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

EPO Strike a Week From Now, After That Strikes Can Become Permanent
A week from tomorrow there will be another strike
Your Site Should Implement Its Own Search (Before It's Too Late)
GAFAM was never trustworthy
 
Streisand Effect and Justice
This weekend this site has served over 8 million Web requests
Gemini Links 22/03/2026: "Woman of Tomorrow" and "First Steps in Geminispace"
Links for the day
SLAPP Censorship - Part 19 Out of 200: They Were Ill-prepared for Tough Questions in Cross-Examination
Very ill-prepared for the deteriorating situation caused by their clients' past behaviour towards many people, including high-profile figures who offered to testify
The Media Sold Out to Slop Bros
If you wish for the hype to stop, then stop participating in it
The Only Non-IBM Staff in Fedora Council/Leadership Attacks Booting Freedom (Just Like the Master Wants)
Last week IBM laid off almost 1,000 people in Confluent and the media didn't write anything about it, so don't expect anyone in what's left of the media to comment on Fedora's demise and silent layoffs at Red Hat
Just Like a Founder of XBox Said, Microsoft XBox is Collapsing, Management Continue to Jump Ship
Nowadays Microsoft tries to promote this idea that Windows is XBox and XBox is Windows
Links 22/03/2026: Slop Triggers Emergency at Meta, Energy Prices Rise Sharply
Links for the day
Links 22/03/2026: Microsoft 'Open' 'AI' in Legal Trouble (Plagiarism, Distortion, Misrepresentation); Facebook/Meta Kills Off "Horizon Worlds"
Links for the day
Racism Dressed Up as "Choice"
Racism is rampant at IBM
Probably an All-Time Record
Our investment in our own SSG is paying off
Gemini Links 22/03/2026: LLM Slop Attacks USENET, Announcing Pig (New Game in Gemini Protocol)
Links for the day
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Saturday, March 21, 2026
IRC logs for Saturday, March 21, 2026
SLAPP Censorship - Part 18 Out of 200: Third Parties Funding Attacks on the Messengers, Lawsuits Against GAFAM-Critical Voices That Uphold Real National Security
Women are like kryptonite to them
Never Trust People Who Write Their Own Wikipedia Pages (Vanity Pages About Themselves) or Ask Friends to Do So. Also: Jono Bacon is Married to Microsoft.
We'd hardly be the first to point out Wikipedia isn't what it seems
No Tolerance for Attacks on Family Members
Being a Free software activist ought not lead to "collateral damage" like attacks on family members, including doxing
Sirius Open Source is Just a Zombie Firm With Shell Entities
Many companies fake their health and their size
Communities Can Only Survive When Trust Prevails
PCLinuxOS is still a vibrant and authentic community
Techrights Was Always a Community Site
The harder we're attacked, the more people participate in the site
Maintenance Reminder
We'll carry on publishing
Behind the PR Smokescreen and Microsoft-Sponsored Chaff, Microsoft Layoffs in "AI" Alleged This Month
In an age when ~1,000 simultaneous layoffs aren't enough to receive any media coverage, what can we expect remaining publishers to tell us about Microsoft layoffs in 2026?
EPO "Cocaine Communication Manager" - Part VIII - Mobbing and Silencing of Dissenting Staff
that's the very cornerstone of functional democracies with real opposition parties
Bluewashing at Confluent: Some Workers to Leave Within 3 Months (IBM Mass Layoffs)
Is the "era of AI" an era when none of the media will mention over 800 layoffs? [...] There's a lesson here about the state of the contemporary media, not just IBM and bluewashing
Microsoft OpenAI, Drowning in Debt and Forced to Make Significant Cuts (as Reports Reveal This Month), Does Hiring Disguised as "Takeovers" to Fake Value or Alleged Potential
Remember what happened to Skype last year
Reader Shares Recent Memes on Slop and 'Coding' by LLMs
"just some funny memes I thought were relevant to current coverage."
Slop Does Not Replace Art, It Contaminates Everything With Reckless Nonsense
many Computer Scientists do not want programs to get contaminated by slop
Coders Don't Just Reject 'Vibe Coding' Because They're "Luddites", They Just Know the True Cost of Slop
if some programmer says slop sucks, don't rush to assume selfishness or defence of one's occupation
When Nobody Else Covers the News
There's an obvious "media blackout" regarding the mass layoffs
Links 21/03/2026: David Botstein Dies, Slop as Censorship Apparatus
Links for the day
Links 21/03/2026: Metastablecoin Fragmentation and Crescent Moon
Links for the day
Gemini Links 21/03/2026: Historic Ada Docs; The Lurking LLM on the SmolNet
Links for the day
HSBC the Latest Failed Bank Using Slop as Excuse for Its Financial Failure
"HSBC is planning on cutting as many as 20,000 jobs in the near future as the company allies with AI revolution."
Invitation to General Assembly After 1,200 EPO Workers Participated in the Demonstration 3 Days Ago
"the strike of 19 March was also very well followed."
A/Prof Susan G Kleinmann, Enkelena Haxhija & Debian-private risk to MIT
Reprinted with permission from Daniel Pocock
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Friday, March 20, 2026
IRC logs for Friday, March 20, 2026
SLAPP Censorship - Part 17 Out of 200: A Long Track Record of Online Abuse, Then Choosing a Low-Cost Law Firm to Muzzle People Who Have Illuminated This Abuse for Over a Decade
Censorship by targeting ISPs and webhosts isn't unprecedented
Plagiarism in "Linux" Clothing (LLM Slop in linuxiac.com, LinuxTeck.com, and linuxsecurity.com)
The net effect of those slopfarms is very negative
Links 20/03/2026: Facebook Weaponised Politically, Openwashing by LF and NVIDIA, Encyclopedia Britannica Sues Microsoft Proxy for Plagiarism
Links for the day
The EPO's Local Staff Committee Munich (LSCMN) Explains to the Administrative Council (AC) How Bad Things Have Become at Europe's Second-Largest Institution, Biggest Patent Office, and Corruption/Cocaine Hub (Jobs Sold to Friends)
We'll say a bit more tomorrow
IBM's Red Hat Diversity: Only 3 Women (Out of 11 Leaders)
For comparison's sake, the FSF is about 50% female
Symptom of Publishers Dying: They Move to Adopt Slop. Symptom of Software Companies Dying: They Move to Adopt Slop ('Vibe').
It'll always fail. It's hype. It's a bubble.
Under IBM, Red Hat Replaces Code With LLM Slop, Fedora is Slopware
Not even hiding it, those things are in plain sight
Gemini Links 20/03/2026: Depictions of Culture and The Social Smolnet
Links for the day
SimilarWeb Was Never a Reliable Yardstick for Traffic
5RB may need some "house-cleaning"
Strangulation, suffocation, Jonathan Carter & Debian toxic culture confirmed
Reprinted with permission from Daniel Pocock
Reports or Hearsay Suggest Ogilvy Broke Up With IBM and Insiders Report Mass Layoffs in "Infrastructure" (Might Impact Red Hat Entrants)
hearsay in Social Control Media
Scheduled Server Maintenance Tomorrow Night
Starting 9PM
None of the Above (NotA) & Debian snubbing Sruthi Chandran
Reprinted with permission from Daniel Pocock
Links 20/03/2026: Cryptography Pioneers Win Turing Award and BMG Sues Anthropic for Copyright Infringement
Links for the day
Even Uganda Understands That Journalists Never Belong in Prison
"Ugandan authorities must respect the spirit of this ruling and abandon any measures that seek to jail Ugandans for the free flow of ideas."
Inaction Helps Your Enemies
Without freedom, there's nothing else left
Windows Down From 99% to ~50% in Republic of Seychelles (République des Seychelles)
Windows fell by a lot
"systemd is essentially a corporate IBM/Redhat project and corporations of course will comply"
Microsoft and IBM care about users' freedom like Cheeto Lump cares about the US Constitution
Confluent Insiders: IBM Laid Over Over 800 at Confluent, Not Just 800
For the record, the layoffs at Confluent won't be over. After the bluewashing there will be "IBM RAs" impacting Confluent folks, aside from PIPs
The Layoffs at IBM Carry on (Shades of Enron)
Is IBM another Enron?
"IBM boss Arvind Krishna... financial package valued at $38 million in calendar 2025 - equivalent to the average collective pay of 765 Big Blue workers."
continues to ruin the company to enrich himself while pretending he has a strategy
Gemini Links 20/03/2026: Digital Identity Bifurcation and a "Return to Gemini"
Links for the day
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Thursday, March 19, 2026
IRC logs for Thursday, March 19, 2026