G 1/19: Did this decision surprise us?
G 1/19’s requirement that the technical effect be present throughout the scope of the claim is understood as meaning that when a patent application is drafted, all possible technical applications of a computer-implemented invention must be included and named in the application text in order to obtain protection for them (e.g. Harald Heiske in Bulletin of the German Patent Attorneys 2021, p. 425ff). This article illustrates how this requirement corresponds to long-established case law on essential features and explains it using an example.
- G 1/19 requires that the technical contribution (technical character) or the technical effect be present over the entire scope of the claim.
- These requirements correspond logically to the jurisprudence on ‘essential features’, which applies to all patent applications regardless of the technical field.
- The result of the decision is a consistent application of previous case law and is therefore not surprising, only the formulation of the requirement that the technical contribution must be present over the whole scope of the claim is not as yet very common in the field of computer-implemented inventions.
Decision G 1/19 is only the second decision of the Enlarged Board of Appeal of the European Patent Office on computer-implemented inventions. Despite the fact that the decision was only issued on 10 March 2021, it has already been widely commented on in the literature.
One of the most notable articles, entitled ‘Die Entscheidung G 1/19 der GBK des EPA, oder “Quo Vadis, Patentsystem 4.0?”’ (Decision G 1/19 of the Enlarged Board of Appeal of the EPO, or ‘Quo Vadis, Patent System 4.0?’) is by Harald Heiske and was published in the VPP newsletter, as well as in the Bulletin of the German Patent Attorneys. The decision gives a detailed statement on the patentability of computer-implemented simulations. The decision is based on the previous legal practice of computer-implemented inventions, in particular the COMVIK decision (T 641/00). This decision can be briefly summarised as follows:
- a patentable computer-implemented invention must solve a technical problem , and
- non-technical features, i.e. features that do not interact with the technical subject-matter of the claim to solve the technical problem, so-called non-technical features ‘as such’, do not provide a technical contribution to the prior art and are thus ignored in assessing novelty and inventive step.
In decision G 1/19, the Enlarged Board of Appeal applied the COMVIK approach to computer-implemented simulations and established the following requirements:
- The technical contribution (technical character) or the technical effect must be present over the whole scope of the claim, and
- The technical effect of the subject-matter of the invention may be based on a further technical effect (’further technical use’), in which case this further technical use must be at least implicitly specified in the claim.
In this decision, the explicit or implicit specification of the further technical effect (further technical use) is required for computer-implemented simulations, i.e. that the patent claim must clearly state which technical application the subject-matter of the independent patent claim relates to.
In the above article by Harald Heiske, the following is stated in this regard:
‘G 1/19 now makes a further distinction between technical and non-technical applications and demands that the scope of the claim be limited to specific applications of digital innovation in such a way that no non-technical applications remain in the scope of the claim. Dogmatically, this is justified by the fact that the protection requirements must be present in the whole scope of the claim.
The consequences of these considerations could be very far-reaching. If you were already faced with the problem of formulating applications in the claims in such a general way that, as far as possible, all future applications not foreseeable at the time of the application were covered by the claim, you may now have to also ensure that no actual non-technical applications, or even non-technical applications that are just theoretically conceivable or that cannot be ruled out with certainty, fall within the scope of the claim. Since the exclusion of a negative complement is known to be difficult or impossible, very complex legal problems could arise here in the protection of digital innovations.’
Computer-implemented inventions, such as computer-implemented simulations, are often tools that can be used in a wide variety of applications. As stated in the above article by Harald Heiske, it follows from decision G 1/19 that when drafting patent applications, as many potential technical applications as possible must be explicitly listed in the patent application, otherwise one runs the risk of losing patent protection for the technical applications not listed in the patent application. Inventors in the field of computer-implemented inventions are at a disadvantage compared to inventors in the field of conventional technologies, since one can easily obtain patent protection for conventional technologies without having to consider any subsequent application.
The principle that the subject-matter of the independent claim must make a technical contribution over the whole scope of the claim was derived by the Enlarged Board of Appeal primarily on the basis of decisions T 939/92 and T 1635/09. Decision T 939/92 concerns a patent application where it was questionable whether all the claimed compositions were herbicidally active. Decision T 1635/09 was based on patent claims which could not be restricted to a non-therapeutic method, since the therapeutic and non-therapeutic uses were inextricably linked.
These decisions are therefore primarily based on case law in the field of chemistry and biology. Here it has long been customary to check whether an invention actually covers the entire scope of the patent claim.
However, reference was also made to a more recent decision of the Technical Board of Appeal (T 625/11), which relates to a method in which at least one limit value for at least one operating parameter of a nuclear reactor is to be determined using a computer system. This is a computer-implemented invention that has been evaluated using the COMVIK approach. Here, also with reference to the aforementioned decision T 939/92, the question was raised whether the claimed subject-matter was inventive over the entire scope of the claim.
With the decision of the Enlarged Board of Appeal, a common decision-making practice in the field of chemistry and biology, which had only been applied once before in a recent decision of a Technical Board of Appeal in relation to a computer-implemented invention, was applied to the problem of assessing a computer-implemented simulation.
In this decision of the Enlarged Board of Appeal, the requirement was established for computer-implemented simulations that the technical effect (technical use) must exist over the whole scope of the patent claim.
This requirement, stated several times in the decision of the Enlarged Board of Appeal, had previously been cited in the case law on computer-implemented inventions only once, in decision T 625/11 cited above.
What does this mean specifically?
If, in the case of a patent claim, the question arises as to whether there is a technical effect over the entire scope of the claim, then it is necessary to examine whether all the embodiments covered by the wording of the patent claim produce the desired technical effect.
If there is an embodiment which falls within the wording of the claim but does not achieve the desired technical effect, then the claim does not meet the requirements of G 1/19.
This sounds simple and logical at first.
If at least one embodiment falling under the wording of the claim does not achieve the technical effect, then the subject-matter of the claim does not always solve the problem underlying the invention. The patent claim must therefore be limited by one or more features so that the problem on which the invention is based is always solved.
The patent claim therefore lacks at least one feature that is required to solve the technical problem on which the invention is based. Such features are referred to in case law as ‘essential features’.
Insofar as the decision of G 1/19 requires a further technical effect over the entire scope of the claim, it logically reflects the consistent application of the long-established case law that an independent patent claim must contain all essential features of the invention, essential features being all those features that are necessary to solve the technical problem that the patent application relates to.
The conclusion that the requirements for computer-implemented simulations set out in the decision of the Enlarged Board of Appeal are logically equivalent to the case law on the essential features of a patent claim has been derived in the abstract above. In the following, I would like to briefly explain this using the subject-matter of T 1227/05 (Infineon 1/f simulation) as an example.
Claim 1, on which this decision is based, relates to a computer-aided method for the numerical simulation of a circuit which is subject to 1/f noise effects, which are simulated using certain method steps indicated in claim 1.
Claim 1 of the corresponding US patent (US 6,795,840 B1) corresponds essentially to claim 1 of the European patent, although it is not limited to the numerical simulation of a circuit, but the application of the simulation is left open (claim 1: ‘A computer programmed to process a method of determining sequences of random numbers of a 1/f noise comprising the following steps: …’). The description also indicates that this simulation can also be used to predict events in the financial market. The subject-matter of the US patent therefore relates to a method that can be used both for commercial purposes, for determining events on the financial market, and for technical purposes, for simulating circuits. The US patent consequently protects a simulation method which, in accordance with European patent practice, can be used for a technical application (= simulation of circuits) as well as for a non-technical application (= financial market analysis).
The simulation as such is a mathematical procedure for describing a specific type of noise. Noise can appear both in non-technical data, such as financial data, and in technical objects (e.g. the noise of a certain current level). Therefore, the simulation of noise as such within the meaning of European patent practice under Art. 52(2)(a) in conjunction with Art. 52(3) EPC is to be assessed as a mathematical method as such and therefore not as technical.
Only by the limitation to the ‘numerical simulation of a circuit’, as contained in claim 1 of the European patent, is the application of the simulation specifically limited to a technical application. In the European patent EP 1 257 904 B1, the following is indicated as a problem:
‘It is the object of the invention to specify a method for generating a sequence of random numbers of a 1/f noise that can be carried out quickly and with little computing effort. It is also the object of the invention to specify an improved method for simulating a technical system that is subject to 1/f noise. Finally, a computer program for determining sequences of random numbers of a 1/f noise should also be indicated, which can be executed quickly and which demands few resources of a computer system.’
The object of specifying a method for simulating a technical system is achieved with claim 1 of the European patent compared to claim 1 of the corresponding US patent in that the subject-matter is limited to the simulation of a circuit. Compared to the patent claim of the corresponding US patent, a feature is required that limits the subject matter of the claim in such a way that non-technical embodiments, such as the analysis of financial market data, are excluded from the scope of the claim. Since this feature is necessary to solve the technical problem, it is an essential feature within the meaning of the case law on essential features of patent claims.
This means that the requirements laid down by the Enlarged Board of Appeal for patent applications relating to computer-implemented simulations would have led to the same outcome if previous jurisprudence on the essential features of a patent claim had been applied consistently. These requirements therefore do not represent any changes to previous case law, but merely a consistent application of case law that has been common for decades on the essential features of a patent claim.
Since the case law on the essential features of a patent claim applies in principle to all patent applications regardless of the technical field, the requirements established by the Enlarged Board of Appeal apply not only to computer-implemented simulations, but also to all other computer-implemented inventions and, in principle, to all inventions.
This will have little impact on the practice of patenting inventions in traditional technologies, such as mechanics, since mechanical inventions are usually defined by physical features, in that a particular mechanical body is specified. The application of such a mechanical body is considered technical according to existing case law. Accordingly, pursuant to the requirements of the Enlarged Board of Appeal, the indication of physical features in a claim implicitly limits the subject-matter to a technical application. This is different for computer-implemented inventions in general. The requirements for computer-implemented simulations established by the Enlarged Board of Appeal apply equally to all computer-implemented inventions, which usually contain a combination of technical and non-technical features in the patent claims, so that it is not inherently clear whether there is a technical effect.
In the case of computer-implemented inventions, the use of features that can be implemented by software poses the problem that it is not inherently clear whether these features represent a technical or a non-technical feature. They only form a technical feature if they contribute to the solution of a technical problem, which must be checked separately.
Therefore, with computer-implemented inventions, unlike with conventional technologies, it is not self-evident that the subject-matter defined in a patent claim relates to a technical application which, in case of doubt, must be explicitly defined in the patent claim.
If a patent claim does not meet this requirement, then an essential feature for solving the technical problem would be missing, and non-technical objects, such as a mathematical method for simulating 1/f noise according to US 6,795,840 B1, could be placed under patent protection.
Even if it initially appeared that decision G 1/19 set out new additional requirements for computer-implemented simulations or computer-implemented inventions, the requirements explicitly formulated therein on the technical effect over the whole scope of the claim represent a consistent application of the previous case law on essential features of patent claims. The result of the decision is therefore not surprising, only the wording of the requirements is not as yet very common in the area of computer-implemented inventions.
 VPP newsletter no. 2, 2021 and Mitteilungen der deutschen Patentanwälte (Bulletin of the German Patent Attorneys) 2021, p.425 ff.
 EPO, G 1/19, para. 30 (G).
 EPO G 1/19, para. 30 (F).
 EPO G 1/19, para. 82, 84, 94, 95, 98, 129.
 EPO G 1/19, para. 137.
 EPO G 1/19, para. 82, 98, 116, 129.
 EPO G 1/19, para. 109, 129-133.
 Case Law of the Boards of Appeal, 9th edition, II.A.3.2 Indication of all essential features.
 US 6,795,840 B1, column 1, lines 19-23.
This article was originally published in “Mitteilungen der deutschen Patentanwälte 12/21” and was prepared by HGF Partner & Patent Attorney Bernhard Ganahl.