EP 3 275 932 B1 relates to a polyethylene for injection moulding and a moulded article, like an injection-moulded fuel tank.
Brief outline of the case
The patent was maintained according to a new MR filed during the OP before the OD.
The opponent appealed. Eventually the patent was revoked for lack of IS.
The case is interesting for the considerations relating the choice of the CPA=D4.
D4 relates to a polyethylene-based resin for an injection-moulded plastic fuel tank which exhibits an excellent balance of injection mouldability, durability, and impact resistance.
The contradictory points of view
The opponent, submitted that the composition described with example 1 of D4 constitutes the CPA.
For the proprietor, the CPA should be selected from the general teaching of D4, as specified in the claims of that document. Selecting a particular example may rely on impermissible use of hindsight.
The proprietor argued in addition that the skilled person, if starting from an example of D4, would rather start from example 2 which concerns a composition having better injection mouldability and higher orientation ratio, the latter properties being also sought to be achieved in D4. In view of example 2, example 1 would not be seen as the most promising springboard.
The board’s decision
The importance of the purpose or objective indicated in the patent/application
It is established case law that ideally the CPA should be a document that mentions the purpose or objective indicated in the patent in suit as a goal worth achieving and having the most relevant technical features in common, i.e. requiring the minimum of structural modifications. The board referred to the CLBA, 10th edition 2022, I.D.3.1.
The board did not agree that in order to avoid hindsight an inventive step analysis should necessarily start from the general teaching of a document as e.g. specified in the claims. This is case dependent.
A preferred embodiment such as a specific example of a patent document which unambiguously describes a combination of concrete and promising measures toward the achievement of the goals set out in the patent under examination constitutes a realistic approach of the skilled person.
Definition of “the most promising springboard towards the invention”
The concept of “the most promising springboard towards the invention” expresses the necessity to assess inventive step from such point, as any non-obviousness finding vis- -vis a prior art, “if not closest to the invention, would be irrelevant and inconclusive to validity without the assessment of the inventive step in respect of the objectively closest state, i.e. the most promising springboard towards the invention which was, available to the skilled person, cf. T 164/83, “Antihistamines” OJ EPO 1987, 149)” and T 254/86, Reasons 15.
It does not mean, however, that a non-obviousness finding vis-à-vis that most promising springboard necessarily means that another route starting from a more remote starting point is non-obvious as well. This depends on the circumstances of the case under consideration.
It is nevertheless required for an objective assessment of inventive step that a more remote starting point for the skilled person is also realistic.
In this respect, the board referred to T 405/14 cited in section I.D.3.1 of the CLBA, in which it was held in essence that there is no requirement that the “closest prior art” should be unique and that it should be selected based on a metric defining the distance between items of prior art and the invention, cf. T 405/14, Reasons 18. As pointed out in T 405/14 “the jurisprudence does not define any such metric beyond indicating what criteria might be considered relevant to it, common features, similar purpose, …, there are frequent situations in which the identification of a unique closest or best starting point is not straightforward or even possible.”
The board adhered to the view expressed in T 405/14, at least insofar the selection of a single starting point as a matter of principle would not correspond to the real situation encountered by the notional skilled person.
On the contrary, an inventor seeking to achieve a given goal would rather find natural to maximize their chances of success by trying more than one promising and realistic starting point.
For the board, the assessment of inventive step should thus start from a situation as close as possible in reality to that encountered by that notional skilled person.
Application to the present case
According to paragraph [0012] of the patent, an object of the invention is to provide a material for a moulded article, which material exhibits excellent durability and impact resistance of the resulting moulded article, particularly a polyethylene-based resin material for a fuel tank while maintaining good injection mouldability.
This is the case for example 1 of D4, which unambiguously describes a material having good injection mouldability, impact resistance, Charpy impact strength, and durability and whose use as a material for a fuel tank is in addition evaluated in example 4 and reference example 1 of D4.
Starting from the general disclosure of D4, i.e. its claims 1 and 2, as proposed by the proprietor, which claims comprise functional features and/or a more general definition for both the High Molecular Weight (HMW) component (a) and Low Molecular Weight (LMW) component (b) might therefore necessitate more experimentation to put into practice the teaching of D4 on which the skilled person needs to build on for a further development toward the achievement of the goals set out in the patent in suit.
Accordingly, it cannot be held that starting from the teaching of claims 1 and 2 of D4 would be considered by the skilled person to be more appropriate.
Any of the examples 1 to 3 are all representative of the teaching of D4 and address the problems set out in paragraph [0012] of the patent in suit.
Comments
Choice of the CPA
The CPA should primarily be a document that mentions the purpose or objective indicated in the patent in suit as a goal worth achieving and having the most relevant technical features in common.
This does however not mean that the CPA is the document having the most technical features in common!
Starting from this document, then the most promising springboard towards the claimed invention can be defined. The only requirement is that it should be realistic. Depending on the case, it is not necessarily the general teaching disclosed, but can well be a specific example.
Only defining the CPA as the most promising springboard towards the claimed invention might be opened to the criticism that is defined with hindsight.
If the CPA does not deal with the same/similar purpose or objective, it cannot represent “the most promising springboard towards the invention”.
On the procedure
D4, which eventually lead to the revocation was patent document 2 cited in the patent in suit as background art.
It was cited neither in the ISR/JP nor in the SESR.
Comments
25 replies on “T 1888/21 – On the choice of the CPA – Same purpose or objective – Most promising springboard”
This case gives me an opportunity to write (again) what I like about the EPO’s problem-solution approach, namely, its efficiency, its management of the hindsight problem, its objectivity and the way it balances the interests of the inventor and the public.
First, it is efficient enough to be used as a filter in all of the hundreds of thousands of patent applications, without clogging up the Patent Office. Second, it recognises that the drafter has hindsight knowledge of the problem and the solution when drafting the patent application, whereby fairness between inventor and public demands “equality of arms” between the party arguing that there was an inventive step and the party arguing that there was none. Third, it bases the decision on a hypothetical (ie objective) person skilled in the art. Sworn statements from so-called experts in the field don’t help much, in deciding what that imaginary person would or would not have found obvious. Finally, despite all that imaginary stuff, EPO P-S, with its “realistic springboard” case law, is intensely concentrated (as here) on what was the “real life” situation in the field, and what real people do, so that between the opposed parties, the one that argues closest to real life has the better chance of winning the debate. Hooray!
Dear Max Drei,
I can only but support you in your appreciation of the problem-solution approach of the EPO when it comes to assess inventive step.
It should however not be forgotten that the PSA is a construction of case law and is not to be found in the EPC. It is a mere tool, but a good one.
Being a tool “not invented here”, allows some people, especially in national jurisdictions, to look down on it.
It will be interesting to see whether the UPC will adopt it or not.
On the other hand, the PSA has proven its efficiency. There is barely no decision of a board which does not apply the PSA.
The advantage is that it can be applied in all areas of technique, although it originates from the chemical boards of appeal. Mechanic boards of appeal were the last to be convince of its utility.
As any tool, it has to be applied correctly, and sometimes one wonders what some people understand under PSA, even a few years having sat the EQE! I will leave it at this….
Yes, Daniel, in my experience, the problem is that so many patent attorneys in senior positions think they know what the problem-solution approach is, and how to apply it, without any need to take the time to learn it properly. Rather, they rely on a lifetime of arguing about problems and solutions, and their intuition, to argue EPO PSA against an EPO Examiner or the other side in an EPO opposition.
Your comment though seems to suggest that the problem is not going to die away as the old guys go into retirement.
Perhaps the educationalists at the EPO, travelling the world to explain EPO-PSA, should try to include in their presentations material which reveals to the audience the dangers of relying on intuition to win obviousness debates at the EPO?
Dear Max Drei,
My experience on the sub-optimal application of the PSA is not only based on my active time, but also on my consulting jobs and training duties.
The sub-optimal application of the PSA starts with the wrong choice of the CPA, as it is always tempting to chose a “CPA” which is further away from the claimed invention, or which has the most features in common, so that IS seems easier to prove or deny. I have also seen a CPA composed of a mosaic of documents (also from examiners).
The second problem is an incomplete assessment of what is claimed and/or disclosed in the “CPA”. Starting on this incorrect basis, the difference between the CPA and the claimed subject matter is not properly appreciated, and hence the effect of the difference is not the right one. It ends then with a wrong objective technical problem (OTP). I have also come across OTPs containing a pointer to the solution!
When it comes to the actual assessment of IS, often the opponent claims happily that the skilled person is super intelligent and would therefore combine any documents passing by. The proprietor often claims, with the same conviction, that the skilled person is as thick as two planks, and would never put two and two together.
There are various places were things can go wrong when applying the PSA, but in most of the cases when it is not correctly applied it simply shows a profound misunderstanding of the PSA and of its mechanisms. Often its application is more a matter of chance than planned action.
There are good ways to learn how to correctly apply the PSA. Either look at decisions of the TBAs, or to delve in past EQE papers. Decisions of the TBAs can be rather complicated. Past EQE papers have the advantage to be technically simple and well understandable in all technical areas. Former B papers show how to argue in favour of IS and former C papers show how to argue against IS.
Self-praise is never good, but I have devised a few examples on how to correctly apply the PSA. They are mainly based on simple decisions of the TBAs.
One example, based on a real case, is to have two different CPA. It happens, even in real life. It is then easy to show that the result can be the same, lack of IS, but the argumentation has necessarily to be quite different. In the patent world, arguing D1+D2 is necessarily different from arguing D2+D1. In general, at this moment, the penny drops, even with experienced representatives in need of a refresher. I then add that in the patent world, addition is not commutative as in the real world. Good to bring in a shin dig! ;-))
My experience is with US patent attorney clients. For many of them, the concept of the OTP is incomprehensible, totally alien, and always, inherently, hopelessly contaminated with illegitimate hindsight.
And so, as the identification of the OTP is decisive to the outcome, they can’t even get past “first base” in their grasp of the obviousness debate at the EPO.
They used to think they could instruct me how to reply to EPO Examiners on Art 56 EPC. These days, because they still can’t understand the EPO’s analysis of obviousness, they don’t even try.
Keep going, with your training activities. There is much still to do.
The overarching objective of the PSA to minimise hindsight is highly laudable. It is based on the welcome recognition by teh EPO that hindsight is inevitable anyway since the search of the prior art has to rely on the foreknowledge of the invention. But the PSA adds to the language of Art 56 methodology steps, the CPA and the « objective technical problem » (« OTP »), which are ill-defined and create additional risks of hindsight-guided reasoning of their own.
Two examples : the reference to the « most promising candidate » (a term I personally find unfortunate) can only be based on hindsight. And how the choice of a CPA be based on the largest number of common features with the claimed subject matter if not with the benefit of hindsight ?
Generally speaking, I think the artefacts introduced by the PSA tend to draw too much attention (this is conspicuous in the BOA decisions) to the detriment of the conditions of Art 56, the person skilled in the art and the assessment of obviousness (« could vs would »).
And the PSA is taken at the EPO as dogma, not as a tool which indeed is valuable but has validity limits like any tool. For example, there are decisions in which « problem inventions » are assessed in accordance with the PSA protocol, even though the PSA is clearly inapplicable to such inventions. Or the need to define an OTP may lead to define the problem as the search for an alternative to the prior art : a purely negative definition, devoid of technical content, which makes it simply impossible to apply the « could vs would » reasoning. This is what I could call a box-ticking approach. All boxes must be ticked off even if the answer is meaningless.
The UPC revocation decision Sanofi vs Amgen of the Munich Central Division, issued on July 16, will certainly bring much food for thought as to the assessment of IS. My first takes are that the UPC court did not follow the PSA. It felt no need to select a CPA. A very significant aspect is the discussion of a motivation for the skilled person. The US Supreme Court in KSR rejected the requirement for the skilled person to find an explicit motivation in the prior art. Needless to say, we need to wait for the decision of the UPC Appeal court. We have already a case in which the local court’s decision has been overturned by the Appeal court.
On the broader topic of a US/EPO comparison, it is also of note that the US Supreme Court held the Amgen US equivalent patent invalid for lack of enablement across the entire scope of the claim. US case law is very stringent on this issue, much more than the EPO case law. This is enshrined in article 112(f) of US patent law, which restricts the scope of claims to embodiments disclosed in the description and their equivalents. So that the US Supreme Court and the UPC Munich court have reached the same conclusion but on the basis of different grounds.
As Francis Hagel observes, EPO-PSA sets up a framework for the obviousness debate. I am yet to be persuaded that this framework is defective in any way, or that dispensing with it leads to higher quality decisions on obviousness.
The risk of ex post facto analysis is ever-present in the obviousness debate. But, as i see it, EPO-PSA inceases the likelihood that during the debate it will be called out for what it is, and set aside in the further analysis. I’m thinking in particular of the call-out “That OTP is erroneous because it includes something of the solution”.
It is all very well saying, as in KSR, that motivation is not the issue but then, what is? One runs the risk of losing objectivity, with judges instead subjectively declaring something is obvious because I say it is obvious. Adversarial English (and US) law relies on evidence acquired under cross-examination. All fine and good, for high quality decision-making, but not viable for examination of patent applications ex Parte. And always devilishly expensive too. Perhaps that’s the point: it encourages parties to settle rather than go to trial. Make EPO-PSA the standard in the courts and everybody will be petitioning the judges.
Except that, if the courts adopted EPO-PSA, everybody would know in advance what will be their answer and so will settle rather than litigate all the way to full trial. Is there anybody who wants every patent dispute to go all the way to full trial anyway? Perhaps only the community of litigation lawyers and their largest international corporate clients with their bottomless reserves of cash to exhaust the resources of the opposing party long before arrival of the case at the highest court of appeal?
Dear Max Drei,
I cannot but support you views. What judges, as mainly legally qualified persons, dislike in the PSA is its determining traits.
When you start arguing and putting every word in balance, it is difficult to accept the systematic approach of the PSA. It was “not invented here”, but by technical members of the boards of appeal of the EPO. How outrageous, and hence not our cup of tea!
This is certainly encouraged by “the community of litigation lawyers and their largest international corporate clients with their bottomless reserves of cash”. That’s why this “community of litigation lawyers and their largest international corporate clients” have been so keen on the UPC.
A late well known French legal practitioner, specialised in patent litigation, Me Mathely, said the following about assessing IS:
“Assessing non-obviousness does not imply any judgement of value, be it about the creative effort of the inventor, or be it about the importance or the usefulness of the invention.
The judge cannot simply assert obviousness by merely saying what he is convinced of.
He must formulate the technical reasons which, by comparison of the invention with the prior art, lead to an objective finding of obviousness.”
This statement is much older than the introduction of the PSA, but you can see the PSA underneath it.
Me Mathely is also known for having said: “the proprietor discovers his patent on the day of the infringement”. This tells a lot!
Cf. Art 69(1) EPC and Art 1 of the Protocol on interpretation.
For the surplus, please have a look at my reply to Mr Hagel.
@ Mr. Hagel.
The PSA is nothing more than a tool, but in my opinion it is a useful tool. I have never used it as an examiner, as it did not exist at the time, but later when sitting in oppositions in my directorate.
I can agree that the choice of the CPA can be considered as a potential drawback of the PSA. My agreement stops there.
It is however extremely rare that an inventor starts afresh and there is no prior art available. I know one example: copying documents by creating a latent image on a selenium drum. In the 70ties this was known under Xerography and for a long time Xerox had a monopoly on this type of copying. Nowadays it is standard. It existed before various ways of copying, one of them being a real photographic picture.
Normally, the inventor starts with prior art known to him. This is the reason why R 42(1,a) prescribes to indicate at the beginning of the description he background art known to the applicant, as it is useful to understand the invention, to draw up the European search report and to examine the application.
An invention is normally the technical solution to a technical problem. In other words the inventor had a problem he wanted to solve starting with the prior art known to him. For the sake of simplification this problem can be classified as the subjective technical problem (STP).
During the search, the EPO might find other prior art which can be considered coming closer to that the inventor started with.
Case law of the boards has devised a set of rules to decide whether a document is the CPA.
Applied correctly, there is no trace of hindsight.
According to case law and of the Guidelines the following criteria have been developed by case law in order to select the CPA:
– same purpose
– same effect
– similarity of the technical problem
– most promising springboard towards the invention
– most promising starting point
– or easiest route to arrive at the claimed invention needs a minimum of changes
I consider that the last three criteria should not been taken on their own, as it could indeed lead to hindsight. They can however best be used as a confirmation of the three first ones.
The OTP is the result of the effect of the differences of the claimed invention with the CPA.
The piece of prior art having the most features in common or the least differences is rarely the CPA. This is only the case if it belongs to the same technical field as that of the claimed invention.
When you look at case law, the OTP becomes an alternative if the claimed invention and the CPA have the same effect. Where is there hindsight in this?
The “problem invention” is a result of T 2/83, OJ EPO 1984, 265.
T 119/82, T 119/82 and T 764/12 come indeed to the conclusion of the presence of a problem invention.
At this time the PSA was not even in the limbs.
In T 1236/03, the board applied the PSA and nevertheless concluded to the presence of a problem invention. T 1001/18 explained why, even when applying the PSA, it can be concluded to the presence of a problem invention. Those two decisions contradict your sweeping statement on the inapplicability of the PSA to problem inventions.
Lots of applicants, or better their representatives, have tried to come up with problem inventions.
They have rarely succeeded. See for example T 971/92, T 252/10, T 703/18, T 2411/18, T 1112/19 and T 437/21.
I am thus also obliged to contradict your opinion that, at the EPO the PSA is taken as dogma. It is a tool which indeed is valuable but has validity limits like any tool. It is certainly more than a box-ticking approach. Very often the strict approach with its clearly set out steps is simply ignored. This is the reason why some of the argumentations I have seen fail to convince. But this is not the fault of the PSA.
All steps=boxes must be ticked off and the result is in the overwhelming majority of the cases anything but meaningless.
I am open to receive any proof to the contrary, but I am not prepared to accept such a succession of sweeping statements.
As far as the UPC decision is concerned, nothing much can be said as it is only a first instance decision. We will have to wait until the Court of Appeal has issued a series of convergent decisions. Anything else is pure speculation.
That legal requirements are different between the US and the EPO is nothing new under the sun.
In T 1097/22, a US applicant argued that, since examiners in 18 countries were able to derive the claimed subject-matter from the original disclosure, it seemed reasonable that the same would be the case under the EPC. You can guess what the board’s reply was!.
“This is enshrined in article 112(f) of US patent law, which restricts the scope of claims to embodiments disclosed in the description and their equivalents”
This is not correct. This applies only to claims which invoke the means-plus-function interpretation. Looking at the statute:-
“(f) ELEMENT IN CLAIM FOR A COMBINATION.—An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.”
This only applies to claims that are without the recital of structure, material, or acts in support thereof.
@ Correction,
Thanks for the clarification intended for Mr Hagel.
In any case I consider it not really relevant or helpful to compare decisions reached in quite different legal systems like that of the US and in Europe, and especially at the EPO or now at the UPC.
The procedures and the legal provisions are quite different.
Whilst claims of the type “means + function” are rather frequent in Europe for device claims, they are not liked much in the US. The same goes for Jepson claims, which are preferred in Europe, but avoided in the US. think also of the notion of file wrapper estoppel, unknown in Europe, not to mention the grace period.
In view of the prior art which can be used on both sides of the Atlantic, any direct reference to one or the other system is therefore rarely conclusive.
Last but not least, one of the biggest differences lies the definition of added subject-matter between the EPO and the US.
Agreed, Daniel, but often it is interesting how different jurisdictions use different legal provisions to get to an end result (patentee wins or loses) which is in substance the same. All First to File jurisdictions revoke old or obvious subject matter. What the EPC sees as “added matter” the USA sees as “lack of written description” in the application as filed. If you want a patent that succeeds, you know what to do, and it’s pretty much the same in every jurisdiction.
That said, i recall a case in Europe, in which the patent owner was a company with a 5 letter name starting with “H”, where the patent owner lost (on various grounds) in every jurisdiction except its home jurisdiction?. One more argument in favour of the UPC, perhaps?
Dear Max Drei,
That the result can be the same in different jurisdictions in spite of differences in legal provisions is not at stake. In view of the difference in legal provisions it is however not very useful to compare the different decisions. And it can certainly not be claimed that one is better than the other. What does this really bring? Judicial independence is a very valuable good. Comparing the ways to the decision can nevertheless be useful as it can show different ways of thinking.
If an applicant/proprietor wants to be successful in different fora, he should know what to do and act accordingly.
That some jurisdictions are not free from governmental influence is a matter of fact. It is thus not such a problem when a patent owner selectively wins in his home jurisdiction. Where it becomes problematic is when foreign patent owners lose in this same jurisdiction. This is however far of convincing me of the necessity of the UPC!. But this is nothing new to the readers of this blog.
Discussing the application of the PSA from the angle of BOA decisions and the comparison between the EPO and national courts & UPC is interesting but tends to be overly focused on inter partes proceedings and to miss the major issue of its application to ex parte proceedings by EDs. MaxDrei’s first remark in his initial comment was indeed insightful.
The PSA has been implemented at the EPO primarily for use in the examination of applications, which requires the review by thousands of examiners of hundreds of thousands of applications filed every year (ca 200,000 in 2023), under the very tight time constraints imposed by production objectives. Needless to say, this is very far from the context of opposition proceedings, leave alone court proceedings.
Thus, when we agree that the PSA is a tool which has its limits e.g. linked to the nature of the invention, we must not lose sight of its context of use – ex parte, huge volumes, time constraints, applicants’ limits on resources vs. inter partes, smaller volumes, resources generously mobilised by the parties, given the commercial stakes in the patent, and the EPO. This is why restricting the analysis to the comparison between the EPO case law about the PSA and the case law of national courts and UPC is somehow misleading.
The referral decision T 439/22 (G 1/24) related to the application of Art 69 provides a glaring illustration of such a narrow angle which utterly disregards the context of examination by EDs. The decision is to a very large extent focused on an analysis of the case law of European courts on claim interpretation under Art 69, but is astonishing in that it is entirely silent on the settled rule of interpretation stated in section I.C.4.1 of the case law of the BOAs (« claims must be given their broadest sensible meaning ») which is simple and general as needed to provide clear guidance to the EDs. The only recommendation of the decision regarding examination by EDs (Reason 6.2.5) is simply unrealistic under the existing time constraints and would open up unwelcome uncertainties and additional costs.
Turning to the practical implementation of the PSA by EDs, I had the opportunity of an unusual angle to review this issue by participating in the audits of granted patents conducted every year as an external assessor within the SACEPO/SQAP panels. Each of these panels, assigned to specific technology fields, include two external assessors (industry and private practitioners) and two EPO experienced examiners or team leaders. Each panel audits 6 files of granted patents and delivers reports as to quality issues. The findings are shared by the entire panel more often than not.
As to the PSA and more broadly the assessment of IS, the panels are generally positive as to quality, but signal the following quality issues as not infrequent :
– Lack of explanation of the ED as to why they feel a need to substitute for the technical problem as stated in the original application a different « objective technical problem ».
– Lack of reference to specific parts of a prior art document relied upon by the ED in their assessment of IS
– Lack of support documentation for the CGK of the skilled person when the ED combines the CGK with a prior art document to justify a finding of obviousness.
Francis, as to your not infrequent “quality issues” I wonder whether they are fair to EPO ED’s or even rise to the level of a quality issue. I don’t see how it can be a lapse of “quality” for an EPO Examiner to set out a plausible obviousness attack which suffers from one or more of the three quality issues you cite. All that’s going on is an Examiner putting it to Applicant that patentability is not “a given”. Basically, the Examiner is inviting Applicant to set out why the claim should be allowed. Applicants should be pleased to receive such invitations, and only too happy to reply with persuasive arguments and evidence that convincingly assuages all the Examiner’s doubts, thereby leading to a more robust grant than if the ED had issued a 71(3) as their first Communication.
Or not? I’m still not seeing a fair criticism of EPO-PSA.
Dear Max Drei,
The PSA is certainly open to critics, but then one waits for a proposal to replace it. The fact that when assessing IS documents have to be combined is not a specificity of the PSA.
If the PSA is correctly applied, it is very efficient and allows rapidly to come to a result which can be defended.
That some categories of users of the EPO system have their difficulties with it is certain, but then it has to be explained over and over in the hope that at what moment the penny drops.
Mr Hagel,
Contrary to your allegation, the PSA has not been implemented at the EPO primarily for use in the examination of applications. It was simply considered at the time as a tool allowing a more systematic approach when assessing IS. The PSA finds its origin of case law of the boards.
It is manifest that EDs have to deal with a large number of applications a year, and are under a corresponding pressure. It can be agreed with you that if the double check suggested in Reasons 6.2.5 of the referring decision for G 1/24 (T 439/22), would inordinately increase the workload of EDs and ODs. However, it would also increase the workload of the boards. My agreement stops there.
I do however fail to see the link with the PSA and its application as such.
That some ED do not apply the PSA correctly is not a surprise. Taking very general conclusions from a sample of 6 files par panel is also quite daring as it cannot be considered as statistically significant.
Lack of explanation of EDs as to why they feel a need to substitute for the technical problem as stated in the original application a different OTP might be correct, but this does not put the usefulness of the PSA in question. This appears more a question of training than a fundamental flaw of the PSA.
Lack of reference to specific parts of a prior art document relied upon by the ED in their assessment of IS might also be correct, but this does not put the usefulness of the PSA in question. This appears, again, more a question of training than a fundamental flaw of the PSA.
Common general knowledge is in principle to be found in standard books and reference works, cf. T 1641/11, T 263/12, and not in patent documents, with the exception of very new techniques, e.g. T 1117/14.
Patent literature is not representative of common general knowledge, cf. T 1819/14, T 263/12 or T 1540/14. This also applies to specialized journals, cf. T 1997/11, or scientific papers, cf. T 1651/13. Also a doctoral thesis is not representative of common general knowledge, cf. T 1832/14.
Unless standard books and reference works are not fully digitised, it will remain difficult for SDs and EDs to easily consult them as the members of first instance division are mainly working in home office. In my active time standard books and reference works were in libraries, be it general libraries or specialised directorate libraries.
The failure of a correct reference to common general knowledge is thus foremost an organisational problem of making available the information contained in standard books and reference works to SDs and EDs, especially in view of the now established working pattern at the EPO. The failure of some properly documented references to common general knowledge as noted by the panels might be correct, but this does, again, not put the usefulness of the PSA in question.
I have been looking at far more files (a few thousands) than the files looked at by the panels and it is apparent that regularly common general knowledge is brought by the opponent.
In quasi none of the ESR/ISR established by the EPO or other searching authorities, I have found a documented reference to common general knowledge. Search reports contain mainly patent literature and occasionally, in search reports established by the EPO, excerpts of specialised journals or scientific papers.
All the objections you raise are not such that the PSA should not be used at the EPO. There are other ways to assess IS, but the PSA is very powerful tool, provided it is correctly applied.
My last comment was not intended to challenge the usefulness of the PSA but to draw attention to quality issues related to its implementation, such as sweeping statements supported by no documentation regarding the content of the CGK, or reliance on a prior art document without reference to specific parts of the document.
I wish to say that the findings of the SACEPO/SQAP panels are globally positive, such issues emerge in a minority of the audited files. It is important though to spot concrete issues on the basis of detailed audits of granted patent files, and to acknowledge their credibility, owing to the fact that audits are performed by panels including EPO experienced examiners and external auditors and most often there has been consensus of the panels as to the findings.
Also, such issues relate to quality in terms of the thoroughness of the assessment by EDs. This is a notion of quality which is to be distinguished from a commonly used definition of quality based on the statistical analysis of the outcome of opposition proceedings (which relates to a small minority of the granted patents and to the different context of inter partes proceedings). An assessment leading to the grant of the patent may actually be correct even if the reasoning supporting the assessment lacks the desirable specificity.
Mr Hagel,
I have taken note that you’re not criticising the PSA per se, but you rather wanted to draw attention to quality issues related to its implementation. This was not entirely clear.
Up to now the EPO management has always been very reluctant to acknowledge any quality problems at the EPO. I am therefore curious to see how it will react to conclusions drawn from a rather small sample of cases, i.e. 6 cases par group.
The quality I have been assessing by looking at oppositions is based on an average of more than thousand decisions of the boards per year for the last five years.
How can it be that a study on the application of the PSA “which relates to a small minority of the granted patents” can be more significant than a study relating to a few thousands of patents, even taking into account “the different context of inter partes proceedings”.
What is noticeable when looking at the opposition cases, is that in less than 10% of the cases, opponents came up with document which could truly not be available in the EPO’s search files. The proportion of patents being revoked is over 50%, and the oppositions rejected is lower than 20%. If this is not a problem of quality, then any notion of quality can be forgotten. A linear extrapolation on all the granted patents, is not correct, but it nevertheless gives raise to concerns.
I suppose that your group took as prior art, the prior art revealed during the search carried out by the EPO. Your results might be correct on the basis of those documents, but if the right documents were not mentioned on the search report, the findings of your group could be questioned on a different level. .
It might be true that an assessment leading to the grant of a patent may actually be correct even if the reasoning supporting the assessment lacks the desirable specificity. However, if later this grant is questioned in view of the fact that the documents taken into account were not the right ones, then there is a bigger quality problem.
At the EQE you might give the correct answer to a question, but if your argumentation is not correct or incomplete, the chances are high that you get the ticket for the next year. Your comment is thus to be taken wich a rather large pinch of salt.
Mr Thomas,
The number of the patent files audited by the SQAP panels is decided by the EPO. But the selection of the files is decided by the panels, it is based on a large number of cases proposed by the EPO for each technology field and each topic studied by the panel.
You seem to be skeptical as to the interest of the findings of the panels. Let me give you an exemple. The panels looking at search reports and WO have noted a number of cases in which the ISR was prepared by an ISA other than the EPO and the ED paid no attention whatsoever to the lSR. In other words the ED started from scratch. This is a quality issue concerning the search.
Francis Hagel, you point to “a quality issue concerning the search” but if I understand the situation correctly, the quality issue lies with the ISA rather than at the EPO. If the ISR is rubbish well then the EPO has no alternative but to “start from scratch”, or not?
And if the search reports produced by other ISA’s (I’m looking at you, USPTO) are indeed sometimes of disappointing “quality”, might that be because these ISR’s are drawn up by a person oblivious to any relevance that the EPO’s problem-solution approach might have, to the issue whether the claimed subject matter is inventive?
Dear Max Drei,
Reality shows that in many of the cases I have been looking at in the last 5 years, neither the non-EPO ISR, nor any ISR/EPO or ESR, reveal the really relevant documents.
Even taking into account the peculiarities of opposition proceedings, how can it happen that opponents come with much better prior art, sometimes even prior art under Art 54(3) from the same applicant or inventor?
Mr Hagel,
The number of files selected by the SACEPO panels is relatively low, so that the statistical value of the findings can easily be queried. See my preceding comment.
I am not sceptical as to the interest of the findings of the panels. I am more sceptical as to how the results of the panels might be taken into account by EPO’s upper management.
It is difficult to see why, after an ISR, starting the European search from scratch can really be a quality issue.
When I look at decisions of the boards in oppositions, in over 90% of the cases, with or without a preceding ISR, opponents come up with documents which should have been found during the European search or the IS carried out by the EPO, as they were in the same IPC classes in which the search as apparently be carried out. This results are based on over a thousand of files/years for the last 5 years.
This is to me a real quality issue, even taking into account the particularities of oppositions proceedings.
The big problem is that, there is still a belief in certain circles at the EPO, that a good search is a search which reveals at least one X or better a series of X and/or Y documents.
When one or more X and/or Y documents are found, the search is then stopped. If later, the applicant amends the claims, some reasonable doubts about the completeness of the search appear justified. This seems to be particularly the case, when the original claims are limited by features from the description or the drawings.
If the really relevant documents are not found by the EPO, it becomes irrelevant whether the PSA has been correctly applied or not, as the end result cannot be the grant of a valid patent. This is also an important quality issue.
When barely 20% of granted patents survive an opposition without being revoked or maintained in a more limited form, this result appears significant, even taking into account the particularities of oppositions proceedings.
I see a quality issue when an ED pays no attention whatsoever to the search report of another ISA, and this opinion was shared by all members of the panels including the EPO examiners. This is a matter of principle. I do not see why the reports from the ISAs should be disregarded as a matter of police. The ISA Reports may cite relevant stuff and the comments should be reviewed by the EDs.
Regarding the use of the SQAP reports by the EPO management, scepticism is justified. What we can do ? share the findings with the patent community so as to trigger a debate on concrete issues.
Mr Hagel,
I would be careful with principles. Sticking to them can lead to catastrophes.
I suppose you are talking about ISR not established by the EPO. It might well happen that examiners do not look at the ISR , which might indeed cite relevant “stuff”. This should not be the policy.
I suppose that when a SESR mentions “No further documents revealed”, the examiner has indeed looked at the ISR, and at the IPER should it have been established.
However, to me it is worse when neither the ISR nor the SESR cite the really relevant documents, but the document is found by the opponent and when this document was in the IPC classes allegedly searched by the ISA and the EPO, not to say in the same IPC class as the patent.
The evaluation of IS might be correct on the basis of the documents in the ISR and SESR, but not when the really relevant document is only revealed during opposition.
As far as quality is concerned, the IPQC has shared its findings with the patent community so as to trigger a debate on concrete issues. This was no avail as the EPO denied any quality problems.