EP 3 158 325 B1 relates to a microwave measuring device and method for the verification or rod-shaped articles or of a rod of material of the tobacco processing industry and machine of the tobacco processing industry
Brief outline of the procedure
The opposition was rejected and the opponent appealed.
The parties were summoned to OP.
In reply to the board’s communication pursuant to Art 15(1) RPBA, the proprietor stated that it would not participate in the OP before the board and withdrew its request for OP.
The OP before the board therefore took place in the absence of the proprietor.
The board decided revocation of the patent. None of the requests on file, as granted and AR 1-8, were novel over E1=US 3 581 190.
The case is interesting as it deals with the absence of the proprietor at the OP before the board and the latter’s decision on AR8.
The board’s decision
The board was aware that the argument that claim 1 of AR8 lacked N in relation to E1 was raised for the first time during the OP before the board. However, the board could not see that this infringes the right to be heard, cf. Art 113(1), of the proprietor, who was absent from the OP.
The written submissions by both parties on AR8 were brief in the written appeal proceedings. In particular, N and IS were dealt with by the proprietor in a generalised sentence: “None of the documents in the proceedings show this combination of features. Therefore, AR8 also meets the requirements of R 80 “.
This raises the question of whether the proprietor’s submission regarding N and IS of AR8 was sufficiently substantiated. AR8 could therefore have been disregarded by the board on the grounds of lack of substantiation, cf. Art 12(3+5) RPBA
At the very least, the proprietor had to expect that a detailed discussion of N and IS, which had not taken place to date, would be held for the first time at the OP.
In its communication under Art 15(1) RPBA, the board held that AR1-7 were considered not allowable due to lack of N over E1.
Finally, the board was convinced that, when examining a request that has not yet been examined in the first instance proceedings, it must at all times be permitted to investigate ex officio in accordance with Art 114(1) or, where appropriate, to take up new arguments put forward by the other party.
This is because, in the case of such a request, it cannot be assumed from the outset that the subject-matter of the request meets the requirements of the EPC. The board cannot consider such a request allowable solely on the grounds that the other party has not raised any objections to it.
This is even less possible if plausible objections have been raised by the other party, albeit belatedly, as in the present case.
Although the board is not obliged to conduct its own exhaustive search and examine the subject-matter in its entirety, it must be satisfied that the facts and evidence already presented in the proceedings and brought to its attention do not preclude the maintenance of a patent on the basis of such a request.
It follows that an objection which it considers to be substantial, which is based on facts and evidence already dealt with and which clearly calls into question the allowability of a claim, may be raised at any time, even if the patent proprietor deliberately fails to attend the OP.
The board saw no reason to disregard the new objection of lack of N of claim 1 of AR8 in relation to E1 pursuant to Art 13(2) RPBA. In the grounds of appeal, the opponent raised an objection of lack of IS against AR8, which would have had to be dealt with in any case.
The board therefore considered it an exceptional circumstance that document E1, which was considered to be prejudicial to N in relation to the higher-ranking requests, was also taken into account in the discussion of the distinguishing features.
Therefore, the board saw no reason to disregard the objection raised for the first time in the OP by the opponent that claim 1 of AR8 lacked N over E1 pursuant to Art 13(2) RPBA.
Comments
It is manifest that the board wanted to get read of the case.
However, when looking at G 4/92, one wonders whether the board could actually act as it did.
The headnote of G 4/92 says:
1. A decision against a party who has been duly summoned but who fails to appear at oral proceedings may not be based on facts put forward for the first time during those oral proceedings.
2. Similarly, new evidence may not be considered unless it has been previously notified and it merely supports the assertions of the party who submits it, whereas new arguments may in principle be used to support the reasons for the decision.
It remains that the way the board dealt with the case is subject to caution. Coming up with a lack of N never discussed before, not even brought forward by the opponent, was quite daring.
I doubt that a board would have accepted such a way of dealing with a case when an OD would have behaved like the present board, and introdced a N objection never raised before. The likelihood that the board would have concluded to a SPV is rather high.
The board considered that AR8 was barely substantiated. This could have been a good reason not to admit AR8, and would also have allowed revocation of the patent.
The present decision has been issued on 21.10.2025, and notified on 24.10.2025. The time limit for filing a petition for review has thus lapsed. As the proprietor did not attend the OP, it is unlikely that he would have filed a petition for review.
It should however be a warning for proprietors not attending OP: insure that the argumentation in favour of any filed requests is complete.
On the procedure
E1=US 3 581 190 was not mentioned in the ISR established by the EPO. It bears the IPC classification symbols G01N22/00 and H01P7/06 as well as G01N27/78.
The patent bears the IPC classification units G01N 22/04 and A24C 5/34. The search was carried in the IPC classification units GO1N and A24C.
It is thus surprising that E1 was not found during the IS carried out by the EPO..
Comments
11 replies on “T 0077/23-Absence of the proprietor at the OP-Revocation on the basis of higher ranking requests”
Dear Mr. Thomas, I have never been an examiner and I have very little experience in doing patent searches, so please excuse me if my question is a naive one.
You often mention — like in this blog post — that it is surprising that the novelty-destroying document was not found during the search, even though it was in the same IPC classes as the opposed patent.
My question is: why is it surprising? Here is why I am asking:
1/ Opposition cases carry a very strong selection bias: an opposition is filed only if the opponent has relevant prior art on hand, and the first place to look for it is in the opposed patent’s classes.
2/ There are so many published patent documents (more than 160 million, according to Espacenet’s home page) that any class is bound to have a large number of documents (thousands? tens of thousands? I honestly have no idea), and from time to time the most relevant one might be overlooked, even after a diligent search. Nothing has a 100% success rate, and after all, opposition proceedings exist precisely because invalid claims can slip through examination.
3/ Here, E1 is a rather old US patent (filed in 1968, published in 1971). In my experience, older patents tend to use different terminology, so they can slip through a keyword search, and/or it can take you a good bit of time to realize that they read on your claims. Time which one does not always have, as any search has to end at some point.
I would be grateful if you could enlighten me (and your blog’s readers). Thank you.
@ Extraneous Attorney,
Your question is not naïve, it deserves an answer like any question. When teaching, I always say to the audience, there are no naïve or stupid questions. Answers can be naïve or stupid, but my aim is to avoid them.
1/ I agree with you that opposition cases carry a very strong selection bias. This is why it is not possible to consider the result of oppositions as fully representative of the quality of the work of the EPO. I do however consider that they are representative of a trend.
If have not yet finalised the figures for 2025, but for 2024, only 15% of patents survived an opposition and were maintained as granted. 40% of patents were revoked, and 35% were maintained in amended form, i.e. with a restricted scope of protection. This goes on par with the fact that in less than 10% of the cases opponents came with documents which could not be found in EPO’s search files. I think here of public prior uses, PhD dissertations or catalogues of manufacturers and the like.
On the other hand, what is a patent good for a proprietor if 85% of patents come out maimed from an opposition before the EPO? Opposition should not become the reparation workplace for search and examination procedures which are not up to what they should be. If the search has failed, then examination can barely be better, as the relevant documents were not found.
2/ It is clear that with the exponential increase of patent and non-patent literature, it is not thinkable to carry out a purely manual search in a paper documentation. In the early 70s until the early 80s this was still the case, even at the EPO. Nowadays a computer aided search is an absolute necessity, be it only for the sheer weight of the documentation. The EPO has done pioneering work on computer aided search, but it seems that it is the way the system is used which causes problems. In the past the EPO used EPOQUE, presently it uses an ANSERA-based search tool. The Academy of the EPO offers training sessions on this tool in 2026, but only to examiners of national offices.
Irrespective of the tool used, a computer search needs much more intellectual input as a manual paper search. One needs to prepare the queries and adapt those to the intermediate results.
When an examiner starts a search he has no idea whether the application under his eyes will be the pearl in the oyster or will end up in the cemetery of patents never having ended in the light. A rather old study has shown that patents which are incorporated in goods ending on the market represent the tip of an iceberg. The vast majority of patents end up beyond the surface. This is not useless, as patent documentation has a further function: publish technical knowledge helping technology to progress.
2a/ Why is possible that the relevant prior art can be missed?
The reasons are multiple:
– Search queries, especially based on keywords, have a limited possibility to find the relevant documents. The problem of synonyms, different terminology and misspelled words can easily be overcome by a human brain, but are much more a problem in a computer search. Some technical areas, e.g. mechanics, have lots of figures. Searching figures by computer is a very difficult task. This is also a problem. Even electronic circuits can be represented in such a way that the circuit function is not immediately apparent. A human brain is quicker to spot this. Chemistry or biotech appear to me to be easier to search, as it is about formulas.
– In the past, examiners used to classify new patent and non-patent literature entering the search documentation, but from the information I gained from “Search and Examination Matters”, that this is not any longer the case at the EPO.
As search examiner I was also classifying in specific IPC classes. This action allows to gain a good idea of what is to be found in the corresponding class. When a document meant to belong the class I was responsible for, did not fit, I could not merely bin the document, I had to think to which colleague I should be sent. A lot of information could therefore be gained from classification work. This has gone.
– Again from “Search and Examination Matters”, I also got the information that examiners do apparently not any longer use classification when searching. The search has apparently become purely statistical. I am therefore not convinced that this is really the best way to find the relevant documents.
– There is no doubt that any search has to be limited in time. Time spent on a search could be infinite, but the probability of finding a relevant document goes quickly towards 0.
If the examiner has found documents he thinks relevant, he will stop searching. Whether the documents are really the most relevant ones is a question which cannot be answered. It is often answered by the opponent coming with more relevant documents.
– There is certainly a time when it necessary to stop the search. This time can be fixed by superiors, so as to increase production/productivity. I therefore link the apparently poor search results to the extreme pressure on the examiners to accelerate search and examination fore the sake of a higher production.
– That IA has made inroads in the patent world, and in patent search is not to be denied. But to me IA is primarily artificial and not intelligent, as it all depends on how it has been trained! Any bias in the training inevitably comes out in the output.
– A computer search is, in principle good at finding documents which can be deemed novelty destroying. To my perception, it is much more difficult to find documents to be combined to assess inventive step. There an intellectual input is a necessity.
– I do not know if it is still the case, but especially in The Hague, there was a theory that a good search is a search revealing a X document. Nothing is more fallacious. I have done some searches where there was truly nothing to find, but those were exceptions rather than the rule. Finding a true X document is a challenge.
– In view of the high pressure on examiners, there is another phenomenon, which I call the “Super BEST Effect”. As search examiner, you do not find relevant documents in the time allotted, and the search report then only contains A documents. When it comes to examination, a direct grant is possible in most of the cases. The problem comes later when the opponent turns up with the relevant document(s), but in the meantime the production target was achieved.
3/ As far as the age of patent documents is concerned, I would like to remind that the PCT minimum documentation goes back to 1920. It follows that a document filed in 1968, published in 1971, cannot be considered as old. In mechanics, some older patents can still be highly relevant.
If the document cannot be found by a keyword search, or any other search, due to the fact that terminology has evolved, then there is something wrong with the search algorithm. It is clear that some fields of technique become obsolete as technology evolves.
In the early 80s, as substantive examiner, I had to deal with magnetic bubble memories. There was a big rush on those as they were permanent and the information was not lost when the power supply was cut off, while electronic ones lost the information when the power supply was cut off. Who has still an idea what those bubble memories could be? Magnetic bubble memories ended in the part of the iceberg below water, but they induced further searches in the field of electronic memories.
Conclusion:
My comment on the failed search is not to be seen as a criticism of my former colleagues. They do what they can in the time allotted to them. It is more the system behind it which is at stake.
When the opponent comes up with documents from the same applicant, and even from the same inventor, then something has manifestly gone wrong. The same is when the opponent comes with prior art under Art 54(3). And this is not so infrequent.
A bad search can have a deleterious effect. When a patent is revoked for a large company is it is not as dramatic, as for a SME. Big players have large patent portfolios. If the patent of a SME is revoked, it can be much more dramatic, as in general they have much smaller patent portfolios. If then, the SME goes to the UPC and claim infringement it can become very dramatic if the patent is revoked with a counterclaim for nullity. They have lost the patent and are burdened with heavy costs. In the past, before a national jurisdiction, the damages were not so high.
I might have been long, but I hope I could shed some light on the situation.
Thank you for your detailed answer.
If EPO examiners indeed no longer use classification when searching, it is nothing short of incredible. In my initial training, I learned that classification searches are a highly powerful complement to keyword searches. In fact, the entire point of having a classification is to enable later searches!
In my experience, many EPO search reports still turn up relevant prior art documents, which necessitate claim amendments or even giving up on the patent application. The system still works, or you could say it isn’t completely broken yet.
I have not personally seen opposition cases where the opponent found Art. 54(3) EPC prior art, or prior art from the same inventor or applicant. But if such cases are on the rise, that is a bad sign indeed.
Otherwise, I can only agree with your conclusion.
@ Extraneous Attorney,
I do agree with you, that, overall, EPO searches are still better than searches carried out by other offices, but in my opinion, they are not as good as it used to be. Quality is not increasing, in spite of all messages to the contrary from the EPO upper management. This is mainly due to the production pressure exerted on the examiners by the latter.
Be it only due to my family name ;-)), if I had not seen prior art under Art 54(3) cited by opponents, I would not have mentioned it. It is not so frequent, but sometimes the priority is not valid under G 2/98 and the ED has not spotted it. There as well Art 54(3) prior art can play a role next to “intermediate” prior art under Art 54(2).
It appears that you are not mentioning one important aspect. I always understood (but correct me if I am wrong) that EPO examiners are supposed to be technically skilled in their field.
True technical skill of the examiner in the particular field should be a safeguard against several of the mentioned reasons of a poor search. In such a case, the examiner will know where to search in order to have the best chance of finding relevant documents, and will understand that something must have gone wrong in the search process if the selected document do not come close enough.
As I already mentioned, I have never been an examiner and therefore I do not know what truly happens within the EPO, but I have strong doubts regarding your points.
First, as far as I know, it is very much the case that EPO examiners are technically skilled. Strong technical qualifications are absolutely required to be even considered for the job; and the EPO’s compensation package is so attractive (compared to other technical jobs available in Europe) that the hiring process is bound to be highly competitive. Besides, I cannot recall a single instance where I suspected that an EPO examiner lacked technical skill. Sometimes an examiner will raise an unfounded N or IS objection, but that is a quite different thing which happens in every patent office, and there are many possible reasons why this can happen. In fact, I’m quite sure that some examiners raise such objections mainly to get the applicant to go on the record and explain why the invention is patentable.
Second, I cannot agree that technical skill should prevent poor searches. Technical understanding and searching are different skills; otherwise no patent office would need to train its new examiners in searching. Besides, unless the patent application is obviously directed to something lacking novelty, like a patent application claiming (the generic concept of) the wheel, it’s practically impossible to predict a priori whether relevant prior art exists. When I had far less experience as a patent attorney, I sometimes tried to make such predictions, and I got every one of them wrong. One invention, which I thought bordered on the trivial, was deemed patentable by multiple patent offices; another, which I thought was an elegant and technically sophisticated solution, was anticipated by a patent application published before 1945. One can only tell in retrospect, after the prior art is found.
@ BJ and Extraneous Attorney,
I think that both have a point, but it is not the whole story.
In order to be recruited at the EPO you needed to have a diploma allowing you to be employed at the national patent office of your home country. For British candidates it meant a BSc (or this was the case when I was involved in recruitment). For most of the other countries it meant a MSc or an equivalent diploma, e.g. an engineering degree at university level. A PhD was never required.
That in biotech you cannot employ a mechanical engineer or vice versa goes without saying, the same applies to electricity/physics.
One aspect should not be underestimated. Some national patent offices require 5 years of professional experience before you can seek employment as examiner at the NPO, e.g. Germany. This condition is not applicable at the EPO, you could be recruited as examiner straight at the end of your studies.
From the opening of the EPO, board members were only appointed for a period of five years, but reappointment was possible. This rule of renewable 5 years appointments has then been transferred to other posts within the EPO. This was then first applied to posts of Principal Director, then to Director posts, and apparently now examiners.
For young graduates, the EPO has devised a Young Professionals Programme lasting three years. It is unclear for an out stander to know in which way it mixes with other possibilities of employment at the EPO as examiner. Whether those people are then becoming examiners is not known.
From information received, it seems that nowadays a permanent contract is only available for an examiner after the completion of two 5 year periods as examiner. This limits somehow the provenience of examiners. A person with good professional experience will hesitate to come to the EPO, especially with a family and children, and at the same time sever all links with national social security or pension schemes.
This does not mean that graduates are unable to perform well as examiners, but I know from experience that in certain technical areas, having some professional experience is of great help. Also for people with experience in industry or research, the grass is not greener on the other side, and they know well what they have when they join the EPO. This cannot necessarily be said from examiners coming straight from university.
On can therefore assume that skills are there, but having skills does indeed not prevent poor searches. Technical understanding and searching are indeed different skills.
What is sure, is that examiners really do what they can, but the production objectives are ever increasing. In my opinion, mishaps are mainly due to the high efficiency and production objectives required from examiners. As examiners do not have any longer dedicated offices, the coffee-machine chats are something from the past, and yet those had a great, but not measurable, impact on knowledge dissemination. When holding OP, examiners, and very often board members, are not sitting together. This is also detrimental to the work. Computer aided tools are at the disposal of examiners, but what is often missing is time.
I can only imagine what the board would have done if an opposition division had behaved in the same way. It is curious what constitutes “exceptional circumstances”. Circumstances seem to be “exceptional” only when it saves the board work. Decisions such as this erode the EPO’s reputation.
@ Anonymous,
There is nothing to add to your conclusion.
If the board wanted to “punish” the proprietor for not turning up at the OP, it would have been better not to admit AR8. There was no necessity to discuss the novelty of AR8. This induces a dangerous precedent, in spite of the fact that the proprietor had not ggiven reasons as to why AR8 was allowable.
It is the same when a board alleges common general knowledge without presenting any evidence, and stating that the personal knowledge of one its members is enough as evidence, see T 1092/12, T 1370/15 or T 2526/19. There as well, an ED or an OD would be faced with the reproach of having committed a substantial procedural violation.
That procedural case law of the boards can still be seen as being a lottery, is also not helping the board’s reputation.
Prompted by your reference to T2526/19, Daniel, I called it up and read its para 3.7. If I understand it right, the Board reversed to OD, relying for its reversal on nothing more than the “view” of the Board members (opposite to that of the OD members) that what was claimed “would have been immediately apparent” to the relevant skilled person. Nothing more than that.
What is it, that brings about such results? The pressure to meet management productivity targets, perhaps? Driving down the stats on pendency of cases, perhaps? Are they afraid of losing their jobs, perhaps?
Or are they doing the bidding of the litigation community, which demands that the speed of getting to a result, any result, shall outweigh any other consideration?
Of course, it is different if the first instance gives a preliminary view about the level of common general knowledge for then the parties are then bidden to respond, with proper evidence and the EPO can then decide on the evidence, rather than gut feeling.
Dear Max Drei,
You have put the finger on the problem.
Board members are only reappointed if they show a certain “performance”. It happens that the performance criteria to be applied have never been made public. Why not?
The chair of the boards, never misses an opportunity to claim that the production of the boards has increased steadily and the pendency time is going down. The board will however never reach the pendency times of the UPC for the simple reason of the time periods/limits built in EPO’s procedures.
When one sees how procedural case law is going in all possible directions, one wonders what the chair of the board and the presidium of the boards are actually doing. A computer can tot up the decisions.
One should also not forget that Art 1(2) RPBA does nothing more than defining for each board member its production target for the year to come. Another aspect which should also not be forgotten, is that the backlog of the boards increased for many years as the then president of the EPO refused to present members and chairpersons to the approval of the Administrative Council.
I do not think that the boards are bidding to the litigation community and speed is not their prime concern. Up to now, the vast majority of decisions are well reasoned and convincing, with some exceptions like T 2526/19.
If there is one important aspect, it is that any deciding body, first instance and certainly not a BA, should let itself go and decide according its gut feeling.