EP 2 516 330 B1 relates to titanium dioxide particles.
Brief outline of the case
The opposition was rejected and the opponent appealed.
The board held that claim 3 as granted lacked N over a PPU.
Claim 3 reads:
3. A particulate titanium dioxide having an E524 of less than 9 l/g/cm, an E360 of 25 to 50 l/g/cm, and an E 360/E308 ratio of 0.5 to 1.0.
Eventually the patent was revoked.
The opponent’s point of view on the PPU
The opponent argued in the notice of opposition and in the statement of grounds of appeal that it had purchased approximately 10 tonnes of a product containing various lot numbers, including Lot 0537002, from Kemira Pigments Oy before the patent in suit’s priority date. This product met the parameters specified in claim 1, thus meaning that this claim was not novel.
The opponent provided, as evidence, an order confirmation (D8) and a certificate of analysis (D9).
Furthermore, the opponent submitted in the notice of opposition that Kemira Pigments Oy had conducted a spectrophotometric analysis to demonstrate the properties of the sold product.
A letter from a former employee of Kemira (D12) concerning the circumstances of the alleged prior sale by their employer to the opponent. D12 showed a graph indicating the results of a spectrophotometric analysis of the sold product.
The proprietor’s point of view on the PPU
The proprietor did not contest the sale as such, neither in the opposition proceedings nor in the appeal proceedings. Concerning the circumstances of the sale, the proprietor argued that it had not been shown that the sale did not involve a confidentiality agreement. When setting up a pilot plant designed for producing hundreds of tonnes, 10 tonnes constituted a reasonable test batch, i.e., entirely for internal purposes.
The proprietor also contested that the parameters defined in claim 3 were properly measured. Neither the method nor the apparatus used to obtain the data contained in the graph of D12 was specified.
For the proprietor, the opponent did not prove the alleged fact “beyond reasonable doubt“. In other words, since there remained “a ‘grey area‘, then the case of anticipation based on such a document has to fail”.
The board’s decision on the PPU
For the board there was no need to give a final opinion on the questions of which standard of proof to apply, whether different standards of proof should be even applied in proceedings before the boards, depending on the context, and whether “proof beyond reasonable doubt” could ever be the right yardstick.
The decisive question is not whether the evidence could be more complete but, instead, whether – on the basis of the evidence as a whole – the deciding body is convinced by the alleged facts when applying the principle of free evaluation of evidence.
The board referred to T 2463/22, Reasons 4.5; T 1138/20, Reasons 1.2; T 733/23, Reasons 3.2; T 1634/17, Reasons 19; as well as to T 832/22, Reasons 3.2, for some more general observations on the standard of proof.
In the course of such evaluation the board has to take into account the probative value of the relevant pieces of evidence under the given circumstances and also in view of the fact that the certificate of analysis (D9) is only accessible to the opponent, the order confirmation (D8) is only accessible to the opponent and Kemira Pigments Oy, which is not party to the proceedings, and also in view of the fact that the spectrophotometric-analysis figure that was prepared by Kemira Pigments Oy.
As far as the alleged “confidentiality agreement” is concerned, the board held that, in the absence of any indications to the contrary, a secrecy agreement cannot be presumed to be included in the sale of this large amount, 10 metric tonnes, of titanium dioxide particles. There is no indication that Lot 0537002 was intended for the opponent’s internal experimentation or for trial purposes only, including pilot plants, under conditions of implicit confidentiality.
In particular, there is no evidence that the opponent operates a plant producing hundreds of tonnes of this or similar products even if – and contrary to the proprietor’s viewpoint, 10 metric tonnes is not a reasonable test batch.
Analysing in detail the parameters, the board came to the conclusion that Lot 0537002 has extinction coefficients E524 of about 8, E360 of about 32 and E308 of about 50, which result in E360/E308 of about 0.65, thus falling within the range of claim 3 as granted. .
Comments
Confidentiality agreement
If a proprietor alleges the existence of a confidentiality agreement, the onus of proof lies with him.
Confidentiality depends on the circumstances.
In T 1311/21, commented in the present blog, it was held that there is no implicit confidentiality of handbook for a sold product.
In view of G 1/23, the notion of confidentiality or of a confidentiality agreement will play an eminent role.
In T 0143/24, commented in the present blog, it was held that products sold under confidentiality agreement are not part of the prior art.
Free evaluation of evidence
T 2463/22 was commented in the present blog.
T 1138/20 was commented in the present blog.
T 0832/22 was commented in the present blog.
T 1634/17 was cited when commenting T 1311/21 in the present blog.
From those decisions, as well as form the present one, a new line of caselaw seems to emerge which does not rely on the notion of “up to the hilt” when the PPU is whether the PPU is carried out in the sphere of the opponent or “beyond any reasonable doubt” when the PPU is carried out by a third party.
The notion which emerges is that of “free evaluation of evidence” as it has been exemplified in G 2/21.
Comments
22 replies on “T 0267/25-A new criterion to decide on Public Prior Use?”
Does anybody see anything unfair in the idea of dumping “up to the hilt” and “beyond reasonable doubt” in favour of the”free evaluation” principle (the FA principle?)? Why make things more complicated than necessary. Having just read para 27 to 34 of G2/21, I conclude (despite having grown up with English ideas of two different tests of evidence) that (provided the judges are competent, both technically and legally) FA alone is all that we need, regardless which type of fact the evidence is trying to prove.
For more debating fuel here is a link to the newest post in the Kluwer blog, discussing Judge Pumfrey’s Decision on prior use (praised as being not only just but also full of common sense) in Mayne Pharma:
https://legalblogs.wolterskluwer.com/patent-blog/
Dear Max Drei,
Thanks for an interesting comment. I have to reply to it, which should not come as a surprise.
There is, in my humble opinion, no need to dump “up to the hilt” and “beyond reasonable doubt” in favour of the ”free evaluation” principle. “Up to the hilt” and “beyond reasonable doubt” are simply two ways of freely evaluating the evidence submitted by the opponent.
It is clear to me that when the whole evidence material is in the hands of the opponent, the evaluation criteria should be stricter than in the case of the evidence material not being completely in the hands on the opponent. Is this not part of free evaluation of evidence?
Conversely when a proprietor is alleging that a sale was made under a confidentiality agreement it has the onus of a proof. A sale of 10 ventilators to a car manufacturer, is most likely a sale allowing the car manufacturer to test the ventilators. With a sale of 1000 ventilators it will be difficult to claim that it was for test purposes. Is this not part of free evaluation of evidence?
It is difficult to claim that a handbook delivered with some machine freely bought on the market, has not render the handbook public. Is this not part of free evaluation of evidence?
It is difficult to claim that a leaflet printed before an exhibition in order to be distributed there, has never been made public and ended in a rubbish dump or was recycled as old paper. Is this not part of free evaluation of evidence?
In any case, the submission made by a party has to be assessed. If it is called free evaluation of evidence, let it be. I do however have problems when the board comes up with having been “convinced” by a party’s statement. In a criminal case, a culprit can be sentenced in the absence of any proof, if this corresponds to a judge’s intimate conviction. In patent matters, I see it differently.
The following statement of Me Mathély, a late, well know French litigation lawyer, when dealing with IS might best express what I want to say:
“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 his 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”.
I read with interest the Kluwer blog on Mayne Pharma. I see here a difference between the English court and the EPO.
At the EPO any experimental data has to be disclosed in full, for the simple reasons that, on the one hand, the other party should be able to verify the experimental data, and on the other hand, for the deciding body to check whether the experimental data is at all relevant to the actual case. Any comparison with another prior art than the CPA is not considered valid when it comes to demonstrate an effect, as it boils down to comparing apples with pears. Is this not part of free evaluation of evidence?
Daniel thanks for that. Three thoughts now arise, as follows:
First, I agree with you that we should not dump the difference between “preponderance of the evidence (more likely than not)” and “beyond reasonable doubt”. But perhaps we can reserve the latter for criminal trials, lest we send somebody to prison only because we think it “more likely than not” that they committed the crime.
Second, the Mayne Pharma case is relevant where, for example, experimental evidence is used to prove that compound X is the “inevitiable result” of following the instructions in prior publication D1. We need to see how the petitioner AB, in following the teaching in D1, planned their programme of experiments, and we need to see how many experiments performed by AB failed, before they alighted upon the experimental conditions that resulted in the creation of X. How does the EPO manage its “free evaluation” in such cases, I wonder.
Third, a propos M Mathely, there is a notorious US case where the issue was whether a specific publication was “obscene”. The judge made himself an object of ridicule by asserting that he needed no definition because, as to what is “obscene”, I know it when I see it.
@ Max Drei,
First, I do not think that the criterion of “beyond reasonable doubt” should be reserved to criminal cases. Although the main criterion used when assessing added matter is that defined in G 2/10-directly and unambiguously derivable-there is a further criterion used by boards when it comes to assess added matter. It is exactly “beyond any reasonable doubt”. This criterion goes back to T 0383/88, but has been associated with G 2/10 in more recent decisions, e.g. T 1224/14, T 2418/13 or T 1710/13. Another criterion for assessing added matter is that “the original application/patent is not a reservoir of features combinable at will”, e.g. T 1251/19.
Second, since any decision on free evaluation of evidence is a specific decision, it would be presumptuous to say how the EPO would react in a case like the Mayne Pharma case. In any case the boards of appeal of the EPO are not bound by decisions of national or supranational courts
Third, you might have found it funny to refer to the US judge with his definition of “obscene”, but you did not realise, or were not willing to realise, that the quotation of Me Mathély was actually defining the problem-solution approach well before the EPO boards of appeal came up with it. The quotation of Me Mathély deserves more than a quick browse and has nothing in common with the (stupid) comment of the US judge.
Just to clarify, Daniel, my mention of the US case was intended as support for the Mathely quote. I was not poking fun at Mathely but, in fact, exactly the opposite. He is right. Of course In contrast to that US judge I mentioned.
As to Maybe, I was merely curious how the EPO assesses evidence in a case where the fact in issue is whether any given result is “inevitable”. But perhaps there aren’t any.such cases
Sorry. Not “Maybe”. Rather “Mayne”.
@ Max Drei,
When reading your reference of the US judge, it is difficult to see it as being supportive for the Mathely quote. The best proof of it is that you felt the need to clarify. Your comment suffered a manifest Art 84 problem.
I do not know of any case at the EPO comparable with the “Mayne” case. Even if there would have been one, it would be difficult to derive general conclusions from it as the EPO’s boards of appeal are not a common law jurisdiction.
I agree with Mr Thomas. As the Board holds, the sale of a commercial product is presumed non-confidential and the sold product is prior art. And the burden of proof is on the party willing to rebut the presumption.
The patentee ‘s argument. that there is no proof that the sale was non-confidential was irrelevant, a party may not require from the other party the proof of a negative fact, such as the inexistence of a confidentiality obligation binding on the opponent.
Regarding the free evaluation of evidence, my understanding of the case law of the Boards is that it relates to points of fact. But the existence of a confidentiality obligation binding on the buyer of a commercial product raises legal issues, some quite difficult depending on the case. For example, a sweeping confidentiality clause in the general terms and conditions of the seller attached in the quotation or the invoice may not by itself be binding on the buyer, when the buyer has not formally agreed on the clause. This depends on the contract law applicable to the sale.
In this case, what is the Board supposed to do ? Should it request the opinion of a legal expert ? Mr Thomas, are there examples in the case law ?
As to proving a fact with evidence, I think the decision which party carries the burden of proof is not difficult. Harder is to decide when that burden has been carried. In criminal cases, you have to get beyond any reasonable doubt. But when it comes to the question, is the claim valid or not, well then both sides have an equal commercial interest and the balance of probability standard “more likely than not” should always apply, between these two commercial combatants.
I can’t see any better way to decide what fact is “more likely than not” than full and unrestricted application of the EPO’s “free evaluation” fact-finding process.
@ Max Drei,
I do not think that it is the role of any commenter, in any blog, to tell a board how they should appreciate any evidence submitted by a party. Why should the boards be limited to the possibility to decide only on the basis of what is “more likely than not”.
“Beyond any reasonable doubt” or “up to the hilt” are criteria which are enshrined in the case law of the boards. There are no reasons to throw them overboard.
I respect your opinion, but your comment was not bringing matters forward and I was on the verge of deleting it.
Your decision, Daniel, not to delete gives me a chance to defend my position. Thank you.
The way I see it, “up to the hilt” is a normal (and essential) part of the EPO’s (and civil law litigation) “more likely than not” evidential standard.
I have in mind situations in which the evidence is of a “fact” that the tribunal finds seriously hard to believe. It might be evidence of a technical effect that defies the ordinary physical or chemical laws of nature. Or it might be the testimony of a witness that strikes the tribunal as so unlikely as to be not believable. One has to have in place ways to punish those who try to mislead the tribunal or offer it evidence that is not the truth. The more unlikely the fact, the more persausive and powerful must be the evidence needed, to carry it over the “more likely than not” standard.
Other jurisdictions have rules-based evidence. The EPO practises rules-free evaluation of such evidence. That’s OK with me.
@ Max Drei,
Your comment allows me bring a real story. See below.
I would not say that the EPO only practises rules-free evaluation of evidence. Free evaluation of evidence is for me a generic notion which can go from “more likely than not” to “beyond any reasonable doubt” or “up to the hilt”. Without G 2/21 we would not have had this discussion.
Evidence of a technical effect that defies the ordinary physical or chemical laws of nature will most probably be ignored and the reasons for this will be part of the decision. A perpetuum mobile remains a perpetuum mobile whatever a party, expert or witness tells you.
As far as not believable testimonies are concerned, I refer to T 0476/91. In the facts and submission one reads the following:
“The Opposition Division considered, on the basis of the oral evidence of witnesses heard on 21 January 1991, that the evidence relating to the construction of the endoscope and to its distribution was not at all convincing, and awarded the costs in full to the Respondent (patent Proprietor) in accordance with Article 104(1) EPC”. Just have a look at the dicision.
I was chairman of the OD. Looking at the file we had suspicions that the PPU was forged for the sake of the opposition. Being informed by the legal member that he will be invited to repeat his testimony in a binding manner before a German court, the first witness, after nearly four hours, gave up and it became clear that the PPU had been forged. The second witness was only briefly heard and did not even tried to hide the truth. When the third witness, a salesman, was heard, the only question put to him was his sales area: Southern Germany, Switzerland and Austria. As the PPU was allegedly done in Portugal, the last nail in the coffin of the opposition was planted.
Once it was clear that the PPU was entirely forged I merely mentioned, in passing, that we could think about costs. The proprietor’s representative was quick and requested a different apportionment which was granted.
A link to the decision:
https://register.epo.org/application?documentId=EMGSXO3A2656E02&number=EP83302443&lng=en&npl=false
Talking about costs when a party takes you for a ride, was, in my opinion, a well deserved punishment. I still remember the face of the opponent’s representative when I mentioned the costs. The proprietor had two representatives at the OP and since the opponent had brought in the procedure 3 of his employees, the proprietor flew in two technicians from Japan. You can tot it up…
Your reply of 15 August about “forged” evidence and life in the trenches, with real cases, reminds me of an interesting old case in which the EPO made a “free evaluation” of the evidence presented by a petitioner. It involved the EPO’s Corneliusstrasse Night Letterbox and a last-minute filing of an oppositon.
Our sworn evidence from our European Patent Attorney was that he personally had driven in his car to the EPO building and duly filed our envelope at around 10 pm. The EPO’s evidence was that it never received our envelope.
A different attorney swore in evidence that, in his personal experience, the letterbox was on occasion so full that one could reach inside it to remove envelopes (and thereby make space inside the box for one’s own last day filing). This was the likeliest explanation, we suggested, for the EPO’s failure to receive our envelope.
Our sworn evidence was that with a hand and forearm extending into the chute one could feel a choke point at a 135° bend in the direction of the chute.
The Board’s answer: no drawings exist (so there are none that we can show you) and, anyway, your evidence is not plausible enough to oblige us to show you the shape of the chute. At the time, this disinclination to accept our evidence I found quite upsetting.
But (as you surely already know) soon afterwards, the Corneliusstr Letterbox was nevertheless demolished and replaced by a correctly-functioning facility on the north side of the EPO’s Isar Building. Perhaps our evidence did carry weight, elsewhere within the EPO building?
Sometimes, it is hard to prove to the EPO, under its “free” evaluation regime, that what you swear on oath that you did, you really did do. For the Board, our evidence did not get us over the “more likely than not” threshold.
Of course, night letterbox cases are now nothing but history. But what remains is the huge freedom that EPO tribunals enjoy, under the “free evaluation” regime, to make of the evidence what they please.
Mr Hagel,
I agree with you that the sale of a commercial product is indeed presumed non-confidential. If a proprietor claims that a sale was confidential, he has the onus of proof and cannot push the onus of proof to the opponent.
There is one case in which a sale was deemed confidential: T 0143/24, commented in the present blog. This decision relates to the application of G 1/23 and the board came to the conclusion that free access to the market is lacking if the product is sold under confidentiality agreements.
If product and safety data sheets relating to the products sold under a confidentiality agreement are public, not much weight should be given to a confidentiality agreement, however it might be formulated.
In a comment of yours you drew the attention to the rather odd confidentiality agreement buried in the general conditions of sale in T 0134/24..
It is not for me to say how a board should act in such a circumstance. The notion of free evaluation of evidence should allow the board to come to a differentiated and well reasoned view on the matter.
In T 375/00, in case of a technical interpretation, the board was of the opinion that an expert in the meaning of Art 117(1,e) is only then necessary when the board does not consider itself in a position to decide upon a matter without technical assistance. As the board includes two technically qualified members such cases will be rare and will only occur in special circumstances.
There are a few decisions in which parties have requested the hearing of an expert in the meaning of Art 117(1,e). All those requests have been turned down. See for instance T 0230/92, T 0045/01, T 0086/12, T 0038/15 to name a few.
In T 1814/18 and T 1841/18 the proprietor’s request to commission an independent expert on Italian law under Art 117(1,e) was turned down. .
If the problem boils down to a legal problem, like contract law, the board can always decide to enlarge itself under Art 9 RPBA. I therefore doubt that in case of a legal problem, a board will appoint an expert in the meaning of Art 117(1,e).
I therefore do not know a case in which a board has ever mandated an expert in the meaning of Art 117(1,e).
Mr Thomas,
I am very thankful for your answers and your review of the case law in which the hearing of an expert has been requested under Art 117(1-e).
The decisions T 0230/92, T 0045/01, T 0086/12, T 0038/15 you have cited related to points of fact.
T 1814/18 and T 1841/18 are very interesting in that the issues to be handled were points of law (or a mix of fact and law). The set of facts is quite different from T 143/24 you covered in a previous post and it illustrates the complexity of legal issues related to the existence of a confidential obligation in the course of business relationships. In T 143/24, there were uncertainties at all levels as to the existence of a confidentiality obligation (whether there was an agreement at all, the applicable law, the validity of the obligation, its applicability to the sold product).
I am not quite convinced by the Board’s conclusion in T 1841/18 & T 1814/18 regarding the testimony of the witness. It seems to me that the question of whether the witness was under a confidentiality obligation was to be assessed in the framework of Italian contract law, and I doubt on matter of principle that the Board can so easily assert jurisdiction over such an issue. In addition, in the very peculiar circumstances of the case, I find fairly convincing the proprietor’s argument that there was an implied obligation.
@ Max Drei,
It is clear that, night letterbox cases are now nothing but history. However, any tribunal, and not only EPO tribunals enjoys a huge freedom under the “free evaluation” regime, to make of the evidence what they please. Isn’t so that in front of a tribunal you are like on high see, in the hands of god?
I have seen cheating not only in T 0476/91, but in other cases. Copies of a loose leaf catalogue had been so arranged that one could not see the holes in the copies, but they were immediately apparent when the actual copies were presented to the OD.
In another case, an eminent colleague of yours, partner in a well know large Munich firm of attorneys, was furious that the OD requested the presentation of a maintenance manual for which only copies had been provided. He wanted to file an affidavit that the maintenance manual was a bound document. During the OP, he claimed that if he alleged something we had to believe him in his quality of qualified representative, and this should have been enough. He further complained that it cost him custom duties to provide the actual manual and he was going to complain to the President that he had unnecessary costs. Here again the maintenance manual was a collection of loose leaves. He also refused a lunch brake as he had more important things to do….. Those kind of representatives cannot be helped.
I can understand that you and your colleagues were upset when your statements were not accepted by the board, but what else should the board have done? Honest representatives have, alas, to pay for dishonest ones. That is also part of life’s rich tapestry.
If it had been the filing of an opposition at the last possible moment, I would most probably not accepted such an excuse either. It looks prima facie like a clearly fabricated excuse.
With regard to “beyond any reasonable doubt, I would like to add the following:
In a previous reply I had drawn your attention that the criterion “beyond any reasonable doubt”, applied not only to the evidence of a PPU, but also to added matter.
In T 0614/26, published on 11.08.2026, the board referred to T 307/05 and reminded that it is established case law that the requirements of Art 123(2) and (3) are assessed according to the standard of “beyond reasonable doubt”. This applies mutatis mutandis to any evidence of a PPU, especially when it comes to evidence exclusively in the realm of the opponent.
Daniel, your reply astonishes me. The attorney who gave affidavit evidence concerning removal of envelopes from the Corneliusstr letterbox was me. So when I read that prima facie my evidence was “clearly fabricated” my eyebrows shot up. Besides, at least in my old firm, it was routine to file oppositions on the last day, even after normal office hours.
It is a pity that I have forgotten the B publication number or the name of our client, the last day opponent. If we could find the Decision we could read how the Board skilfully found wording to close the case without offence to the opponent and without opening up the “appalling vista” of a night letter box not fit for purpose. Fact is though, soon after the case was heard, the Corneliusstr letterbox disappeared.
As to how to handle witness evidence that the court is unable to accept, in a case of alleged prior user, I recommend Robin Jacob’s elegant judgement in the Leonard Studio camera dolly case.
In the interests of justice and the Rule of Law, I assert that the reasoning in published decisions must be of high enough quality that it squashes any complaints from one party or the other, that the Decision is somehow unfair. The reasoning is just as important as the bare Yes or No decision itself. I hope you agree, Daniel.
@ Max Drei,
I gave you my honest opinion when I said that, should I have been faced with a late filed opposition and the opponent would have come with a story like yours, I would prima facie consider it as a fabricated excuse.
I know that you are a decent and reasonable person, but in opposition, the OD has to take a decision irrespective of the person filing affidavit evidence. Whether you like it or not, the odds were against you, and irrespective of the person, it looks really prima facie like a fabricated excuse. or a least woth a film scenario. That the Corneliusstr letterbox disappeared later is not actually relevant.
Why on earth was it not possible to file the opposition during office opening hours at the counter which was open to this effect? Filing it at night shortly before the deadline was looking for problems.
You have, alas, been paying for some of your colleagues having cheated. I can agree with you that the reasoning in published decisions must be of high enough quality that it squashes any complaints from one party or the other. I would never have used the term fabricated excuse, but the decision should be such that the reasons why the excuse cannot be admitted has to be given in all clarity, in a decent form and not by a mere yes or no.
I would be interested to read Robin Jacob’s judgement in the Leonard Studio camera dolly case with the proviso that it cannot be binding on the EPO. I could not find it.
Daniel, I could not find Leonard Studio on the internet either. But I have a paper copy somewhere and as soon as I find it I will scan it and email it to you. I was in court when the witnesses were cross-examined. It was very interesting.
As to “up to the hilt” and “beyond reasonable doubt” Daniel, I accept that this is the burden a party must carry, when it alone has the evidence, orv when the issue is Art 123(3). I would see this as real life pragmatism by the EPO, common sense, and in the interests of justice. But it is an exception to the general rule, right?
You mention T0614/26. I think you mean T0614/24. It concerns Art 123(3) EPC. Its para 11 construes claim 1 of the MR and finds that it covers embodiments not within the scope of claim 1 of the patent as granted. That strikes me as a question of law rather than fact. But perhaps at the EPO that distinction is meaningless.
@ Max Drei,
I have to repeat that, “up to the hilt” and “beyond reasonable doubt”, as well as “more likely than not”, are within the framework of the general rule of free evaluation of evidence.
You are right that the decision I referred to is T 0614/24 and not T 0614/26. Tomorrow 19.08.2026 the will be a blog entry on this case with the right number. I apologise for my mishap..
@Max Drei,
In J 0002/25, the applicant could not file by electronic means a divisional application before the mention of the grant for the earlier patent application was published in the European Patent Bulletin due to a technical problem, with the consequence that at the time when the European patent application was actually filed, the earlier application was no longer pending.
The European divisional application could not be filed as the representative systematically received an error message. In protection of the legitimate expectations due to the content of an EPO Notice, cf. OJ EPO 2020, A120, and due to the fact that this notice does not indicate that all available means of filing must be used in order to have access to the safeguards.
This story is quite different from being allowed to withdraw documents from an EPO letter box.
Although the case is about filing of a divisional, it should logically also apply to filing an opposition on the last possible moment.