Manner of manufacture and software-based inventions: Patent eligibility test re-affirmed but limited further clarity provided by Australian Appeals Court

After much anticipation and a 10 month wait since the appeal was heard in November 2018, the expanded panel of five Federal Court judges in Encompass Pty Ltd v Infotrack Pty Ltd have delivered their decision, finding that they were “not persuaded that the primary judge erred in concluding that each [Encompass patent in suit] does not claim an invention that is a manner of manufacture within the meaning of section 6 of the Statute of Monopolies”.  Each of the Encompass patents in suit have accordingly been found to be invalid.

As reported in our blog article published in late 2018, central to the dispute between the parties were Encompass’ patents relating to methods for an electronic processing device to access information from different databases (such as Land Titles Office, ASIC and a motor vehicle registry) in a seamless manner, and to produce reports obtained from such searches to subscribers of the service for a fee.

At first instance Justice Perram found that the invention as claimed in each of the relevant patents failed to meet the requirements of being a ‘manner of manufacture’ on the basis that it did not involve any improvement in the functionality of the computer nor the computer itself.  His Honour acknowledged that Encompass’ patents described a method that resulted in the computer being used to do something it had not been used to do before – by using a network representation and querying remote data sources – and that requiring the users to purchase the search results provided an innovative step.  However, while it was further accepted that this enhanced the experience for the user of the computer, His Honour considered that these matters did not improve the functionality of the computer because, after the consideration of the prior art, the methods, individually, were not new and were well-known.

On appeal, Encompass sought to argue that Perram J erred in directing his attention to whether the alleged invention results in “an improvement in the computer” and erred in referring to methods known in the prior art.  Encompass argued that this resulted in His Honour erroneously incorporating the “machine – or – transformation” test into his consideration and resulted in concepts of inventive step being conflated with considerations as to whether the inventions are a manner of manufacture.

In rejecting these submissions, the Full Court found that despite Encompass’ “oblique attacks” on the correctness of Research Affiliates LLC v Commissioner of Patents [2014] FCAFC 150; (2014) 227 FCR 378 (Research Affiliates) and Commissioner of Patents v RPL Central Pty Ltd [2015] FCAFC 177; (2015) 238 FCR 27 (RPL), they did not think that the correctness of those decisions was seriously in doubt.  The Full Court therefore considered that the burden of the appeal lay in whether Perram J applied the principles espoused by these earlier authorities to the case at hand.  Notably, and in contrast to what many had anticipated, the Full Court observed that the appeal “did not provide the occasion for this Court to set out the metes and bounds of patentable computer-implemented inventions, assuming that be an appropriate task…”.

While these findings by the Full Court can be regarded as somewhat anti-climactic, particularly for those hoping for greater clarity on whether software-based inventions or computer-implemented methods are patentable according to the ‘manner of manufacture’ test, the Full Court did provide a summary of certain propositions that it considered cannot be in dispute.  In particular, in the Full Court’s view, the High Court in NRDC [1959] HCA 67 made plain that, in the context of considering whether a claimed invention is a manner of manufacture, the inquiry is not into the meaning of the words but in relation to characterization of the invention involved.  This question of characterization is to be addressed as a matter of substance, not merely as a matter of claim form.

Accordingly, in considering the decisions and language used in Research Affiliates and RPL, the Full Court noted that while the Judges in those earlier decisions (in grappling with whether the mere implementation by a computer of an abstract idea or scheme is enough to transform unpatentable subject matter to patentable subject matter) used the language of ”artificial” or “physical” effects and “technical contributions”, they were doing no more that explaining that the claimed method in those cases “did not transcend, as a matter of substance, what remained an abstract idea or mere information of a kind that has never been considered to be patentable subject matter under Australian law”.  That is, use of the word “physical” and the like must be understood in a very broad or extended sense, and in no way different to the High Court’s reference to and use of “physical phenomenon” in NRDC.

On this basis, the Full Court considered that Perram J’s inquiry as to whether the implementation of the claimed method results in “an improvement in the computer” must be seen in the context of the fulsome discussion in the earlier Research Affiliates and RPL decisions.  In particular, the Full Court did not understand His Honour to be “positing a hardware specific test… The judge was simply directing his mind to whether the claimed invention was something more than ‘mere generic computer implementation’ of an otherwise abstract idea”.  On the facts of this case, it was found by the Full Court that Encompass’ claimed method in truth involved “no more than an instruction to apply an abstract idea (the steps of the method) using generic computer technology”.

While the Full Court’s decision in this matter provides some clarity in relation to how the Court will approach the manner of manufacture test for computer implemented inventions, it remains to be seen whether those interested in obtaining patent rights in this field consider that the developments they most wish to protect will be able to be patented.  Where those developments pertain to methods which use ordinary or ‘generic’ computer technology, it is unlikely that they will be found to meet the test.

Perhaps, the Full Court’s observation that the matter was not one that was appropriate to set out the ‘metes and bounds’ of the manner of manufacture test is an acknowledgement that there will likely be continued calls for reform in this commercially important area.