A rare five member bench of the Full Court of the Federal Court delivered its much anticipated decision in Encompass Corporation Pty Ltd v InfoTrack Pty Ltd  FCAFC 161.
Although the Full Court found that the two computer-implemented inventions lacked patentable subject matter, the failed appeal does not spell the demise of computer-implemented inventions generally. Rather, the Court re-affirmed that implementation of a ‘mere scheme’ or ‘abstract idea’ by ‘generic software’ will not transform an alleged invention into patentable subject matter.
It was, however, a lost opportunity for the Court to take the matter further and set out definitively what is, and is not, patentable subject matter in computer-implemented inventions. The Court’s decision leaves companies in an uncertain position when assessing their ability to obtain patents to protect key software developments.
Background of the case
Encompass and its exclusive licensee, SAI Global Property, alleged that InfoTrack had infringed certain claims of two innovation patents. InfoTrack admitted infringement, but cross-claimed to revoke the patents including on the basis that they were not directed to patentable subject matter (i.e., lack of manner of manufacture).
Encompass’ patents claimed the generation (by ‘a method using an electronic processing device’ and ‘an apparatus in an electronic processing device’) of representations of networks of relationships between entities (such as people and corporations) where a user could select searches of databases to be performed in respect of a given entity, the results of which would then be added to the network. For example, this would allow a user to see a visualisation of the relationships between different companies, and select different searches (e.g., a title search) to add extra detail regarding a given company’s holdings or directors.
Importantly, the patents did not identify any particular software or programming to perform the invention. This proved critical to the Court’s reasoning that the claimed inventions were not directed to patentable subject matter.
Manner of manufacture and computer-implemented inventions
Sections 18(1)(a) and 18(1A)(a) of the Patents Act 1990 (Cth) require that an invention needs to be a manner of manufacture within the meaning of section 6 of the Statute of Monopolies to be a patentable invention. In answering this question, the Court has been guided by two questions set out in National Research Development Corporation v Commissioner of Patents (1959) 102 CLR 252 (NRDC):
- does the invention consist of an artificial state of affairs; and
- is the significance of the invention economic?
This test has been applied with difficulty to computer-implemented inventions. Arguably, any invention involving a computer is likely to consist of an artificial state of affairs, including inventions relating to traditionally unpatentable subject matter, such as business methods or schemes. The Court has therefore focused on whether the invention is the implementation of an idea in a computer (unpatentable), or whether it is the implementation of an idea which results in an improvement in a computer (patentable).
In the first instance decision of Perram J, Encompass Corporation Pty Ltd v InfoTrack Pty Ltd (2018) 130 IPR 387;  FCA 421, the Court acknowledged the difficulties in determining what an improvement in a computer actually is. On one view, any software installed in a computer is an improvement, as it allows the computer to perform tasks that it otherwise could not.
In resolving this difficulty, his Honour determined that the claimed inventions of the patents in suit were not an improvement in the function of the computer, as the enhanced user experience was achieved through the interconnection of three previously known methods – that is, the computer was not doing something which could not already be done. These findings underpinned his Honour’s decision that the patents were invalid on the basis that they did not concern a manner of manufacture.
The appeal decision was greatly anticipated – an expanded Full Bench was appointed to hear the appeal and two parties intervened (Commissioner of Patents and the Institute of Patent and Trade Mark Attorneys of Australia (IPTA)).
In an unanimous decision, the Full Court upheld Perram J’s findings that the inventions were invalid on the basis that they lacked patentable subject matter. Although this outcome was not a surprise, it came as more of a surprise that the expanded Full Bench determined that the case was not the appropriate vehicle to provide clear boundaries for computer-implemented inventions, stating ‘[w]e do not see this appeal as raising any significant question of principle’ and that ‘[t]his appeal does not provide the occasion for this Court to set out the metes and bounds of patentable computer-implemented inventions’.
In discussing the principles concerning whether an invention is a manner of manufacture, the Court reiterated the established principle that a ‘mere scheme’, ‘abstract idea’ or ‘abstraction’ will not be patentable. Implementation of unpatentable methods via ‘the instrumentality of a computer’ will not transform an otherwise unpatentable abstraction into patentable subject matter.
The Court explained that previous decisions had been directed to drawing distinctions between patentable manners of manufacture and unpatentable abstractions, and that specific terminology used in those decisions (such as a suggestion that the invention must produce a ‘physical effect’ in Grant; or that it must result in ‘technical contributions’ or ‘improvements in a computer’ in Research Affiliates) were merely indicative of the Courts attempting to demarcate this conceptual distinction.
Turning to the inventions in suit, the Court concluded that they were nothing more than instructions to apply an abstract idea, in which a computer was an ‘intermediary’ in carrying out the steps of a method described in terms which amounted to an abstract idea or scheme. Notably, the claims of the patents did not incorporate any particular software or programming, nor did the specifications identify any such programming or software. The user must develop their own software to use the method, as the invention as claimed was little more than an idea for a computer program.
The Court also rejected the appellants’ submission that Perram J’s consideration of whether the invention resulted in ‘an improvement in a computer’, imported into the test for patentability a requirement for the creation of a new physical device or improvement in the hardware. The appellants also argued that the primary judge had impermissibly imported considerations of novelty and/or innovative step into the patentability analysis. The Court accepted that the primary judge’s language ‘suggested that other, conceptually distinct elements of patentability’ might have intruded into his Honour’s consideration of manner of manufacture, but ultimately found no error in the primary judge’s conclusions on patentability.
Submissions from the Commissioner and IPTA were considered but ultimately given little weight by the Court. IPTA challenged the correctness of some of the Commissioner’s decisions in this field, also providing ‘editorial comments’ on the drafting of the Australian Patent Office Manual of Practice and Procedure (the Examiner’s Manual). The Court declined to comment on these issues, noting they were outside the scope of the appeal.
Tracking the implications
The Full Court’s decision offers little new guidance to parties considering whether a computer-implemented invention is patentable. Rather, the Court has confirmed that where a computer-implemented method is no more than a mere scheme, abstract idea or intellectual information that requires only “generic computer implementation”, it cannot be a manner of manufacture.
The decision clarifies (albeit tentatively) that decision-makers should not have regard to prior art when considering whether an invention results in an ‘improvement in a computer’. Their Honours re-emphasised the caution previously expressed by the Courts against importing concepts of novelty and inventiveness into the assessment of manner of manufacture.
Despite this, the Examiner’s Manual (under ‘principles for examination on patentable subject matter’), provides the following direction:
In order to determine the substance, examiners must identify the central underlying invention embodied in the claim, rather than merely consider the literal form of the claim. This is the contribution the claimed invention makes to the art and is determined based on a reading of the specification as a whole and examiners’ understanding of the common general knowledge and prior art.
The consideration of prior art in assessing patentability in examination has resulted in numerous applications for computer-implemented inventions being rejected as concerning unpatentable subject matter. This trend in examination practice (which motivated IPTA’s intervention), is unlikely to change following the Full Court’s decision. Notably, IP Australia has subsequently announced that it would not be altering current examination practice in relation to computer-implemented inventions.
For potential applicants for computer/software implemented inventions (and those seeking freedom to operate), the upcoming appeal from the decision in Rokt Pte Ltd v Commissioner of Patents  FCA 1988 may yet offer much-awaited clarity on patentability.