[Author: Yan Wang]

In the practice of patent examination in China, to determine whether an invention has prominent substantive features is to determine, to the person skilled in the art, whether the claimed invention is non-obvious as compared with the prior art. Usually the following three steps are followed to determine whether a claimed invention is obvious as compared with the prior art.

i. Determining the closest prior art;

ii. Determining the distinguishing features of the invention and the technical problem actually solved by the invention;

iii. Determining whether or not the claimed invention is obvious to a person skilled in the art.

The closest prior art refers to a technical solution in the prior art that is the most closely related to the claimed invention, which shall be the basis for determining whether or not the claimed invention has prominent substantive features. The closest prior art may, for example, be an existing technology in the same technical field as the claimed invention, and its technical problem to be solved, technical effects, or intended use are the closest to the claimed invention, and/ or has disclosed the greatest number of technical features of the claimed invention; or be an existing technology which, despite being in a different technical field from the claimed invention, is capable of performing the function of the invention and has disclosed the greatest number of technical features of the invention. It should be noted that, when determining the closest prior art, whether they are in the same or similar technical fields should be considered first.

As a comparison object, in the opinion of EPO, the closest prior art is generally the subject-matter disclosed in the prior art document that has the same aim or is directed to the same objective as the application being examined, and the prior art should have the most relevant common technical features with the application being examined. When selecting the closest prior art, the first consideration is that it must have the same purpose or effect as the claimed technical solution, otherwise, the prior art cannot lead the skilled person to deduce the claimed invention in an obvious manner.

As the closest prior art, it must be described as being able to fit the claimed purpose of the claimed technical solution, rather than showing superficial structural similarity to the claimed technical solution. Ideally, the above objects or objectives should have been mentioned in the prior art as worth achieving. Even under non-ideal conditions, as the closest prior art, it should at least involve the same or similar technical problems as the claimed technical solution, or at least should involve the same or closely related technical fields as the claimed technical solution.

In contrast, determining the closest prior art based on the said "three-step" method does not necessarily consider or satisfy the above conditions, although the "three-step" method also stipulates that when determining the closest prior art, the prior art in the same or similar technical field should be considered first, and further consideration should be given to the prior art with the technical problem solved by this application and the closest technical effect achieved, but even under the premise that there is no such prior art, the prior art with the most disclosed technical features can also be further considered.

So why do the above differences between EPO and CNIPA appear, and what practical significance do the above differences exert in specific examination practice? First of all, the reason why the "problem-solution" law stipulates selecting the closest prior art in this way is to make the starting point of the judgment process of inventiveness closer to the situation faced by the inventor, allow the technical problem to be considered as the driving force of technical improvement, and enable sufficient consideration of the existing technology. If a prior art does not mention the technical problems recorded in the patent specification or the technical problems related to the technical problems that can be derived from the patent specification, then no matter how many common technical features the document has with the patent, the prior art typically does not qualify as the "closest prior art".

In the case of T 835/00, D1 was considered as the "closest prior art" but it does not mention any problem addressed by the involved application. As a consequence, a technical problem to be solved in the involved application is unrelated to the disclosure of D1 while the solution to the technical problem was then found obvious in the light of the disclosure of D2. The Boards of Appeal of EPO stated that it was a fatal defect that no relevant technical problem could be formulated from the closest prior art without inappropriate hindsight, because, without such hindsight, any attempt to establish a logical chain will run into difficulties at the start.

It can be seen from the above comparison that when selecting the closest prior art, EPO pays more attention to reproducing the process of determining the inventive step as the invention process of the inventor, and fully analyzes whether a person skilled in the art can obviously find a suitable solution from the prior art to the technical problems encountered by the inventor. In contrast, when selecting the closest prior art, CNIPA does not care about the invention process of the inventor, but pays more attention to whether the claimed invention has sufficient inventiveness as compared with the overall level achieved in the prior art. Therefore, when selecting the closest prior art, it is not important whether it records the technical problem to be solved by the claimed invention, but it is particularly important whether it is the existing technology closest to the claimed invention in the prior art, and such “closest” is further reflected in the similarity of technical fields, technical problems and technical features.

Of course, the criterion adopted by the EPO for choosing the closest prior art can greatly avoid "hindsight bias" in the judgement, but in a context where technology is becoming increasingly complex nowadays, such criterion also inevitably makes the judgment of inventiveness mechanical, maybe lowering the criterion of inventive step in practice. In contrast, although the criterion of choosing the closest existing technology adopted by the Chinese Patent Office enables sufficient examination of the difficulty of improving the technical solution based on the existing technology, it will inevitably introduce more "hindsight bias" into the judgement and excessively raising the criterion of inventive step in practice. How to seek a more objective and accurate solution is still an urgent problem to be studied.