Bonum Certa Men Certa

Consequences of Straining Staff: Patent Quality at the European Patent Office Has Gotten Rather Terrible

Squeezing not only staff but also the EPO's (traditionally) good reputation for short-term gains

Strainer



Summary: A look at some recent news and personal perspectives on the status quo of today's European Patent Office, where examiners' performance is measured using the wrong yardstick and patent quality is severely compromised, resulting in overpatenting that the public pays for dearly

THE EPO 'results' are being debunked already, but the EPO's PR team keeps spreading that today, regardless of all the scrutiny [1, 2]. How much longer is this sustainable for?



Earlier today we spotted English language Chinese media citing the EPO's 'results' which are basically embellished if not a half-truth/lie (or intentionally bad statistics).

"How much longer is this sustainable for?"Why is the EPO risking damage to its integrity and record on truth? Well, it already lies to journalists and to staff, so there's probably not much reputation left to lose.

The other day we noticed EPO-friendly media citing the USPTO's Senior Counsel as follows:

USPTO Senior Counsel Mark Cohen notes that aggressive antitrust enforcement like that sought in the Hitachi Metals case could affect patenting activity in general, adding: “This can be of great concern in the non-SEP space, where a patentee may have a choice whether to disclose an invention or keep a proprietary method secret.”


Actually, this is what many in the software field do. They rely on hiding source code, copyright on Free software (that everyone can study, inspect and modify as long as copyleft is respected), and there's hardly any room for antitrust enforcement in such a setting. There's no rigid requirement imposed, nor are there patent lawsuits and shakedowns. India has been getting it right on software patents for many years and as WIPR put it a few days ago:

The revised guidelines on computer-related inventions by the Indian Patent Office imply a reversed position on whether software inventions should be patentable. Abhishek Pandurangi of Khurana & Khurana reports.

After the Indian Patent Office (IPO) published the first set of guidelines for examining patents for computer-related inventions in August, in February the office introduced an amended set of rules.

While the previous guidelines were kept in abeyance in response to strong protests by critics, a revision was expected, but surprisingly the IPO has replaced an excessively liberal set of guidelines indicating that any software is patentable with a contrary one which almost indicates that no software patents are allowed.


This article from Abhishek Pandurangi serves to remind us that much of the world (large populations) does not accept software patents, whereas the EPO increasingly does, unlike the US where Alice keeps marginalising them.

"The answers to the questions about patent leniency may actually be found in anonymous comments from insiders."Simply put, under Battistelli there is a huge patent maximalism problem. Patent scope gets broadened in pursuit of additional profits, rendering any performance requirement invalid (comparing apples to oranges, if not actually patenting apples and oranges, which now seems possible at the EPO). Yesterday we saw this announcement titled "EPO Revokes a Patent of Biogen, Inc.’s (NASDAQ:BIIB) Top-Selling Tecfidera". Why was this patent granted in the first place? Working under pressure or in rush? Inclination to lower the patent bar and issue/grant bogus patents? Whatever it is, as the article put it: "The European Patent Office (EPO) has revoked European Patent EP2137537, a method of use patent concerning Tecfidera, last week. If left unresolved, the move will take a big hit on Biogen’s balance sheets because sales of Tecfidera account for a third of its overall revenue in 2015."

The answers to the questions about patent leniency may actually be found in anonymous comments from insiders. While many comments on the debunking of EPO 'results' have come from EPO apologists (if not worse) who are simply shooting the messenger, some of them come from insiders who acknowledge the problem (we have been hearing about these problems for a while). To quote:

I accept that life always involves compromises. But it is distressing to see EPO examiners slowly turning into the three wise monkeys (that is, if you don't look too hard for problematic prior art, and don't think too hard about strict compliance with all of the provisions of the EPC, then examination becomes a lot simpler... and faster).

I fully understand what is driving this process, as applicants, the EPO and national patent offices all stand to benefit. However, it does look like it could be the beginning of a process of erosion of the fundamental bargain with the public that underpins the whole patent system.

I am not saying that where we stand now is definitely unacceptable. Instead, I am merely pointing out that what appears to be a drive from the EPO for “examination light” represents a potentially dangerous trend that needs counterbalancing with strong input from voices representing the public interest.

I say this not as a "patent denier" but rather someone who believes in the patent system, and who wants to cherish it for many years to come.

Think about it. If the pendulum swings too far in terms of permissiveness, then there are certain to be cases where aggressive patent owners assert blatantly invalid patents against competitors with shallow pockets – potentially aided by the €11k fee at the UPC for filing a counterclaim for invalidity. It will not take many cases where a patent owner can be painted as a bad actor for there to be overreactions in the opposite direction. If you have any doubts about what can happen, then witness the effect that lobbying by anti-patent pressure groups has had on “gene” (or other “natural phenomenon”) patents in the US and Australia. Scary stuff!

I agree that some of the professed aims of the ECFS system are laudable. Indeed, there is no point prioritising cases where everyone is happy to let sleeping dogs lie. However, it is not hard to see that much of what is prioritised by ECFS are the “easy wins”, where examination is very straightforward.

The inevitable short term hike in productivity figures produced by ECFS is not to be welcomed for two reasons. Firstly, it will leave a rump of “clearly difficult” cases that no examiner wants to tackle – because the time taken to sort them out will be too detrimental to the examiner’s apparent productivity. Secondly, it is likely to provide a strong temptation to examiners to keep their productivity figures high by waiving through “borderline” cases – ie. treating them as if they are also “easy wins”.

Nobody expects the EPO’s search and examination to be exhaustive. However, they will be doing us all a favour if a way is found to reward examiners for doing their job properly – and not just speedily. In this respect, it is important to acknowledge that it is impossible to constantly drive down the time taken to search and examine applications without compromising on quality. The best that you can hope for is an acceptable balance between speed and quality. Thus, management initiatives that seek to constantly increase productivity look increasingly like a drive to reach the bottom of the barrel.


Here is another noteworthy and long comment:

Based on our own experience and talking to others in the profession, it seems that for some examiners getting examination reports issued quickly involves being totally unhelpful, simply not dealing with issues or throwing in a load of amendments and gambling on the applicant just accepting what is given to avoid remaining in examination.

I have recently seen a first examination report to issue on amendments filed in 2011 in response to the EESR. Unfortunately, an amendment shown on a manuscript amended copy of the claims filed with the response did not make it into the clean copy of the claims. The amendment was described at length in the covering letter and is shown clearly on the manuscript amended copy. Rather than examine the claims as including that amendment or call the representative to ask him to submit a clean copy of the claims that included the amendment, some four plus years after the filing of the response, the examiner examined the claims as those the amendment did not exist. Furthermore, despite the fact the representative's letter explained various reasons why other features of claim 1 distinguished over the cited prior art, the examiner has just parroted the objections from the EESR without giving any clue as to why he/she disagrees with the representative's analysis. So about as helpful as a chocolate teapot. However, somewhat craftily, an allowable dependent claim has been allowed.

In a case of my own, we submitted amended claims on entering the regional phase accompanied by a two-part letter explaining the basis for the amendments. Ahead of the search report we got a note from the examiner saying that no basis for the amendments had been supplied when the amendments were filed and if this was not supplied within a month or two, I don't remember which now, the amended claims would not be searched. We were given no more information than that so wrote back pointing out that we had filed a two page letter explaining the basis for the amendments. The response from the examiner was to issue a partial search report with the comment that neither our first letter or our second letter explained the basis for amendments. As far as I can make out, since the **** has not been sufficiently helpful to provide any useful indication, he just did not consider the explanation of the amendments to a couple of claims sufficiently complete. Leaving aside whether he is entitled to ask for further detail at that stage, we might have been able to move things forward if he'd just said asked us to provide additional explanation of the amendments made to particular claims instead of sending out a communication which was misleading and, basically, factually incorrect.

These are just two examples I am aware of and I am guessing that neither I nor those I know in the profession are being singled out for special treatment.

Basically we are seeing cases that already suffer from long delays in examination making no meaningful progress because examiners are simply bloody-minded, unhelpful or do not take the trouble to explain the issue. How much this is down to the mindset of individual examiners and how much a response to management pressure, I really do not know; whatever , it is not doing the reputation of the EPO any favours.


As the above put it, “cases that already suffer from long delays in examination [are] not doing the reputation of the EPO any favours.” Discriminatory practices aren't the solution to this.

Not only delays are the problem; patent quality too is a serious issues, including grants of software patents in spite of the EPC.

Recent Techrights' Posts

SLAPP Censorship - Part 161 Out of 200: Low Standards in Defamation Cases Just Muddy the Waters and Distract From Legitimate Cases
The judge at the trial said that Garrett's case was a waste of the court's money
Congrats to Linux.org for Adopting or Getting Back to IRC
This is the way the Net ought to work and was originally designed to work [...] IRC as a protocol turns 38 this month
 
GNU/Linux Did Not Start in 1991 and America Wasn't Discovered by Europeans
it'll be 43 next month
Rolling Out Some Changes Soon
To the regular reader no change will be seen
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Monday, August 24, 2026
IRC logs for Monday, August 24, 2026
Gemini Links 25/08/2026: Separated by Plexiglass, Low-Tech Information Networks, Jörg Rippel Comes to Geminispace
Links for the day
The Tragedy of Software Developers Making Up Narratives and Making Excuses for Plagiarism (of Their Own Work, Too)
one lingering issue is that many who vote in Debian GRs receive money from slop companies
Evri Makes Us Optimistic About the Collapse of the Slop Pyramid Scheme (Bubble)
Do not be seduced by false promises of "automation" or "intelligence" where only automation may exist but no intelligence at all
Links 24/08/2026: "Journalism Can Help Expose Bad Science And Trigger Real-World Change"; Further Suppression and Censorship in China/HK
Links for the day
Gemini Links 24/08/2026: Soul Mentality, Words to Live by, Mozz.us Gemlog Resumes, Smol Conversations
Links for the day
Today The Register MS Published Greenwashing Spam for the Slop Pyramid Scheme, It Mentions "AI" 29 Times
More people need to talk about the role of the media in this pyramid scheme
Slop Plagiarism and Chatbots Are Killing Evri (They Infuriate and Insult Clients)
Slop destroys companies and leads to misery (miserable clients, time-wasting)
Links 24/08/2026: Re-defining the IndieWeb and "Data Center Backlash Bursts Into the Midterms"
Links for the day
The Issue With Omarchy is the Slop, the Politics Are a Side Issue
Those corporations do not oppose slop, they participate in it
In South Korea, Steady Increases for GNU/Linux
authorities said they would migrate to GNU/Linux or consider moving in that direction
SLAPP Censorship - Part 160 Out of 200: In Astounding Repetition of Last Year, Brett Wilson LLP Deliberately Ignores Holidays of People It is Attacking and Crushes Principles of Access to Justice
Disconnected from the law
Links 24/08/2026: Vision and Skill, Doing Good, Chiperia Project
Links for the day
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Sunday, August 23, 2026
IRC logs for Sunday, August 23, 2026
Gemini Links 23/08/2026: Turning 60, PineTime, and Simulation Hypothesis
Links for the day
Will Your "Modern" New Car Still be on the Road in 2060?
Are people now expected to change a car as often as they replace a mobile phone and, if so, how fast would costs add up?
The Sniff Test
Be sceptical of "CoCs"
Cloudflare Sees a Quarter of Requests in China Coming From GNU/Linux Today
Will 23% ever be the average for GNU/Linux rather than a mere peak?
You Can Lose Some Battles and Still Win the War
Winning or losing isn't something for a scoreboard
GNU/Linux Usage Measured at 23% in Russia Today
the average was almost 10%
Things Not to Fear Missing Out on
Life is too short to pay attention to social control media
Gemini Links 23/08/2026: Exercising Again, Tackling Phone Addiction, and Putting Graal Online Back Online
Links for the day
Northern Africa: GNU/Linux Measured at Around 5%
by Clownflare
PIPs and Lawsuits: On the Future of IBM Trying to Spit Out Its Own Staff at Minimal Cost
"I'd rather be laid off than be put on a PIP"
France: GNU/Linux Averaging at 7%, Peaked at 14% Today
When will 14% be the average?
Links 23/08/2026: Water Crises, Illegal Tariffs, and Carney Confronts US Over Patent Imperialism
Links for the day
Links 23/08/2026: Slop "Children's Stories Contain Bizarre Patterns" and "Butterflies at the One Garden"
Links for the day
SLAPP Censorship - Part 159 Out of 200: Telling Courts False Information and Spoon-feeding Them Insults, Hoping or Expecting Them to Repeat These Insults
When lawyers trick courts into repeating something false
Microsoft is in Double Trouble in Singapore
Android+GNU/Linux+ChromeOS are near 20%
The Slop Industry Bribed the Media to Pretend It Has Something New and Revolutionary. Now It Bribes Politicians Too. Anything to Avoid Scrutiny and Regulation.
borrowed money has long been used to bribe the media
"AGI" is Decades Old and It Never Found a Viable Business Model or Real, Actually Useful Use Cases
I keep reminding people of local firms (right here in Manchester) doing the same thing the media now calls "AGI"...
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Saturday, August 22, 2026
IRC logs for Saturday, August 22, 2026
Microsoft to Its Workers in April-May: You're Too Old, Go Away. Microsoft in August: Young People, Go Away.
What does this company even sell anymore?
Clownflare Data US-Centric
We are assuming that for national security reasons not many sites in Chinese (or based in China) outsource their traffic to Clownflare
Gemini Links 22/08/2026: Broken Car, Devuan, and Thundermail
Links for the day