The EPO’s recent Enlarged Board of Appeal hearing on patentability of computer simulations was important not only for its legal ramifications. The participants were present, but for the first time ever, and in an immensely practical move, the proceedings were made open to the public by video. Alex Worley attended and muses here on the experience, the law, and the potential outcome.
Lights, camera…
On 15th July 2020, at 9am, the members of the Enlarged Board in the case of G1/19 entered the brightly lit hearing room to a perfectly silent audience of 1600 remote onlookers. There were as many cameras cast over their actions as one could expect in any live television studio.
From the viewers’ standpoint, the unreality of the situation was stark. It is never going to feel normal to watch people cover their faces any time they stand up to leave the room. This unnerving feeling is only intensified by images of a hearing room, purpose built to house the many usual translators and other onlookers, standing awkwardly empty. Nevertheless, everything else about this process feels familiar. The stilted interactions between the board and the speakers, satisfying the strict order of the procedure, and eventually the same ultimate decisions being made. Once we’d got past the weird procedure, there was some good law going on, and whilst we wait for the written decision to be handed down, we round-up our observations below.
Taking the path of least resistance
We’ve been slightly disappointed before when waiting for a really detailed decision from the Enlarged Board on software inventions (G3/08). There’s something quite underwhelming about discovering, after such a long wait, that what had been seen as a potential inconsistency in the way the law has been applied before is deemed not an inconsistency at all. And, whatever the outcome of the present decision on software patentability in Europe may be, we expect more unanswered questions are to come (well, at least one).
The G1/19 referral arose from a case where the claims include features directed to the simulation of pedestrians moving through a space. The simulation would be carried out on a computer, and this particular simulation is supposed to be clever because of the way the pedestrians are individually modelled. If you follow the case law you can skip the rest of this paragraph, but in essence technicality of the claims is a key issue in Europe, stemming from the philosophy that non-technical things are not worthy of patent protection. Other non-technical things include scientific discoveries and theories, purely mental acts, and mathematical methods. A mathematical method cannot in itself be protected by a patent, but if it is applied in a technical way, it can become worthy of a patent. If a mathematical method actually affects the operation of a standard computer, then the method may be patentable under those circumstances. A good example of this would be where an algorithm is designed for better operation of the computer based on internal functioning of the computer, say the algorithm is designed such that it suits the way in which data is passed between a CPU and a GPU. Another example where a mathematical method may have a technical effect is where there is a technical application. Examples that the EPO has given include “digital audio, image or video enhancement or analysis, e.g. de-noising, detecting persons in a digital image, estimating the quality of a transmitted digital audio signal”. Even from this example, you can imagine how the line between real and virtual starts to get a little hazy.
Simulation type claims have been allowed by the EPO before, the key case being T1227/05, which relates to a simulation of an electronic circuit subject to noise. The Board of Appeal in that case ordered the claims to be allowed, the claimed subject matter determined to be “technical”. G1/19 arose because there was felt to be some sort of distinction with simulating pedestrian movement. Simulations by their nature involve mathematical methods but, also due to their nature, do not physically exist outside of the computer on which they are run. A key use of simulations is in the development of engineering products without the need for costly production of physical prototypes. Basic simulation tools even now turn up in free versions of CAD software, and you can perform these simulations on a laptop computer with meagre amounts of memory and processing power. Improvements in simulations can radically speed-up the design process by providing more reliable data about the ultimate performance of a design. Some of these improvements will inevitably be based on the mathematical methods being used.
What did we hope for with this case? Well, the questions posed included a few points regarding whether simulations claimed as such could be considered technical. High up the list of things we wanted to see was a response to the first part of question 2: “what are the relevant criteria for assessing whether a computer-implemented simulation claimed as such solves a technical problem?”. Sadly, the Enlarged Board have made it quite clear that they do not feel they have to provide these. So, for now, we will not be getting any kind of test which lets us know when a simulation is technical. For now we have been introduced to some new terms which will undoubtedly make their way into the European patent attorney’s vernacular. “Potential technical effects” and “Virtual technical effects” look set to provide some determination over whether subject matter will be deemed technical.
We’re in limbo until we see the written decision, but we thought we’d end by taking a punt at what we think the answers to the questions might be. Don’t hold us to it, but at least it might help understand what the case was directed at!
1. In the assessment of inventive step, can the computer-implemented simulation of a technical system or process solve a technical problem by producing a technical effect which goes beyond the simulation's implementation on a computer, if the computer-implemented simulation is claimed as such?
AJW – I think we’re going to see a “yes” from the Enlarged Board here. This will be consistent with the T1227/05 case. But this does not necessarily mean the present claims will be allowed.
2a. If the answer to the first question is yes, what are the relevant criteria for assessing whether a computer-implemented simulation claimed as such solves a technical problem?
AJW – we are very unlikely to have a response to this query. The Enlarged Board made it clear that they did not have to address every point put to them.
2b. In particular, is it a sufficient condition that the simulation is based, at least in part, on technical principles underlying the simulated system or process?
AJW – here, we are likely to see the description of “virtual” and “potential” technical effects. Personally, We cannot see right now how the virtual and potential technical effects will distinguish the pedestrian simulation from the simulation of noise in an electronic circuit. It will be interesting to see how this line is drawn.
3. What are the answers to the first and second questions if the computer-implemented simulation is claimed as part of a design process, in particular for verifying a design?
AJW – in view of how we expect the first two (one and a half?) questions to be answered, the answer to this question will likely be that there is no difference even if the simulation is claimed as part of a design process for verifying a design. The detail will be in the virtual and potential technical effects.
If you would ever like to have a word with Alex about this article, feel free to email him.