Securing software patents through the European Patent Office
In June 2014 the US Supreme Court provided its opinion in Alice Corporation v CLS Bank. This is the latest in a number of decisions from the higher US courts which confirm the considerable uncertainty which has existed in the United States in recent years as to the boundary between what is and is not patentable. The Supreme Court’s opinion follows a high-profile decision from the US Court of Appeals for the Federal Circuit in which the court sat en banc and reached what one of its judges referred to as “judicial deadlock”, such was the disagreement as to what should be patentable and how patentability under the US statute should be assessed.
The recent line of US jurisprudence shows an appetite to restrict the bounds of patentability in the field of computer-implemented business methods. It now seems to be settled that where a patent claims an abstract business method implemented on a computer, the mere computer implementation will be insufficient to remove the invention from the realm of the abstract and is therefore unpatentable. Rather, there is a need for an inventive concept which transforms the abstract idea into something which is patentable. The Supreme Court, citing its own earlier opinion, identified a requirement for “an element or combination of elements that is sufficient to ensure that the patent in practice amounts to significantly more than a patent upon the ineligible concept itself”. Put another way, the court seemed clear that the invention should lie outside the sphere of the abstract method and lie instead in the computer implementation itself.
Some European practitioners could be forgiven for feeling a sense of déjà vu on reading the Supreme Court’s opinion. After a rocky few years in the late 1990s and the early part of the 21st century, the European Patent Office (EPO) has now adopted an approach which is intended to focus only on the technical features of a claim when assessing patentability. That is, the mere presence of a computer or a generic computer implementation cannot in itself save subject matter from exclusion from patentability. In order to be patentable, an invention must provide a technical solution to a technical problem defined with reference to the closest prior art, and features excluded from patentability cannot contribute to the technical solution which the invention provides. In practice, it seems that this is a slightly different route to arriving at the position advocated by the US Supreme Court in its latest opinion – the invention itself must lie outside the abstract (in the United States) and outside the specifically excluded subject matter (in Europe).
Since the creation of the European Patent Convention in 1973 and its entry into effect in 1977, European law has made clear that some things are not inventions for the purposes of patent law. These include computer programs, business methods and mathematical methods. However, the exclusions are limited to the listed ‘things’ as such. Unsurprisingly, this less-than-clear statement as to what is and is not patentable has created considerable uncertainty and debate as to the bounds of patentable subject matter in Europe. There can be no doubt that this uncertainty has caused considerable difficulties for users of the European patent system over many years.
The uncertainty in Europe has created a situation where many European companies – particularly small and medium-sized enterprises – take the erroneous view that their developments cannot be patented because they are, for example, implemented by way of computer programs and computer programs are one of the listed exclusions from patentability.
This can be contrasted with the historically more lenient approach in the United States. Only very recently, the US tribunals took the view that anything “under the sun made by man” should be patentable. This led to thousands of patents directed to entirely abstract methods of doing business (eg, methods of operating a pension scheme or trading shares were routinely fair game for patent protection), and many thousands more for computer-implemented methods which many would (often wrongly) consider unpatentable in Europe. It is therefore unsurprising that companies based in the United States were and are more likely to seek – and therefore obtain – patent protection for their developments in the many fields where computer software is now commonly used. This problem became more significant as computers were applied to more and more fields of everyday life.
The stable European approach
While the law in Europe has created considerable uncertainty by its drafting, in recent years there has been greater consistency and certainty as to what is patentable and how patentability should be assessed. The EPO now adopts a two-stage approach.
First, there is a consideration as to whether a claimed invention includes any technical features. Any claim including a computer, computer network or any tangible technical element will pass this test.
However, the substance of the approach is considered in a second stage which is applied together with the assessment of inventive step. Under European practice, inventive step is considered by defining a problem which the claimed invention can be said to solve relative to the closest prior art (normally a document describing something similar to, but different from, the invention). There is then a consideration as to whether the solution of the invention would be obvious to the nominal person of ordinary skill in the art. The approach to the exclusions from patentability makes clear that the problem defined for the purposes of the analysis must be a technical problem and the solution must likewise be a technical solution. This has effectively excluded from the assessment of inventive step non-technical features directed to, for example, business or administrative schemes.
While there is not – and probably never will be – any definition of what ‘technical matter’ constitutes, case law suggests that one useful way of determining whether something is technical is to ask whether its implementation would be the preserve of a technically skilled person (eg, a computer scientist) or the preserve of a business person (eg, an accountant). Features which are the preserve of those of technical skill can be used in the assessment of inventive step. Perhaps more importantly, those regularly advising clients in this field now have a good feel for what the EPO or a national court is likely to consider technical, allowing practitioners to advise clients reliably as to the patentability or otherwise of particular inventions.
It can also be observed that the EPO’s focus on technical features moves away – at the second step of its approach – from the exact wording of the exclusions. This is appropriate because the exclusions are a non-exhaustive list. Given that the specified exclusions are excluded only as such, it is important to consider what an invention provides over and above the excluded subject matter.
Two examples of the EPO’s approach may aid understanding.
First, consider an invention directed to a computer-implemented method of operating a pension scheme. The claim is defined in terms of the method for operating a pension scheme, but is limited to the method being carried out by a suitably programmed computer. No detail of the computer implementation – over and above the requirement that the computer be programmed to carry out the steps of the method – is provided in the claim.
This invention clearly meets the first stage of the EPO’s approach. The limitation relating to computer implementation is a technical feature of the claim, so the claim does not relate to excluded subject matter as such.
At the second stage, however, it is necessary to identify a technical problem to which the invention provides a technical solution. In a case such as this the non-technical features (ie, those relating to management of the pension scheme in a financial or administrative sense) are included in the definition of the problem, and the technical problem would likely be defined as ‘how to implement the method of managing a pension scheme which includes the claimed features’. The solution to this problem then becomes the implementation of the method on a computer. Given that there is nothing inventive in the implementation itself, the solution would likely be considered obvious (ie, it is obvious to implement a pension management method using a computer) and the invention would therefore not be patentable for that reason.
As another example, consider a computer-implemented method for trading shares between traders in a computer network, the method of which includes features relating to an encryption protocol used to encode communications between the traders and the communications of which include share-dealing transactions. The method has as its aim the provision of secure communications between traders for effective share dealing.
Applying the two-stage approach outlined above, the invention clearly includes technical features arising from the computer implementation. However, the technical features go further. They include features relating to the encryption protocol used to encode communications between traders. While the communications include share-dealing transactions (ie, data of a business nature), the concept of encryption is a technical concept and as such the detail of encryption (in addition to the generic specification of computer implementation) saves the method from being excluded.
At the second stage, it is again necessary to identify a technical problem to which the invention provides a technical solution. Here, it is again legitimate to include the non-technical features in the definition of the problem. As such, an appropriate definition of the problem might be how to provide for secure communication of share-dealing transactions between traders. The solution is now the provision of computer-implemented communication using the encryption protocol specified in the claim. Given that data encryption is a technical rather than business matter, the detail of the encryption can legitimately contribute towards the method’s inventive step. It is then necessary to ask whether the solution involves an inventive step or is obvious. This will depend on the nature of the encryption specified relative to other known encryption protocols, but in principle the invention here is patentable if the encryption used is not obvious.
Certainty of EPO’s approach
The EPO’s approach has now been applied for some 10 years and is not without its critics. The UK courts, for example, have repeatedly refused to follow the European approach (although the outcome of their decisions will in most cases be the same). An attempt to codify the approach into a European directive famously failed, but a request that the Enlarged Board of Appeal of the EPO (the EPO’s highest judicial body) consider the issue on the basis that the law was unclear was refused on the basis that there was no divergence in the case law, which seemed to the Enlarged Board of Appeal to be settled.
There is therefore every indication that the EPO’s position is unlikely to change. This is surely good news for users of the European patent system. The approach is predictable and applicants should therefore be able to obtain reliable advice at an early stage as to how their applications are likely to fare in the EPO.
European practitioners will find much familiar in the reasoning which underlies the US Supreme Court’s most recent opinion. In essence, the court held that the invention itself must be more than merely abstract. This is very similar to the EPO’s approach of requiring that the invention provide a technical solution to a technical problem, particularly because in many practical cases ‘non-technical’ in European practice will be similar to ‘abstract’ in US practice. It does seem that the once-lenient US approach has seen considerable limitation, which in practical terms brings it close to the hard-won certainty of the EPO approach.
Additionally, the US Supreme Court’s guidance is explicit that substance, not form, is what matters in questions of patentability, thereby confirming that equivalent method and apparatus claims will stand or fall together. This is again consistent with the European approach. There is a common misunderstanding that the EPO’s approach prohibits claims directed to computer programs or data carriers. This is wrong. If the underlying method is patentable, the EPO will allow it to be claimed in any (reasonable) form, including as a computer program, a data carrier carrying such a program or a computer programmed to carry out the method.
Of course, while the EPO approach is settled, it will inevitably take the US patent office and the lower US courts time to develop their approach based on the Supreme Court’s latest guidance and it will be necessary to wait and see how the approach develops. However, there is now reason to be optimistic that the much-hoped-for transatlantic consistency might be close at hand.
Every organisation should have an IP strategy. This is particularly the case for organisations engaged in work which might create patentable material. Some organisations might take a conscious decision not to seek patent protection, which might well be reasonable in the specific circumstances. However, what is important is that the decision be taken from a properly informed position, particularly given the confusion which still ensues in this difficult area.
Where organisations are interested in obtaining protection in Europe, it is important to bear this in mind at an early stage. Relatively straightforward steps can be taken at the stage of drafting an application which greatly enhance the prospects of patentability at the EPO. As such, those based outside Europe and consulting with non-European counsel would typically be well advised to seek input from a European patent attorney with specific experience of the exclusions from patentability at the earliest stage possible. This is particularly important where an international (PCT) application is filed, because it is beneficial to ensure that the invention is presented in the best possible light in the PCT application. However, there are advantages to obtaining European advice at the time of filing of the first application so as to ensure that all information which is likely to be required to support patentability in Europe is included in the first filed application, thereby maximising the chances that the application in Europe can fully benefit from the priority date of the first filed application.
Finally, it is important to be aware that some inventions will not be patentable in Europe. The consistency of the EPO’s approach now lets those inventions be identified at an early stage, and obtaining patentability advice at an early stage will allow organisations to spend their patent budgets on the cases which are most likely to mature into valid granted European patents.
Software is certainly patentable at the EPO. The rumours to the contrary need to be quashed. However, those seeking to protect software-based inventions should be aware what is likely to be patentable and what is likely needed to demonstrate patentability to the EPO.
Marks & Clerk
1 New York Street
Tel +44 161 233 5800
Fax +44 161 236 5846
Mark Kenrick has a background in computer science and considerable experience of the European Patent Office’s (EPO) approach to excluded subject matter. He regularly appears before the UK Intellectual Property Office and the EPO on the issue of software patentability and has handled appeals from the Intellectual Property Office to the High Court.
Mr Kenrick graduated from the University of Manchester with a first-class honours degree in computer science and holds the degree of LLM in IP litigation with distinction, as well as postgraduate certificates in IP law and IP litigation. He is a professional member of the British Computer Society and a member of the Society for Computers and Law.
European patent attorney
David Robinson graduated from the University of Manchester with a degree in computer science and maths, and obtained a master’s degree in mathematical logic and the theory of computation and a PhD in automated reasoning before training as a patent attorney. He now specialises in patent matters relating to computer software, computer hardware and business methods. Dr Robinson has particular experience in data storage systems, computer-implemented mathematical methods, image analysis, data analysis, communications protocols, computer graphics and computer-implemented business methods. Dr Robinson’s practice includes work emanating from a broad range of clients, from large multinational IT companies to universities and small and medium-sized enterprises.