of manufacture’ may require a detailed assessment of the prior art, along with extensive expert evidence.  I do not believe that this is what the Full Court intended in Research Affiliates or RPL Central, nor do I think that IP Australia is correct in its view that this approach is endorsed by the High Court in Myriad.  Certainly, if the extent of evidence presented in Rokt is anything to go by, it is potentially a very bad thing for the state of patent law in Australia.

Tags: Australia, Business processes, Computer programs, Patent Office, Patentable subject matter