On 10 March 2022, the High Court of Australia granted Aristocrat’s application for special leave to appeal the decision of a Full Bench of the Federal Court, which found its claims directed to an Electronic Gaming Machine (EGM) implementing a new ‘feature game’ to be ineligible for patenting under Australia’s ‘manner of manufacture’ test of subject matter. The case – which is Case No. S40/2022 – has since been making rapid progress. Aristocrat filed its written submissions [PDF, 711kB] on 20 April 2022. Two parties have applied for leave to appear as amici curiae: the Institute of Patent and Trade Mark Attorneys of Australia (IPTA), which filed its written submissions [PDF, 483kB] on 5 May 2022; and Fédération Internationale des Conseils en Propriété Intellectuelle (FICPI), which filed its written submissions [PDF, 382kB] on 4 May 2022. The Commissioner of Patents filed her written submissions in response [PDF, 718kB] on 11 May 2022. Finally Aristocrat filed its reply [PDF, 310kB] on 20 May 2022. The scene is now set for a hearing, which is set down to take place on 9 and 10 June 2022 before a Full Court of seven judges, which is somewhat unusual (a panel of five judges is more common) and suggests that the High Court may have decided that it is finally time to settle the question of patent-eligibility of computer-implemented inventions in Australia.
The Aristocrat case raises interesting questions. The claims of the patents at issue recite various hardware elements, some of which are common to all computers and some of which are specific to EGMs, along with software-implemented processes comprising the ‘feature game’. (A ‘feature game’ is a secondary, or bonus, game triggered by the occurrence of a defined event in the ‘base’ game of spinning reels.) It is common ground that the hardware elements are well-known in the field of electronic casino gaming, and that the new and inventive contribution resides wholly in the feature game implementation.
At first instance, the primary judge (Justice Burley) considered that the appropriate approach to such a case involves a ‘two step’ analysis. The first step is to construe the patent specification, from the perspective of the person skilled in the relevant art, to identify the claimed invention. This is to be determined as a matter of substance, and not merely based on the particular form of the claims. If the invention thus identified comprises patent-ineligible subject matter, such as a mere scheme, idea, or business method, then the second step involves an enquiry into whether the claimed computerisation involves some further contribution sufficient to render the invention patent-eligible. Applying this approach to Aristocrat’s claims, Burley J found at the first step that Aristocrat’s claims were for ‘a machine of a particular construction which implements a gaming function’. As a result, the question of patent-eligibility was resolved immediately – an EGM is not a mere scheme or a business method, and is thus a ‘manner of manufacture’. There was, the judge concluded, no need to proceed with step two.
On appeal, a Full Bench of the Federal Court rejected the test proposed by the primary judge. The majority (Middleton and Perram JJ) proposed an alternative two step test which asks firstly whether the claimed invention is ‘a computer-implemented invention’ and then – if so – ‘can the invention claimed broadly be described as an advance in computer technology’. They determined that Aristocrat’s EGM was in substance a computer-implemented invention, and that the asserted contribution of the claimed invention ‘pertains only to the use of a computer’ and not to ‘the development or advance of computer technology’, such that the claims were not directed to patent-eligible subject matter. In a separate judgment (which Aristocrat characterises as a ‘dissent’), Nicholas J arrived at the same ultimate conclusion, but disagreed with the majority that ‘an advance in computer technology’ is necessary for a computer-implemented invention to be patent-eligible. In his view, such an invention may be patent-eligible if it results in innovation in ‘different fields of technology’ where technical problems that lie ‘outside the computer’ may be solved using ‘generic computing technology. He proposed that the proceeding therefore be remitted to the primary judge to consider whether the inventions may be patentable as an advance in the field of gaming technology.
With all of these different lines of reasoning present in the judgments of the lower courts, the stage is surely set for a showdown before the High Court. So what are the main arguments that each of the parties will be relying upon?
Aristocrat
Aristocrat identifies the following issues that (it says) are raised by the appeal:
- Do the general principles of manner of manufacture apply to an invention implemented utilising a computer (including a machine of particular construction, such as an EGM, that includes computer components)?
- Alternatively, is the patent eligibility of a “computer-implemented invention” to be determined by asking whether it involves an advance in computer, or any other, technology? Does this wrongly cause the enquiry into manner of manufacture to overlap with the separate inquiries into novelty and inventive/innovative step?
- Does the earlier decision of the High Court in D'Arcy v Myriad Genetics Inc [2015] HCA 35 (Myriad) support an identification of the invention for the purpose of determining whether it is a manner of manufacture, other than by reference to the language of the claims defining it, read in the context of the body of the specification?
In its submissions, Aristocrat asserts that the finding that its claims are patent-ineligible is the culmination of a series of errors into which the Full Federal Court has fallen through a number of its recent decisions. In particular, it says that recent Full Courts ‘have erred in (a) construing claims by focusing upon those integers thought to be “inventive” or “novel” to the exclusion of other integers; (b) creating fetters on the patentability of computer-implemented inventions; and (c) drawing statements from EU and US cases, without the necessary caution identified by [the High Court] given the different legislation and jurisprudence in those jurisdictions.’ It further says that the Full Court in the Aristocrat case ‘wrongly merged the two questions identified by Burley J’, that it ‘treated statements in recent [Full Court] decisions as though they are statutes and used them to create a bright line test’, and that ‘Nicholas J, in dissent, rejected the [Full Court] majority’s bright line test, but erred in rejecting Burley J’s analysis’.
Aristocrat argues that the principles to be applied to assessing manner of manufacture are universal, and that there is no foundation for computer-implemented inventions to be treated differently from other categories of subject matter. It says that, through a series of recent decisions, the Full Court has created a ‘collection of restrictive principles’ specific to computer-implemented inventions that are founded upon specific exclusions to patent eligibility codified in the UK 1977 Act and the European Parent Convention (EPC) and developed under §101 of the US statute, and that are inconsistent with High Court and earlier Federal Court authority. In doing so, Aristocrat says that the Full Court has:
- overlapped manner of manufacture with novelty and inventive step;
- created an anomalous qualification on the principles governing patentability;
- placed an unjustifiable gloss on the statutory language; and
- fettered by contrived constraints the patent eligibility of computer-implemented inventions
With regard to its own claims, in particular, Aristocrat says that the Full Court majority failed to distinguish between a mere scheme (which would not be patentable) and a patentable combination by failing to consider whether the integers of the claim have a working inter-relationship that characterises a patentable combination. It says that the majority:
- wrongly rejected as incorrect the primary judge asking if the invention was a mere scheme or business method, while acknowledging no authority on manner of manufacture has ever required the characterisation of an invention as “computer-implemented”;
- wrongly focused on whether the claimed EGM was a computer when such a characterisation is irrelevant as a matter of law in determining manner of manufacture; and
- in advancing the ‘new test’ (of whether the invention claimed can broadly be described as an advance in computer technology), the majority neither followed the common law methodology identified by the High Court authorities, nor identified any other jurisprudential foundation.
Aristocrat also says that all of the judges of the Full Court erred in by isolating from its claim only those integers relating to the ‘rules’ of the feature game which they considered to be the novel or inventive aspect of the claim, and using only these as central to their enquiry as to whether the invention was a manner of manufacture, even though they recognised the need to read the claim as a whole. Manner of manufacture, Aristocrat says, should be determined independently of questions of inventive step and novelty.
Characterising the opinion of Nicholas J as a ‘dissent’ (despite his reaching the same conclusion as the majority on the issue of whether Aristocrat’s presented claim is for a manner of manufacture), Aristocrat argues that he wrongly equated the question of whether there is an artificially created state of affairs present with an enquiry as to whether the claimed invention resulted in a technical innovation, technical contribution or technical effect. It also says that Nicholas J did not identify what constitutes a “mere scheme”, nor why the claimed invention was not to an EGM or a patentable combination.
According to Aristocrat, the majority’s new test is not based on analogical reasoning from established legal principle, but instead seeks to give effect to a new policy consideration, identified as “the encouragement of novel uses of computers to advance computer technology”. It says that the majority defined a computer in terms that encompass, practically, any device containing a processor and memory which, under the new test, will be excluded from patentability if it only results in an advance in areas other than computer technology. Aristocrat asserts that this will have significant ramifications for industry, affecting manufacturers of a wide range of goods, as diverse as digital clocks, refrigerators, writing instruments, and EGMs, indeed the whole Internet of Things.
In its conclusion, Aristocrat submits that, in order to find its claimed invention not to be for a manner of manufacture, the Full Court had to:
…first, erect an exclusion for inventions involving a computer from the general principles of patentability and, second, devise new fetters contrary to these general principles of patentability. The [Full Court] majority imposed a requirement of “an advance in computer technology”. Nicholas J imposed a requirement of “a technical contribution” or “technological innovation” in the field of technology to which the invention belongs. In doing so, the [Full Court] fell into error by characterising the claimed invention by reference to perceived novelty conferring or inventive integers as opposed to by reference to the invention “so far as claimed”. (Emphasis in original.)
The Commissioner of Patents
The Commissioner’s position, unsurprisingly, is that the Full Court got it right. She observes that the question of whether an allegedly patentable invention is a manner of manufacture depends upon the construction of the patentee’s claim in the light of the specification as a whole and the relevant prior art, which in this case is the common general knowledge. The proper construction of the claim is to be determined as a matter of substance, not merely by the form of the claim. She says that, so characterised, Aristocrat’s claimed invention is an abstract idea, being a scheme or set of rules for playing a game, which is implemented using conventional computer technology for its well-known and well-understood functions. She argues that recent Full Court decisions, including the Aristocrat Full Court, do not involve any ‘new test’ or ‘rigid formula’, but correctly recognise that an unpatentable abstract idea does not change its legal character merely because it is implemented by a computer.
The Commissioner accuses Aristocrat of changing its case significantly in that, in the courts below, it did not challenge the correctness of the other recent Full Court decisions. She says that, in now arguing that other recent Full Court decisions involved numerous errors, and that the decision of the Full Court in this case represents the “culmination of a combination of these errors” and “is not an outlier”, Aristocrat’s appeal now depends upon it establishing that the other recent decisions were wrong. In its reply, Aristocrat disputes this, and it will be interesting to see how this plays out before the High Court.
There is, as the Commissioner rightly points out, a difference between an abstract idea, which is not patentable, and a practical application of it to produce a useful result, which may be. In the context of games, game rules are not patentable, but physical apparatus used for playing a game (such as cards or a board) may be. The Commissioner’s argument, however, is that where the physical apparatus is a computer, this principle does not apply. She says that the point made by the recent Full Court authorities on computer-implemented inventions is that the implementation of an abstract idea in a computer, using conventional computer technology for its well-known and well-understood functions, does not make it a manner of manufacture and that such an invention, considered as a matter of substance, remains an unpatentable abstract idea. As regards the requirement for an improvement in computer technology, the Commissioner argues, there must be invention in the way in which the abstract idea is implemented in the computer, as distinct from the idea itself.
With respect to the decision of the primary judge, the Commissioner argues that the approach taken by Burley J is problematic because it allows the patenting of a new set of game rules by the device of framing the claim as a claim to a gaming machine with standard hardware and software components configured to implement those rules. She says that Aristocrat did not “invent” any gaming machine in this case, but merely devised a scheme or set of rules for playing a game, which is not a manner of manufacture.
With respect to the reasoning of the Full Court majority, the Commissioner denies that focussing on the question of whether the claimed invention could broadly be described as an advance in computer technology involves any error or the imposition of a ‘new test’. In her submission, the majority was employing the term ‘computer-implemented invention’ in a more limited sense, and had in mind an invention involving the implementation of an otherwise unpatentable abstract idea, such as a scheme or plan, in a computer. She contends that Inventions in other fields of technology are not excluded merely because they involve the use of computers, and denies propounding such an approach.
his focus on the question whether the invention involved the use of computers for their well-known and well-understood functions did not involve any fundamental difference in principle. She says that the relevant distinction is between an abstract idea implemented using conventional computer technology for its well-known and well-understood functions, on the one hand, and an invention involving some advance or improvement in computer technology, on the other, and that these are ‘two sides of the same coin’.
In the Commissioner’s submission:
…the gaming machine in [Aristocrat’s claim] is a conventional gaming machine used for its well-known and well-understood functions, to carry out an electronic game involving the selection and display of symbols on reels and the awarding of prizes. The substance of the invention, and what distinguishes this gaming machine from others, resides in the scheme or rules of the game itself. In terms of the recent Full Court decisions, the invention, as a matter of substance, involves an instruction to carry out an electronic game, using conventional computer technology for its well-known and well-understood functions; and there is no ingenuity in the implementation of the game, as distinct from the scheme or rules of the game itself. A scheme or set of rules for playing a game is an abstract idea of the kind that has never been considered patentable.
The Interveners – IPTA
The main focus of IPTA’s submissions is on the implications of the Full Court’s recent manner of manufacture jurisprudence on a broad range of fields of endeavour, and not only upon the gaming industry that is addressed by Aristocrat. As stated by IPTA:
Aristocrat has addressed the effect of this test [requiring an advance in computer technology] in its own industry. However, because of the number of fields of endeavour in which computers are now used, a test that excludes from patentable subject matter any invention implemented in a computer that is not an advance in computer technology represents a seismic shift in patent law. It renders unpatentable swathes of inventions that would previously have been patentable subject matter, and that have been found as such.
For example, the new test excludes from patentability medical imaging and diagnostic machines, speed detection camera systems, green energy solutions, and biotechnology inventions such as COVID-19 diagnosis systems, where those inventions include features (as is often the case) implemented by way of software running on known computer processors.
This is – as I have already noted – a point of contention with the Commissioner, who has submitted that the Full Court’s test should be understood as limited to computer implementation of abstract ideas. In IPTA’s submission, other parts of the majority’s reasons indicate that they did mean to exclude from patentability inventions in other fields that include computer implementation. IPTA further argues that if the same approach were applied in the field of pharmaceuticals, it would eliminate patents on inventions that are new uses of a known substance.
Australia-United States Free Trade Agreement (AUSFTA).
IPTA also submits that recent jurisprudence on patent-eligibility of computer-implemented inventions is having a deleterious effect on the prosecution and grant of patents in Australia. In particular, it contends that examiners commonly now disregard computer hardware integers, unless the examiner considers those integers themselves to be novel or inventive, with the result that subject matter that would likely have
been found to be a manner of manufacture by the Commissioner in the past, so that patents were granted, is no longer being considered to be a manner of manufacture, and patents are being refused.
In its conclusion, addressing what it submits to be the correct test, IPTA states that:
Aristocrat … is correct to distinguish between a mere scheme, and a scheme that “involves something more” that renders it a manner of manufacture. [Aristocrat] goes on to say that the Full Court in [Commissioner of Patents v RPL Central Pty Ltd] should have first considered whether the invention was a mere scheme and then, if so, asked, “whether, either in its implementation or otherwise, there was something that rendered the claimed invention patentable.” There need not be two steps. In determining whether an invention is a mere scheme, the Court should consider the claim as a whole, without distinguishing between the idea of the invention and its implementation. If it is the way a claim is implemented that makes it more than a “mere scheme”, then it will be patentable if it meets the ordinary requirements for patentability….
The Interveners – FICPI
The primary focus of FICPI’s submissions is the current state of US law in relation to patent-eligible subject matter. Aristocrat has described the US position as ‘in chaos’, and FICPI’s submissions support that sentiment. FICPI discusses the outcome of the US Supreme Court’s decision in Mayo Collaborative Services v Prometheus Laboratories Inc. 566 US 66 (2012) (Mayo) and Alice Corporation Pty Ltd v CLS Bank International 573 US 208 (2014) (Alice), leading to the so-called Mayo/Alice test for patent-eligibility. It also outlines criticisms of this test by Federal Circuit judges, experts, and academics.
In particular, FICPI contends that in efforts by the courts and USPTO examiners to apply the Mayo/Alice test, US practitioners have found decisions to appear to turn largely on whether the decision-maker considers an alleged invention to have inventive merit, or not. This, it argues, has led to a large body of case law from which it has not been possible to discern a clear objective standard.
In its conclusion, FICPI states that it:
…is unaware of any jurisdiction other than Australia which provides a two step test devised for determining the patentability computer-implemented technologies (which includes a requirement to show “an advance in computer technology” ), being a distinctly different test to that which is applied for inventions in other fields of endeavour. In FICPI’s view, the Aristocrat test puts Australia out of step with every other jurisdiction around the world.
Conclusion – A Hard Case to Call
Despite the best efforts on each side to make out that their respective positions are clearly and incontrovertibly correct, there is merit in the arguments presented by both Aristocrat and the Commissioner.
On the one hand, it would appear anomalous if an unpatentable idea, such as a business scheme, could be made patentable merely by providing generic instructions to ‘do it on a computer’. In such a case, the Commissioner’s position that an unpatentable abstract idea does not change its legal character merely because it is implemented by a computer is plainly sound.
On the other hand, there are countless modern products that have innovative functionality that is implemented entirely in software. In some of these the hardware may be specialised, in that it is not configured for general-purpose operation (such as a desktop computer), but is nonetheless – like Aristocrat’s EGMs – conventional within its particular field of endeavour. It would appear no less anomalous for inventions of this kind to be unpatentable simply because they do not represent an advance in computer technology, particularly when an alternative implementation (e.g. in mechanical form) would be uncontroversially patent-eligible. In these cases, Aristocrat’s position that the invention can and should be considered as a patentable combination, rather than reducing it to perceived novelty-conferring or inventive integers, is arguably preferable.
Assuming that there is some continuum from clearly patent-ineligible computer-implemented schemes to clearly patent-eligible inventions in established fields of patentable technology that happen to be implemented primarily in software, the question arises as to where Aristocrat’s EGMs sit on that continuum. Is it enough, as Burley J concluded, that an EGM is ‘a machine of a particular construction’, even when the only novel contribution of the inventors is the implementation in software, on that machine, of a new system of game rules? Or is something ‘more’ required, such as the ‘advance in computer technology’ sought by the Full Court majority?
The difficulty here is that the Full Court has thus far failed to provide a clear, principled approach that can be applied in a consistent and coherent manner to enable judges, patent examiners and hearing officers, and patent practitioners to determine and agree on which side of the patent-eligibility line any given computer-implemented invention falls. In attempting to do so, it has (as Aristocrat contends) looked to approaches taken in other jurisdictions, despite significant differences in the law, apparently in the hope of achieving analogous outcomes in Australia. This has not resulted in clarity. It has resulted in confusion and uncertainty.
We can hope that the High Court will be more successful. But the experience of the Federal Court demonstrates that it faces a challenging task. I can see the case going either way. I can also see it leaving us not much better off with regard to the patent-eligibility of a broad range of computer-implemented inventions.
0 comments:
Post a Comment