Computer-Implemented Inventions and the ‘Ball Point Pen Principle’ – Why the Australian Law on Patent-Eligibility is a Mess

The law – and Patent Office practice – relating to the assessment and examination of patents and applications for computer-implemented inventions in Australia is currently a complete mess.  I challenge anyone – whether an inventor, applicant, patent attorney, patent examiner, Patent Office hearing officer, IP lawyer, barrister or judge – to provide a coherent explanation of how to go about deciding whether an invention involving the use of a computer is eligible for patent protection.  I myself cannot do it, so I would genuinely love to have someone explain it to me.

In this article, I will show you where a unanimous Full Court panel of three judges of the Federal Court of Australia – the highest legal authority to have considered patent-eligibility of computer-implemented inventions in this country – clearly stated that it is impermissible to consider the state of the prior art in assessing whether or not the subject matter of an invention is patent-eligible, i.e. a ‘manner of manufacture’ in the terminology used in the Australian law.  Notably, subsequent Full Court panels have thrice accepted the correctness of this decision without criticism.  Then I will show you where a single judge of the Federal Court recently found an invention to comprise ineligible subject matter on the basis that various components of the claims are known from the prior art. 

I will also show you where a Patent Office hearing officer, in upholding the rejection of a patent application by an examiner, expressly stated that it is legitimate to consider the prior art when assessing patent-eligibility of computer-related subject matter.  I will point you to a recent opposition decision in which the hearing officer found that the claimed invention was not patent-eligible, despite an examiner having accepted the application, and the applicant being successful in establishing that the claims involved an inventive step over prior art raised by the opponent.  And I will identify a further four Patent Office decisions – in addition to 13 published since the beginning of 2017 that I listed last November – in which applications for computer-implemented inventions have been refused, including one relating to an application by Google which cites the aforementioned decision to justify considering prior art when deciding subject-matter eligibility.  In these decisions, the Patent Office is continuing to apply its adaptation of the England and Wales Court of Appeals’ four-step test for patentable subject matter, as set out in Aerotel Ltd v Telco Holdings Ltd; Macrossan’s Application, [2006] EWCA Civ 1371 (Aerotel/Macrossan), despite a lack of any clear authority for this approach in Australia.

Meanwhile, one of two Federal Court appeals of Patent Office decisions has been terminated following unsuccessful mediation, while the other has been delayed after the parties decided that expert evidence is required.  Yes – expert evidence to address whether or not an invention is a ‘manner of manufacture’, which is supposed to be a question of law, on which technical experts have no authority.

This is all, frankly, terrible for the state of patent law, and for high-tech innovation, in Australia.  Literally nobody really knows what the law is.  It is not just that patent applicants, and their attorneys, do not agree with the Patent Office about what is, and is not, patentable.  The Patent Office does not agree with itself!  Inventions that are receiving approval in examination are being rejected in opposition.  Attorneys, examiners, hearing officers and judges are finding it all but impossible to extract coherent principles from the relevant Full Court decisions, without exposing inconsistencies in those decisions.  In Federal Court, technical experts are being called in to shed light on something that is supposed to be a legal issue.  As I recently reported, the examination section that deals with inventions related to computing has the lowest acceptance rates in the Australian Patent Office.  All of this speaks to the uncertainty that currently surrounds what is, and is not, patentable when it comes to computer-implemented invention.

What’s Prior Art Got to Do With It?

It has long been recognised that the requirement in sections 18(1)(a) and 18(1A)(a) of the Patents Act 1990 that an invention must be ‘a manner of manufacture within the meaning of section 6 of the Statute of Monopolies’ relates purely to eligibility of the subject matter of the invention as claimed, and is distinct from other requirements, such as novelty and inventive (or innovative) step.  In Lockwood Security v Doric Products [2004] HCA 58, the High Court made a very clear statement on exactly this point (at [73]):

…it is important whether it is s 18(1)(a) [i.e. manner of manufacture], or s 18(1)(b)(ii) [i.e. inventive step], or s 40(3) [i.e. fair basis in the description for the claimed invention] that applies … because none of these complaints is identical with or overlaps with the others: they are conceptually distinct…. (Emphasis added.)

Earlier, in NV Philips Gloeilampenfabrieken v Mirabella International Pty Ltd [1995] HCA 15, made the point that whether or not a claimed invention is a ‘manner of manufacture’ is question to be determined from the specification alone, without recourse to the extrinsic evidence appropriate to enquiries into novelty and obviousness (at [9]):

That does not mean that the threshold requirement of ‘an alleged invention’ corresponds with or renders otiose the more specific requirements of novelty and inventive step (when compared with the prior art base) contained in s 18(1)(b). It simply means that, if it is apparent on the face of the specification that the quality of inventiveness necessary for there to be a proper subject of letters patent under the Statute of Monopolies is absent, one need go no further.  (Emphasis added.)

In the specific context of a computer-implemented invention relating to a method for storage and retrieval of Chinese characters on a computer (CCOM Pty Ltd  v Jiejing Pty Ltd [1994] FCA 1168), a Full Bench of the Federal Court of Australia (in a unanimous judgment) also observed the conceptual distinction between different requirements under the Patents Act (at [112]):

The immediate significance of this is that in the 1990 Act, manner of manufacture, novelty, inventiveness and utility are stated as distinct requirements of a patentable invention. Thus, whilst a claim for the ball point pen now would fail for anticipation and obviousness, it would still be a claim for a manner of manufacture.

And, further (at [126]):

It may be that such a claim lacks novelty, is obvious, or lacks utility, or there is a failure to comply with one or other of the limbs of s. 40 because, for example, the invention is not fully described or the claim is not clear and succinct. But if such hurdles all are surmounted, then in our opinion in a case such as the present there does not remain an independent ground of objection as to patentability, within the sense of para 18(1)(a) of the 1990 Act.

The analogy of the ball point pen is apt: particular subject matter does not 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 and inventive.  That would, of course, be absurd – as is plainly apparent when the subject matter at issue is something so self-evidently patent-eligible as a mechanical device.  So why, then, does all this sensible, coherent, logical reasoning appear to go out the window as soon as a computer is involved?

Computer-Based Inventions – Not a High-Tech Ball Point Pen!

A recent decision of a single judge (Justice Perram) in the Federal Court of Australia is a good example of a failure to adhere to the ‘ball point pen principle’: Encompass Corporation Pty Ltd v InfoTrack Pty Ltd [2018] FCA 421.  This case concerns two innovation patents owned by Encompass, claiming inventions relating to conducting searches across multiple disparate databases (such as motor vehicle, land titles, and company registries) when gathering business intelligence.  The system covered by the patents does this by generating a graphical ‘network’ representation of related entities (e.g. companies, individual directors/executives, addresses and so forth), enabling the user to select one or more entities from the network, automatically identifying the relevant databases to search for information about the selected entity, and then conducting the required searches, including making any payments that are required to access information held in the databases.

InfoTrack challenged the Encompass patents on various grounds, including lack of novelty, lack of innovative step, deficiencies of the disclosure in the patent specification, and lack of ‘manner of manufacture’, i.e. that the subject matter of the claims is not patent-eligible.  All of these attacks failed, except for the subject matter ground.

However, the challenges based on prior art were significant in the court’s decision on ‘manner of manufacture’.  In particular, it was established that certain prior publications disclosed all elements of the claimed invention except for the ability to search multiple remote data sources, and the ‘payment’ element.  Had the case involved standard patents, where an ‘inventive step’ is required, it seems likely that configuring a computer-implemented system to access multiple databases, and to automatically complete payment when accessing databases for which such payment is required, would be considered obvious modifications, and that the patents would therefore be invalid on this basis.

Here, however, the patents were innovation patents, which require only an ‘innovative step’, i.e. that some element that distinguishes the invention from the prior art makes a ‘substantial contribution to the working of the invention’.  On the basis of expert evidence, the court found that searching multiple remote data sources does not make a ‘substantial contribution’ (at [132]), but that ‘[a]lthough I would not describe the payment step as being weighty I do accept that it makes a substantial contribution to the working of the invention as claimed’ (at [134]).

With regard to ‘manner of manufacture’, the court reasoned that, according to authorities such as Research Affiliates LLC v Commissioner of Patents [2014] FCAFC 150 and Commissioner of Patents v RPL Central Pty Ltd [2015] FCAFC 177, a computer-implemented invention, in order to be eligible for patent protection, must result in ‘an improvement in the computer’, or not, on the contrary, merely require ‘generic computer implementation’ (at [193]).  Applying these principles, the court found that the inventions claimed in the Encompass patents were not eligible subject matter (at [195]-[196]):

In this case, the method disclosed in the Patents (and the apparatus) result in the computer being used to do something it has not been used to do before. But it is not clear to me that in doing so they have improved the functionality of the machine. This is because the method (and apparatus) merely involve a concatenation of three other methods, none of which is new: the use of a network representation, the querying of remote data sources and the use of a purchasing step.

It is true that I have accepted that the addition of the purchasing step has the consequence of providing an innovative step. But it does not follow that the method and apparatus involve any improvement in the computer. None of the three elements of the claims is, in itself, an improvement in the functionality of a computer. I would accept the combination of the three elements in the innovative way disclosed in the Patents certainly provides an enhanced experience for the user of the computer. But it is not self-evident that there is a necessary connexion between an enhanced user experience and an improvement in the computer. In this case, so it seems to me, the enhanced user experience results from the combination of well-known computing mechanisms. But unless that combination can itself be said to have brought about an improvement in the computer this will be beside the point.

This, it seems to me, is analogous to saying that a ball point pen with a pocket clip and a comfortable hand-grip is not a ‘manner of manufacture’ because ball point pens, pocket clips, and hand-grips are all individually known, and that combining them together does not result in patent-eligible subject matter because it does not result in any improvement in the basic functionality of the ball point pen, despite providing an enhanced experience for the user of the pen.  (Of course, such a pen would not be validly patentable today on grounds of lack of novelty, but that is not the point here.)  To put it another way, the court could not have reached this conclusion without the knowledge obtained via the prior-art-based attacks on novelty and innovative step that informed its assessment of what constituted ‘known’ methods of using a computer.  And it is hard to imagine that it would have reached the same conclusion had the claimed invention involved a system with physical components that together provide an enhanced user experience, rather than one in which the combined components are software-implemented.

The problem here appears to be the lack of clear understanding of what it means to make ‘an improvement in the computer’.  According to the conventional framework for assessing patent-eligibility, there should not be any consideration of the prior art, which is properly within the domain of the distinct requirements for novelty and an inventive step.  However, when it comes to computer-implemented inventions, it is increasingly difficult to avoid considering the prior art in order to determine what ‘improvement’ the invention can be said to have made.

Has the Patent Office Stopped Pretending?

Considering the above discussion, it should perhaps be unsurprising to find that some decision-makers have altogether abandoned the charade of pretending not to be influenced by the prior art in assessing patent-eligibility of computer-implemented inventions.  McCormick & Company, Incorporated [2017] APO 62 is a decision of a Patent Office hearing officer upholding an examiner’s rejection of a patent application relating to a computer-implemented method of generating a visual representation of different elements making up the flavour profile of food.

After reviewing considerations arising from the various judicial authorities (which were themselves derived from an earlier Patent Office decision of the same hearing officer in Aristocrat Technologies Australia Pty Limited [2016] APO 49), the hearing officer concluded that ‘[i]t is difficult, if not impossible, to see how these considerations can be addressed without an understanding of the art, including relevant prior art’ (at [49]), and that ‘I am satisfied that it is legitimate to consider the prior art in order to assess the actual contribution to the art, and thus to determine the substance of the invention’ (at [54], emphasis in original).  It is important to note, however, that this notion of the ‘actual’ contribution to the art is derived from the English law as expressed in Aerotel/Macrossan, and the importation of this approach into Australia is substantially an initiative of the Patent Office that is only weakly-supported by case law (as I have discussed previously).

Other Recent Developments at the Patent Office

Since I last reported on Patent Office decisions relating to the patent-eligibility of computer-implemented inventions, and in addition to the McCormick decision discussed above, there have been five further decisions published in which patent applications were rejected for failing the ‘manner of manufacture’ test.

1. Sportsbet Pty Ltd v Diogenes Limited [2017] APO 60 resulted from a patent opposition relating to an application for a computer-based method for implementing a so-called ‘partial buy-out’ option in an electronic wagering system.  The examiner must have considered the invention to be patent-eligible, since it had been accepted for grant before being opposed by Diogenes.  However, despite surviving attacks on other grounds, including inventive step, the hearing officer found that the claimed invention was not, in fact, a manner of manufacture.
2. Patent Investment & Licensing Company [2018] APO 1 considered another patent application directed to developments in the gaming industry.  Specifically, the claimed invention related to providing information to punters in relation to the status of gaming machines at a venue, and jackpots available on those machines. The hearing officer considered a number of possibilities for where the ‘substance’ of the invention may reside: in processing of data; in generating recommendations and information; in the presentation of the recommendations and information; in modification of player behaviour; or in some combination of these.  None of these were found to ‘provide a technical contribution to the art’, and thus the claimed invention was not patent-eligible.
3. Google LLC [2018] APO 13 related to a patent application directed to a method of pricing online advertising ‘spots’ (you know what these are – you see them occupying rectangular ‘slots’ on web pages every day).  In particular, the claimed method applies a discount when a user takes an action (e.g. clicking on an advertisement) after being exposed to an advertisement in multiple spots.  The theory seems to be that the advertiser should not pay the full price for multiple spots when a user can only be expected to react (at most) once.  The hearing officer determined that the contribution of the invention to the art lay only in the unpatentable process of discounting the cost, based at least in part on the state of the prior art, which was considered relevant in light of the arguments presented in McCormick.
4. BGC Partners, Inc. [2018] APO 27 considered a patent in the field commonly-known as ‘high-frequency trading’, which refers to the practice of using automated systems to speculate on markets by making large numbers of rapid transactions to take advantage of short-term price differentials.  While computers can be programmed to make rapid decisions on trades, such systems remain limited by the latency (i.e. delays) in the network communications links that are used to transmit bids and other trading information.  The invention in this case was directed to mitigating the impact of such delays.  The hearing officer employed the Patent Office’s adaptation of the UK Aerotel/Macrossan approach to conclude that ‘the “actual contribution” of the invention lies in the scheme for generating a group of trading orders and this scheme is not technical in nature’ and therefore not patent-eligible.
5. Eris Innovations, LLC [2018] APO 17 concerned a patent addressing issues in futures trading.  Again, the hearing officer adopted the ‘actual contribution’ approach derived from the UK law in Aerotel/Macrossan and, again, the contribution was found to reside in patent-ineligible subject matter, namely ‘an algorithm that offsets the effects of convexity bias and/or the net present effect (NPV) effect’.

Note that as in my earlier article identifying recent Patent Office decisions on patent-eligibility of computer-implemented inventions, my point is not to suggest that any of these decisions is necessarily wrong – I have not reviewed the patent specifications in the detail that would be required to form a considered view on the matter.  It remains relevant, however, that there is often substantive disagreement between applicants and the Patent Office – as well as between different decision-makers within the Office – as to how the ‘manner of manufacture’ eligibility test should be applied to this subject matter.  If the law was clear, this would not occur with such regularity.  Furthermore, acceptance rates in the examination section dealing with these inventions would not be notably lower than in other sections, because applicants and their attorneys would have the same levels of confident expectation as exist in relation to other fields of technology about what can, and cannot, be patented.

Other Recent Developments at the Federal Court

There have also been some further developments in cases currently before the Federal Court.  I have written previously about two Patent Office decisions, Todd Martin [2017] APO 33 and Rokt Pte Ltd [2017] APO 34, that resulted in appeals to the Federal Court of Australia: Todd Martin v The Commissioner of Patents, QUD374/2017; and Rokt Pte Ltd v The Commissioner of Patents, NSD1292/2017. 

The Todd Martin case was set down for mediation, which took place of 8 February 2018 but which (unsurprisingly) did not result in a resolution.  The appeal has since been dismissed by consent between the parties, and thus will not be proceeding further.

Meanwhile, the Rokt matter was originally scheduled to be heard on 5-6 February 2018.  However, on 28 November 2017, Rokt filed an affidavit, which I infer comprised expert evidence that it considers pertinent to its appeal.  On 12 December 2017, the court issued an order vacating the February hearing dates, and setting down a timetable for the preparation and filing of further evidence by both parties.  Four days have now been set down for the hearing, from 17-20 July 2018.  The very fact that technical expert evidence is required for a matter that involves what should be purely a question of law is itself indicative of the current state of affairs regarding patent-eligibility in Australia.

Finally, on 3 May 2018, Encompass Corporation filed a Notice of Appeal to a Full bench of the Federal Court of Australia against Justice Perram’s finding that its patents do not claim patent-eligible subject matter.

Conclusion – Uncertainty is the Enemy of Investment

The primary economic purpose of the patent system is to incentivise innovation.  By providing an exclusive right to exploit an invention, for a limited term, the system aims to encourage investment in the development and commercialisation of new technologies, with the investors being secure in the knowledge that a patent can provide protection against copycats for a period sufficient to enable a return to be received on their investment.

Unless, that is, the new technology involves computer software, in which case – as the above discussion illustrates – the availability of patent protection is entirely uncertain.  I read almost daily about how fintech, blockchain technologies, and machine learning are areas in which Australian innovators can excel, and deliver huge benefits to the economy.  And yet these appear to be precisely the areas that are currently least well-supported by the country’s patent system.

Uncertainty around the ability to protect an invention is the worst possible scenario, from the point of view of investment.  An investor typically wants to know what factors will act to minimise their risk when investing.  If technology can – at least in principle – be protected against copying, then that is a definite positive.  Conversely, if the technology is fundamentally incapable of protection via relevant IP rights, then at least the investor can be confident that a competitor cannot hold valid rights that would block commercialisation of the technology.  However, where there is uncertainty around the availability of IP protection, the investor has no confidence in either the ability to defend their own IP position, or freedom-to-operate in the relevant technical space.

Yet uncertainty continues to prevail in the realm of computer-implemented inventions.  Patent law in this area is a complete disaster.  There is nobody who can advise confidently and reliably on what can, or cannot, be protected.  The Patent Office is making up the rules as it goes along, importing tests from the UK and adapting them according to its own interpretation of Australian court rulings.  Federal Court judges appear to be using subject-matter eligibility to plug what they presumably perceive as ‘gaps’ in tests for inventive step and innovative step.  And there is no longer any clear distinction between eligibility and inventive/innovative step, with technical experts now being brought in to inform decision-makers on the ‘state of the art’, so that they can distinguish between what is conventional and what is new, so as to identify the ‘actual contribution’ that a claimed invention makes to the art.

Uncertainty is the enemy of investment.  It is precisely the opposite of what the patent system is supposed to encourage.  At least in the US – where the situation is arguably even worse – the current USPTO Director, Andrei Iancu, appears to recognise the problem, and is willing to stand up and say that people have a right to know what is patent-eligible.  But do not hold your breath – I see no sign that there is anybody of influence in Australia who wants to fix our version of this mess.

Until, and unless, there is the same level of certainty around the patentability of computer-implemented inventions as around those related to mechanical technologies, such as ball point pens, the only thing I can say with confidence is that I will be writing about this problem again in the future.

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