G1/19 (EPO Enlarged Board of Appeal): NO Contradiction between COMVIK and T1227/05 (INFINEON) ?

G1/19 (EPO Enlarged Board of Appeal): NO Contradiction between COMVIK and T1227/05 (INFINEON) ?

Just recently I have seen an aide-mémoire by Dr. Michael Stadler, of the oral proceedings in the case G1/19, here https://kandidatentreff.de/2020/08/vorlage-g1-19-nach-der-muendlichen-verhandlung (remark: Google Translate is not a reliable translator for that content).

I have watched those online proceedings for a short time but I could neither clearly identify the various opinions that have been ventilated by the persons involved, nor could I recognize any clear structure in the oral proceedings. But that is probably my fault, and not of those people that we could see on the screen.

Dr. Stadler confirmed to me that he has no additional insider knowledge and that he took all of his information exclusively from the video stream and from the file inspection, and that the main source of his remarks was the oral hearing itself.

So I take his above-mentioned article as a very credible eye-witness report.

Further Information

Further information on the G1/19 case can be found here:

Patent in suit before the European Patent Office (EPO):



Claims on file for the oral proceedings:


This is a must-read: Enlarged Board of Appeal Communication of 22.06.2020:


Surprise Surprise: “Are the Proceedings Admissible At All?”

What immediately jumped at me in Dr. Stadler`s aide-mémoire was the following remark:

“First of all, the Enlarged Board of Appeal surprised with a question that apparently only emerged in preparation for the hearing. It was discussed whether the decisions might be compatible with each other anyway, so that an opinion is not necessary.”

It is my strong opinion that the Enlarged Board of Appeal is right. COMVIK and T1227/05 (INFINEON) are compatible with each other.

In short words: the patent claims in INFINEON fulfill the patentability criteria that were set out in COMVIK.

It is just that INFINEON came from a direction that was different from the direction COMVIK.

The INFINEON 1/f Noise EP Patent

In order to verify these statements, one needs to study the INFINEON decision in the light of the 1/f Noise Patent EP 01964907.8 for which it was rendered, here

INFINEON, T 1227/05 https://www.epo.org/law-practice/case-law-appeals/recent/t051227ep1.html

1/f Noise Patent EP 01964907.8 https://worldwide.espacenet.com/publicationDetails/biblio?DB=EPODOC&II=0&ND=3&adjacent=true&locale=de_EP&FT=D&date=20021120&CC=EP&NR=1257904A1&KC=A1

That exercise is a time-travel for me, 20 years back in time. It was summer 2000, right after honey-moon holidays, that I have drafted the 1/f Noise Patent.

I was at that time a very young patent attorney with only 4 years of professional experience as an attorney and I already had several landmark decisions for patents about Computer Implemented Inventions (CIIs) under my belt. But I am digressing.

I still remember that it was very clear to me that I had a patentable invention in front of me, but I also wanted to lay a trap for the patent examiners to come. This is why I put a sinister remark on page 1 of the 1/f noise patent specification: the invention can be used “… for modeling the effects of noise in a large number of technical and physical systems as well as for systems for assessing and predicting events in the financial markets.”. Repeat after me: for. assessing. and. predicting. events. in. the. financial. markets. After all, we were in 2000 in the midst of the dot-com stock market bubble bursting.

The original patent claims were silent about any specific application of the claimed method.

It was my hope that an unexperienced examiner would object to the technical character of the claimed algorithm itself, and that we – at one point – stop to argue about that and file an appeal. As a fallback position for appeal it was my hope that we can make the claim patentable by including its real purpose – simulating electronic components such as pn diodes and MOS field-effect transistors – as a limitation into the claims.

And my strategy worked out. To an unexpected extent.

The INFINEON Appeal Proceedings

The case went as I have thought: the EPO examiner rejected the patent application, we filed an appeal, and I ended up in ex-partes oral proceedings before the Board of Appeal.

In fact, all that has been discussed in the oral proceedings was about how to solve my trap of creating a patent that potentially also protects a clearly non-technical method of predicting financial market events. There was never the slightest doubt that the algorithm, being applied on a computer, is patentable. But the board of appeal did not want to give me the broad claims on file that also seemed to protect a pure business method.

I have filed two or three sets of auxiliary requests that have all been rejected until the Board of Appeal came up with their own idea of patentable claims. Which you find in the granted 1/f Noise EP patent.

Please note that the parallel US patent has a much broader scope of protection, given that the events all happened in pre-Alice times. But I am digressing again.

I consider the INFINEON 1/f Noise EP Patent to be a personal success. It took me 5 years to drive that success home.

And I am today still grateful that my client Infineon, and later Qimonda, allowed me to represent them in court for that case.

The After-Math of T1227/05 INFINEON

What came after that success was largely unexpected for me.

My 1/f Noise INFINEON decision T1227/05 was then published in the EPO Official Journal 11/2007, together with the new Guidelines for Examination that have been adapted to my decision. Now that is a powerful success indicator.

Not everybody shared my personal joy about that success. Some of my peers in my own firm showed signs of envy about that which I found quite hurtful for a short time. But, as a consequence, life for us CII patent attorneys became much easier after my 1/f Noise INFINEON decision T1227/05 being rendered because we could better predict what is patentable and what not.

And that was worth the effort.

Not Everything is Sunshine

But I also saw weird things happening.

CII patent applications came up that only relied on a computer simulation of technical subject matter per se, without also fulfilling the COMVIK criteria for patentability. Similar to a new indication for a known pharmaceutical. And EPO actually granted some of these barren patent applications. I have not carefully studied the patent in suit of the G1/19 case but it is my suspicion that there might be a relevant point here, as Dr. Stadler´s aide-memoire says:

“… there is no technical effect in the entire claim that the (Enlarged) Board (of Appeal) is willing to believe.”

It is my strong conviction that a patent application for a CII must have at least one embodiment that shows each step of how the claimed algorithm transforms a given set of sample (input) data into a second set of (output) data. And best is to also describe how the data is handled by the various components of the computer system that executes that algorithm. One may still get a granted patent for a CII patent that does not have these detailed parts of the description but that joy of a successful shortcut may only last until the same patent undergoes revocation proceedings. Such poorly drafted patents are seldom found enabling and therefore revoked, and I have many examples for that.

A second point worth mentioning is that I have been attending talks and meetings where people tried to apply my 1/f Noise INFINEON decision T1227/05 to a specific case, and I have never met anybody who would know or understand what I have described above. That is also why I personally find the Enlarged Board of Appeal`s remark about a potential basic flaw in the case G1/19 so refreshing and enlightening.

What The Applicant May Feel

I have inherently never been good in identifying other people`s feelings, but being put into the boots of the applicant in the case G1/19, I would not feel comfortable with the ongoing proceedings.

As interesting as the underlying case is, I would not understand what is going on. I was in appeal proceedings against an EPO decision, and suddenly I find myself within a large audience before the Enlarged Board of Appeal, and it is unclear who is fighting against whom. Applicant against EPO against Board of Appeal Against Enlarged Board of Appeal, whom should I support, and whom not? Who pays my attorneys for reading 20+ Amicus Curiae submissions? Do I need to attend oral proceedings before the Enlarged Board of Appeal, which means further costs? And will there be further oral proceedings before the Board of Appeal, which again means further costs?

Questions over questions.


I find it interesting to be only indirectly involved in proceedings before the Enlarged Board of Appeal. I am not representing any party, and I do not need to explain anything here. I can ask questions without being liable for the answers. And I am good with any outcome of these proceedings.

But what gives me confidence for the future of CII patent applications in the future is that the systematic efforts of the European Patent Office, the Boards of Appeal, and the Enlarged Board of Appeal will lead to even greater predictability of future CII patent applications before the EPO.

My unbroken credo is that the European Patent Office is THE place to prosecute CII patent applications in this world.

    Did you like what you just read? Then subscribe to my free Tip of the Week!