Bonum Certa Men Certa

Patent Quality at the EPO Has Collapsed Since Battistelli Came, According to EPO Staff

EPO grant rate
From an internal EPO publication



Summary: Norway highlighted strong concerns about declining patent quality at the EPO several years ago under the Benoît Battistelli administration; like Battistelli before him, António Campinos uses the propaganda mill "IAM" to say the opposite of what is true (IAM is also being paid by the EPO's PR agency), so today we reveal what actual EPO staff says about quality of European Patents

THE following publication is not new, but it is still very relevant because Team Campinos with its propaganda mill "IAM" lied a lot in recent weeks. We repost it below with no further comment (none is needed, the charts speak volumes)



Munich, 04.07.2018 sc18025mp – 0.2.1/1.3.3

Clarification – Grant Rate



Some colleagues approached us with questions about the grant rate at the EPO. In a recent Discussion paper on Patent Qualityi staff representation argued that the current grant rate is about 69% whereas the EPO management sees it at around 57% in 2017. During the September 2017 meeting of the administrative council, the German delegate raised concerns about the ever increasing grant rate, which they found had reached 70%ii. We decided to inquire by first seeking clarification from DG1. Unfortunately we did not receive an answer yet. So we looked at past press releases, articles and management presentation. We found that the grant rate at the EPO increased by 15 point from 42% in 2009 to 57% in 2017.

According to the IP5 Statistics Reportvii the EPO defines its grant rate as the number of published EP grants divided by the total number of completed EP examinations in a given year (#grants + #refusals + #withdrawals). Unfortunately, we were not able to determine the total number of completed EP examinations for a given year since the number of withdrawals and refusals are not available to us.

As we are not able to calculate the grant rate, we collected the grant rates for recent years by looking at press releases, management presentations and news articles: see Figure 1 and Table 1. Particularly notable is the rise from 50% to 57% from 2016 to 2017.

EPO grant rate increased from 42% to 57%; +15 points
Figure 1: Grant rate as communicated by the EPO management



Grant rate as communicated by the EPO
Table 1: Grant rate as communicated by the EPO (sources: 2007 and 2008iii, 2009iv, 2010 and 2011v, 2012vi, 2013 to 2014vii, 2015viii, 2016ix, 2017x)



Grant rates (as well as refusal and withdrawal rates) are of course a dubious indicator of patent quality. The underlying hypothesis is that the quality of the incoming patent applications remains stable over the years. However, this is unlikely, as it can be argued that the legal framework over the years improves. Furthermore, search tools for applicants to check available prior art before filing an application are much better today than some years ago.

Hence, whether the grant rate gives much indication about patent quality is open for discussion. However, the grant rate communicated by the EPO indicates an increase of 15 points in comparison with 2009. We think that this deserves a proper and transparent analysis.

Discussion Paper on Patent Quality

We are pleased that the Discussion Paper on Patent Quality finally triggered some discussion on patent quality within the management. In our view this is urgently needed. The statement that the grant rate increased from 50% to 70% was based on incorrect data; now thanks to new data provided by the management we are able to give more precise figuresxi. The result is that absolute values are different but the trend of an increasing grant rate is real: The grant rate increased from 42% to 57%.

Unfortunately, when investigating the grant rate we need to rely on press releases by the EPO, news articles and management presentations. We cannot directly verify the numbers. In particular, we could not find the yearly total number of completed EP examination proceedings for the years 2009 to 2017, i.e., the exact number of grants, withdrawals and refusals. Before 2009 this number was published in the Annual Report which the EPO publishes every year. We asked the EPO to provide us with the necessary data, albeit with no luck until now.

The error in the calculation of the Grant Rate in the Discussion paper on Patent Quality The main data sources for the grant rate as published in the Discussion Paper on Patent Quality were the Annual Reports of the EPO. The wrong assumption was that the total number of completed EP examinations in a given year corresponded to the number of “Examinations” as published in the Annual

Reports. For example for 2017 “Examinations” amounted to 153.858 (B) and the number of published grants to 105 635 (A), see Figure 2 and Figure 3, for 2017. Hence, a calculated grant rate of 0,69% for 2017 would follow by the division A/B. However, “Examinations” does not correspond to the number of completed procedures in EP examination, but to the total number of communications sent out by examiners. In the yearly report for the EPO every single communication counts as a product, as can be seen in Figure 4. Internally for examiners only final actions count as products, hence the error of interpretation.

Examinations at EPO
Figure 2: Examinations (Annual Report 2017xii , EPO Intranet)



EPO grants
Figure 3: Number of grants in 2017 (Annual Report 2017xii, EPO intranet)




Figure 4: Total products (Annual Report 2017xii, EPO intranet)



____ i Good enough? A discussion paper on patent quality, 2018, https://munich.suepo.org/archive/su18003mp.pdf ii CA/91/17, Draft Minutes of the 153rd meeting of the Administrative Council, paragraph 88, 24.11.2017, http://main23.internal.epo.org/projects/micado/micadn.nsf/198832a7132e4641c1256fcc002de3ed/e2ca77d3ad94c8cbc12581e20046cee0?OpenDocument iii European Patent Office takes a tougher stance on quality as European patent applications continue to rise, IAM, 2009, https://www.epo.org/news-issues/press/releases/archive/2009/20090317.html iv EPO patent grant rate plunges as the backlog grows, IAM, 2010, http://www.iam-media.com/Blog/Detail.aspx?g=76431827-efdb-430f-84d6-e595e6789902 v Low grant rate is an indicator of quality, says EPO president; UK lags however you look at it, 2012, http://www.iam-media.com/Blog/Detail.aspx?g=4d81a9f1-03bb-4438-b538-63928cd398a7 vi IP5 Statistics Reports 2012 Edition, https://www.fiveipoffices.org/statistics/statisticsreports/statisticsreport2012edition/IP5statistics2012.pdf vii The EPO defines its grant rate as “... the number of applications that were granted during the reporting period, divided by the number of disposals in the reporting period (applications granted plus those abandoned or refused).”, IP5 Statistics Reports 2014 Edition, https://www.fiveipoffices.org/statistics/statisticsreports/2014edition/ip5sr2014.pdf viii EPO achieves major performance gains in 2015, press release, 2016, https://www.epo.org/news-issues/news/2016/20160113.html ix EPO Quality Report 2016, https://www.epo.org/news-issues/news/2016/20160113.html x Management Presentation, May 2018, Our commitment to Quality, DG1 xi We approached the management prior to publication of the Discussion paper on Patent Quality to discuss and verify the data, but unfortunately to no avail. xii Annual report 2017, Granted Patents, https://www.epo.org/about-us/annual-reports-statistics/annual-report/2017/statistics/granted-patents.html#tab1


Aside from the above, the Munich SUEPO committee had also published at the time "Good Enough?"

It described it as "[a] discussion paper about patent quality at the EPO [...] drafted by colleagues of the local staff committees of Berlin and Munich. Until now the EPO administration refused to publish this document on the EPO intranet. We think this discussion paper should be read and discussed by as many EPO stakeholders as possible. That is why we publish it now on our website and we hope it will result in many fruitfull discussions amongst the EPO stakeholders."

At the time, Team Battistelli was crushing any attempt to study actual patent quality and validity (or compliance w.r.t. the EPC), scuttling any existing attempts to do so. We wrote about it back then. Here's the illuminating full paper, which is 31 pages long. [PDF]

Recent Techrights' Posts

Linux is Released Too Often, Tested Insufficiently (Same as Chromium, Firefox, and Systemd)
Driven by schedule, not quality (objective criterion)
When I discovered people trafficking in open source software
Reprinted with permission from Daniel Pocock
Over at Tux Machines...
GNU/Linux news for the past day
 
Debian: Let's Pretend We Never Knew Daniel Pocock
Ad hominem is what happens when the message is hard to dispute
DPL Sam Hartman proves blackmail is alive and well in Debian
Reprinted with permission from disguised.work
What is a safe space?
Reprinted with permission from the Free Software Fellowship
Does Debian deserve an independent news service?
Reprinted with permission from disguised.work
Linux.com So Neglected If Not Abandoned That It Promotes Deals That Expired 4 Weeks Ago
Quite some "stewardship" by the Linux Foundation
The Fall of Meritocracy in Tech
nuff said
Microsoft Has Lost Malta
Android has caught up
In Asia, Baidu Has Become Bigger Than Bing and Yandex is Getting There Too
XBox and Bing are going through existential crises
"Having IBM Next to Your Name is a Scarlet Letter"
IBM staff just motivated not to work
Techrights Browsing Made Easier
a draft for discussion
Links 12/05/2024: XBox Founders Say Microsoft Lost Its Identity
Links for the day
Gemini Links 12/05/2024: Enshitification and Mind Maps
Links for the day
Aside From Red Hat Spam and Partisan Media There's a Lingering Rumour of Layoffs
Some rumour said IBM had second thoughts about a WARN notice and delayed that a bit
The Albanian open source community is very healthy indeed
Windows nosedives from 99.1% to a lot less
Web Sites Hijacked by WIPO on Behalf of Microsoft-Sponsored SPI (and People Looking to Hide Embarrassing Facts)
debian.chat; debiancommunity.org; debian.day; debian.family; debian.finance; debian.giving; debiangnulinux.org; debian.guide; debian.news; debian.plus; debianproject.community; debianproject.org; debian.team; debian.video
Julian Assange on Privacy of People, Even Little Children
Facebook/Google (or GAFAM, an acronym I coined with Assange) knows you better than your mom knows you
[Meme] Miscomprehension of GDPR
Social control in general is a ticking timebomb
In Haiti, the Market Share of Windows Collapsed (From 97% to 27% on Desktops/Laptops)
A couple of months ago Windows was measured at 3.04%
In Most Countries It's Still Possible Not to Have a 'Smartphone' and to Pay for Nearly Everything With Cash
Withdrawing money will be possible as long as enough people use many ATMs (cash machines)
Expect Lots of Material From Daniel Pocock as Election Day Nears
The experiences of Daniel Pocock were an excellent example of reprisal or retribution against either whistleblowers or people who give a voice to whistleblowers
I've Been Promoting Free Software for Over 25 Years
I wrote my first computer program when I was about 14, maybe a little younger (I have visual memory of it)
Reminder: Richard Stallman's Talk is This Week in Paris (and in French)
Defending rms isn't the same as defending everything he has ever said
IRC Proceedings: Saturday, May 11, 2024
IRC logs for Saturday, May 11, 2024
Online Bullying (Trying to Make People Unhappy)
Narcissists and bullies behind mice and keyboards, no honesty or fact-checking required
Talk About Software Freedom
"Linux" and "BSD" may mean a lot to more and more people, but they're still just brands or acronyms
Windows in South Korea: From 98.5% in 2010 to About 30% (Android Rises to Almost 50%)
Samsung ships like a million Linux devices per day
Improving Site Navigation for Easier Discovery and Catch-ups
This site is run by code we wrote ourselves
LibrePlanet 2024 Recordings
Let's hope independent recordings by viewers can help recovery of "lost talks" (recordings)
GNU/Linux Reaches 11% Market Share in the United States Of America - an All-Time High
The United States Of America is where the operating system started (Boston) and where Linus Torvalds works (Portland)
[Meme] Being Believed, Not Censored or Defamed
Daniel Pocock, Zini, and John Sullivan (FSF)
Links 11/05/2024: XBox Crisis, Spotify Exodus Continues
Links for the day
Gemini Links 11/05/2024: Why to Delete GitHub
Links for the day
In Europe, Bing Fell Every Month This Year, Lost a Considerable Share Since "Bing Chat" and All the Chatbot Hype
Microsoft's Bing has had many layoffs lately
Links 11/05/2024: Analysis of the Microsoft Crisis and Backdoor-Looking Bugs
Links for the day
Attacking the Messenger?
Stack Overflow and LLM licencing
Microsoft Fired Loads of Staff in Kenya, Which is Another Large Country Where GNU/Linux Has Grown a Lot
Microsoft pays Kenyans only 2 dollars an hour for an IT/office job
Knowing the True History of Debian, Owing to Irish Debian Developer Daniel Pocock (Currently Running to Become Member of the European Parliament)
Irish-Australian and scapegoat of a highly dysfunctional 'Debian family'
Attacking by Credentials
Modest people do not demand fancy titles
Microsoft Windows Used to Have 99% of the OS Market in Jordan, Now It's Just 13% (Less Than iOS)
Based on the data of statCounter, GNU/Linux in Jordan climbed from 0.62% in May 2014 to nearly 5% right now
More Nations Are Reaching and Exceeding 5% Market Share for GNU/Linux, Microsoft Wants to be Bailed Out Again
Microsoft is once again reaching out to Biden for a bailout - a subject we'll cover in a video some time this weekend
Over at Tux Machines...
GNU/Linux news for the past day
IRC Proceedings: Friday, May 10, 2024
IRC logs for Friday, May 10, 2024
[Meme] What Do You Call a Woman Who Does BDS on Free Software? Elana Hamasman.
Here are some confused thoughts
[Meme] Mission Aborted
Mission Aborted: cancel RMS
Taking Things Up a Notch
we strive/aim towards 15-25 new pages per day, i.e. around 500 per month or 6,000 per year