CASELAW-EPO - reviews of EPO Boards of Appeal decisions

T 143/24 – Products sold under confidentiality agreement are not part of the prior art-Conform with G 1/23?

chat_bubble 6 comments access_time 5 minutes

EP 2 651 581 B1 relates to a low-emission cold-setting binder for the foundry industry.

Brief outline of the case

The opponent brought forward a series of public prior uses. Witnesses were heard by the OD. The OD did not find the PPU sufficiently proven. The OD decided that claim 1 as granted was lacking IS, but the patent was maintained according to AR1.

Proprietor and opponent appealed.

The board confirmed that the PPU were not sufficiently proven.

The board infirmed the OD’s decision to maintain the patent according to AR1, and maintained the patent as granted.

The decision is interesting as it relates to the sale of products under a confidentiality agreement.

The opponent’s point of view

Binders according to claim 1 had already been openly marketed, delivered, and used prior to the priority date of the patent within the scope of the three prior uses “FDM,” “FSC,” and “ASK.”

The sale of the binders to customers of the opponent was carried out freely. There was no confidentiality agreement between the respective contractual partners.

The general terms and conditions were sent or submitted independently of the products and therefore could not impose any confidentiality obligation.

The composition of the binding agents used in the claimed prior uses was not secret. Rather, customers were informed of the type of binding agent before purchasing it when it was offered for sale. Furthermore, the composition could also be analysed, even if the production process itself was not public knowledge.

A confidentiality agreement would also not have been in the interests of the parties to the contract.

At the OP, it was conceded that it could be assumed that the content of the terms and conditions had always been the same.

For the opponent, the witness statements confirmed that neither the sellers nor the customers were aware of any confidentiality agreement. The products of prior use and their product and safety data sheets could have been freely passed on to third parties. The data sheets had also been freely distributed to any third party for information purposes. These were publicly advertised mass-produced products.

The composition of the binding agents of the asserted prior uses was not secret.

The proprietor’s point of view

The sale of the binding agents under the prior uses ‘FDM’, “FSC” and ‘ASK’ was carried out in accordance with the applicable general terms and conditions.

The general terms and conditions proved that the sellers of the binding agents, Ashland Group, usually based sales within the scope of prior uses ‘FDM’, “FSC” and ‘ASK’ on terms and conditions that included confidentiality agreements.

They concerned ASK Chemicals Espana S.A.U., a subsidiary of the opponent, and Iberia Ashland Chemicals S.A. of the Ashland Group.

Such a confidentiality agreement was also in the interest of the seller in order to prevent third parties from copying the binding agents.

Furthermore, this intention to maintain confidentiality is also evident from the fact that the product and safety data sheets submitted by the opponent did not contain a complete list of the raw materials. The redactions in the documents submitted and the use of non-specific designations that did not reveal the chemical substances and their quantities showed that the composition of the relevant products was not to be made available to the public, which should have been known to the business partners in these long-standing business relationships.

This is also consistent with the limited customer base of the selling  at the time and the statements of witnesses Aramburu and Gonzales, who testified that the composition of the products was an internal matter for Ashland and that the disclosure of the data sheets was not desired.

The end products also did not readily allow conclusions to be drawn about the raw materials used, as these would have to be decomposed during the use of the binding agents.

The board’s decision

The board considered that the opponent has not demonstrated and substantiated beyond doubt that the sales or deliveries claimed and the associated data sheets for binding agents at stake were not subject to confidentiality and thus became public knowledge.

In the case law of the boards, it is recognised that the opponent bears the burden of proof and presentation for any prior public use it asserts.

Only what has been made available to the public before the filing date belongs to the prior art. In order to satisfy its burden of proof, the opponent must therefore, in particular, demonstrate that the prior-used subject matter was publicly available.

If the sale of an object and its delivery to a customer is claimed as prior public use, the recognisable features of the object sold generally become publicly accessible if the transfer of the object sold takes place without any confidentiality obligation and can be analysed by a specialist.

Contrary to the proprietor’s view, reproducibility is not necessary for a product that is freely accessible and analysable on the market to be considered prior art under Art 54(2), see G 1/23, Reasons 73, 74.

For the board, free access to the market is lacking if the product is sold under confidentiality agreements. Contrary to the opponent’s view, nothing else follows from G 1/23, as this relates to products that are on the market and accessible to the public. This condition is not met if the product is only marketed under confidentiality agreements.

In the present case, the terms and conditions governing the sales claimed as prior use provided for a confidentiality agreement, which meant that the products sold were not accessible to the public and therefore did not form part of the state of the art under Art 54(2).

Comments

The conclusion of the board is understandable, but is the decision conform to G 1/23?

Could we see in the present decision a dent in the conclusions of G 1/23?

In G 1/23 it was held that:

  • A product put on the market before the date of filing of a European patent application cannot be excluded from the state of the art within the meaning of Art 54(2).
  • Technical information about such a product forms part of the state of the art, irrespective of whether the skilled person could analyse and reproduce the product and its composition or internal structure before that date.

In G 1/23, Reasons 53, the EBA linked confidentiality with CGK. The EBA held that the knowledge of the skilled person is derived from the fact that some specific technical teaching is considered to be known by a large community of skilled persons working in a given field, it being understood that the particular technical information is considered by all of them to be so well known that its public availability – in the sense of not being confidential and known only to a restricted group – is beyond any doubt.

The EBA added that it may be discussed why and to what extent the state of the art can be broader than common general knowledge,…., but this is not relevant in the present context. What matters is that CGK cannot cover more than the state of the art.

in the present case, there is no doubt that the products at stake were marketed, however with the proviso of a confidentiality agreement.

This could however not withhold the recipients of the products to analyse the products. The recipients might not allowed to give this information to third parties, but it is undisputable that the products were marketed before the effective date of the claim.

Furthermore, the product and safety data sheets could be freely passed.

Circumvention or negation of G 1/23?

As the products were sold to a third party, there is no doubt they were marketed.

Following the present decision, when marketing products under a confidentiality  agreement, the whole aim of G 1/23 could easily be circumvented, if not totally negated.  

I am not sure that the EBA ever envisaged such an outcome.

In view of G 1/23, it appears difficult to accept that the status of marketed products can be made dependent from a confidentiality agreement between private parties.

T 143/24

Note

In the parallel case, with the same parties, T 141/24, the same board came to the same conclusions with respect to the prior uses of the opponent.

We will have to see if other boards interpret G 1/23 in the same way.

Share this post

Comments

6 replies on “T 143/24 – Products sold under confidentiality agreement are not part of the prior art-Conform with G 1/23?”

Anonymoussays:

I think the summary of G1/23 leaves out an important point. The headnote in G1/23 says:

Technical information about such a product *which was made available to the public* before the filing date forms part of the state of the art (…)

Without this addition it might seem that information regarding composition should, in some kind of legal fiction, be regarded as part of the state of the art, even if the skilled person wasn’t able to obtain it. I don’t think that G 1/23 means that.

Avatar photoDaniel X. Thomassays:

@ Anonymous,

What did you intend to prove with your comment?

I can reassure you, I can read, and have understood G 1/23 as well as you can do.

In my summary of G 1/23, “such a product forms” in the second bullet point refers to the “product put on the market before the date of filing of a European patent application” in the first bullet point. With a mind willing to understand, you would not have come up with your comment.

Furthermore, you have manifestly not looked at the details of the case. Product and safety data sheets relating to the products sold were also discussed during the hearing of the witnesses.

Experience shows that only rarely product and safety data sheets are kept secret, as they serve to advertise the products, but they rarely give any precise indication about the composition of the products. This was for instance the case with the product Engage ® 8400 at stake in the decision leading to G 1/23.

Product (and/or safety) data sheets often also comprise a statement of the kind. “We reserve the right to modify the composition of the product depending on the evolution of technical knowledge”. This means clearly that the composition of the product might change with time, but that a given composition has existed at various points of time.

Any technical teaching that can be derived from those documents, made available to the public before the effective date of the claim, manifestly belongs to the prior art.

What is not made publicly available can by no means belong to the prior art, whether by a legal fiction or any other means. Interpreting G 1/23 in this way would be non-sensical.

Your comment boils down to what I would call in French “Enfoncer une porte ouverte”, i.e. breaking down an open door.

The point raised in the blog is the following: can the sale of products to third parties before the filing date of a patent, but under a confidentiality agreement, lead to the products not belonging to the prior art, i.e. to circumvention of G 1/23?

I have some doubts about this, as it would mean that G 1/23 could be circumvented by a private agreement between parties.

After all, being delivered to a third party, they are in principle analysable by the latter, cf. G 1/92.

I was hoping for an answer on this specific issue, and not a comment manifestly beside the point.

Max Dreisays:

What comes to my mind is a 1950’s case in England, called Dalrymple. Link below. Just because something is labelled CONFIDENTIAL does not make it so. If it is disseminated to a controlled number of selected users, the confidentiality is likely preserved. But if it is distributed widely throughout an entire industry, then it ain’t “CONFIDENTIAL” any longer. Commonsense, or what? 1957 RPC 449.

It’s a bit like labelling a document PRIVILEGED. That status can also be lost by uncontrolled dissemination of the document. 1957 RPC 449.

https://swarb.co.uk/in-the-matter-of-j-r-dalrymples-application-for-a-patent-1957/

Max Dreisays:

OOPS. Sorry. The reference to 1957 RPC 449 is to the official publication of the Dalrymple decision. The second inclusion of the reference, at the end of my second paragraph, was an error and should be ignored.

Avatar photoDaniel X. Thomassays:

Dear Max Drei,

Thanks for your comments.

In my opinion, the Dalrymple case corresponds to what is envisaged in Headnote 2 of G 1/23: technical information about a product which was made available to the public before the filing date (I would say before the effective date of a claim), forms part of the state of the art within the meaning of Art 54(2).

Common sense would agree that the mere fact that a document has been labelled ‘confidential’ will not automatically confer confidentiality on it if it has been made generally available. I do not know of any corresponding board decision on this topic.

The case at stake is slightly different in that the products were sold to a third party and were considered by the board to have been sold under a confidentiality agreement, here according to general terms and conditions. Therefore, the sold products were not part of the prior art.

If tomorrow sales are made under a confidentiality agreement, merely according to general terms and conditions, this could have a massive impact on the applicability of G 1/23.

That relations between parties for instance for prototyping or on collaborative projects between parties, imply confidentiality is not at stake here. Also, a transfer from a company to a subsidiary, cannot be considered to be public, as it does not leave the reach of the company. It is only if the subsidiary sells the product to a third party, that the last sale can be held to be public.

When looking at the CLBA, 11th edition, there are mainly very old decisions quoted, and none which could have an impact on the applicability of G 1/23.

francis hagelsays:

In this case, I think the opponent could have challenged the patent owner’s reliance on the confidentiality clause of their general conditions of sale (« CGS »).

I have looked at the CGS, which is available in the online file. The online file contains the original in Spanish language and a German translation.

The first remark is that the law applicable to the interpretation and execution of the CGS is Spanish law. The opponent could have required from the patent owner some evidence of the validity and scope of the confidentiality clause under Spanish contract law. I suppose it was not up to the OD or BOA to raise this issue of their motion in the absence of opponent action.

Other remarks relate to the wording of this clause. It defines « Confidential Information » so broadly that it makes it impossible to distinguish non-public information from publicly available information, which is clearly the case for many of the items covered by the super-broad definition. It sets no time limit to the confidentiality obligation.

In addition, the restriction to the use by the buyer of « Confidential Information » is worded in a very unusual manner, if I compare with standard practice in confidentiality clauses and non-disclosure agreements. The clause says that the buyer may not use« Confidential Information », except if it benefits the seller « in the performance of the agreement ». This language is odd. First, the CGS provides no definition for « the performance of the agreement ». Second, a definition one could find reasonable would relate to the relationship between the seller and the buyer – but this would say nothing about the scope of authorised uses by the buyer in the course of its business activities.

Leave a Reply

Your email address will not be published. Required fields are marked *