Sunday, April 18, 2021

e-EQE – epi discussion paper - comments

Update 21 Jun 21: I have also made some of these points in epi Information 02/21
8 Jun 21: forum link added.
The epi recently published their discussion paper on proposed changes to the EQE from 2024. See below for some comments, following on from my earlier comments on the future of the EQE.  It is also possible to discuss using the epi-learning forum (open to all).

My conclusions: 
  • Some good ideas, like allowing an earlier start of learning, but I miss any proposals to improve the organisation. 
  • I miss a proposal for permanent education - the EQE tests whether a small number of representatives are fit to practice at that moment. 
  • I hear more and more candidates choosing to follow a law degree instead of the EQE because of the unpredictability of passing. Having everything modular will help prevent this getting larger.
    • But we also need a clear syllabus for each part, transparent marking and a proper way to address the concerns of the average non-native speakers.
  • Infringement has to be added - it is an essential part of being a patent attorney.
My comments:

1. Is the current EQE adequately testing the “fit to practice” criterion? - p.1
  • Some Exam Papers are ... jigsaw puzzles ... rather than testing real skills ...
    • It is not a problem if questions are designed to have one answer - the legal questions are set up like this. Having small puzzles that fit together, like the current Pre-Exam - claim analysis and D2 does mean that you can feel when you are on the right path.
    • However, the general advice to pass currently is to not think too much and give the answer in the Guidelines because alternatives are almost never accepted, even if properly argued. Perhaps that way that the questions are structured needs to be clarified - instead of asking, is it novel/inventive/extension etc, you can ask whether the EPO will object (the Guidelines answer) - candidates should know what will happen during proceedings.
    • The biggest current "puzzle" is the C exam - it is completely out of control (not just this year). There are too many different documents and the pieces you need are not always in the most likely places. Way too long, too many documents, with a ridiculous amount of detail require for argumentation. 
    • The B exam is now going the same way. 
  • All papers should start with 0 marks, and award marks for correct parts. A and B suffer the most because they are negatively marked - you start with 100 marks, but lose marks for each deviation from the expected answer. The problem is that by misreading or misunderstanding a couple of things, you end up running out of materials to use. There is a randomness to passing - if you are on the same wavelength, you pass. If not, you fail. But if you retake it a year later, then you can pass without preparing any differently.
  • Marking of the practical papers should always consider whether a mistake would be correctable in real-life 
      • Why should you be afraid for A of submitting a non-novel claim or non-inventive claim? As the argumentation supplied tends to be brief for A (as it reflects the brief argumentation typically included in a patent application), you should not lose so many marks if a dependent claim is what they wanted. There should be no marks lost for extending a feature beyond the scope of the client's letter, for example.
      • For B, where argumentation is required, the reasoning can be judged better for possibly having a non-novel or non-inventive claim. There should be no marks lost for not abandoning scope based on an instruction by the client if you think it is patentable (and argue this). 
      • Why is matching EVERY feature and correctly defining closest prior art in C such an issue? For an opposition, the main thing is to attack all the claims, get all the documents in, and argue the non-trivial aspects of the feature matching and inventive step. You would not write an opposition like that.
  • Methodology courses - p.1, p.2
    • I may be a little biased ;-), but I am not sure why it is a goal to get rid of these. Most of the methodology courses are based on reverse engineering and save candidates a lot of time that they can use for actually practicing papers. I don't think these will disappear, even with a more modular exam and shorter pieces. 
    • These are currently "needed" because the exams are much too long (so very difficult to do in the time available, especially for non-native speakers), there is no official info about what is to be tested, how the papers should be made, or how they should be marked. The Examiners' Reports are not easy to read and deliberately leave out a lot of details on alternatives that were accepted, or the exact marking. Even if candidates file an appeal, they will never see the details of their marking to see where they lost marks. 
    • If there are tricks that is being taught to pass with little preparation, obviously this should be prevented. But I don't consider pointing out where marks are awarded and how to score those marks as a trick - all marks count equally to the scoring, and candidates should always focus on their strengths and understand what is required. I don't see weak candidates passing just through methodology - more that there are a lot of well-prepared candidates who fail because they do not know how to show their working properly within the time available. Hardly anyone ever gets about 80 marks for certain exams.
    • I agree that it should not be necessary to follow such a course to pass. The committees and boards can influence this already => a clear syllabus that they stick to, openness about the marking and what is expected, using similar formats and styles each year, announcing new subjects to be tested or changes to the marking in advance, publishing several alternatives that were accepted, allowed appealing candidates to see exactly where they lost marks etc.
  • Practical training
    • This varies greatly, depending on where you work and the opportunities available - there should be more guidance on what you should be doing and some help with this. 
    • I agree that that it is better to prepare earlier and not make exam participation dependent on a fixed number of years - for example, a lack of drafting and amendment experience can be compensated by doing more old exams to learn the skills needed. You need the knowledge as early as possible.
  • Suggestion - allow qualified attorneys, lawyers, employees, formalities officers etc. to also sit the modules (without overwriting their scores) to allow them to stay up to date in a structured way. This could already be done for the Pre-Exam (no consequence - they just get marks separately for legal and claim analysis).

Sunday, April 11, 2021

Answering D1 Exam 2021 with EPC.App & PCT.App

While doing the D Exam as an official benchmarker, I kept track of how I used the EPC.App and PCT.App references to answer questions. I did this to check usability and completeness. Below are notes on how and where I found the main legal issues, so there are spoilers. The D Exam can be downloaded from the EPO Compendium. See also an earlier post with more about D Exam technical issues
I have not included my answers - these were made under exact exam conditions, so there are plenty of mistakes. I have concentrated on D1 use as you need to find things quickly. My main conclusions:
  • Both books contained the legal info needed, and allowed it to be found quickly. Exception was Situation B of Q.5 - although this is unlikely to be asked again for many years, it can be important for real life. R.26.3ter is included in PCT.App, but we need to cross-reference that in other sections.
  • The fastest way to search in EPC.App and PCT.App is to use the Detailed Contents. Based on your knowledge and the point in the procedure, you should be able to quickly see the most relevant provisions.
  • The fastest way to answer PCT questions is to look for an overview of the procedure at the beginning of many sections.
  • It is sometimes quicker to search directly in the Guidelines / Applicants Guides as not everything can be neatly sorted under the structure of a legal reference book.

The set of references I used was:
  • EPC.App (IP.Appify) => EPC and Euro-PCT legal provisions, case law (Nov 20 - standard version from Amazon with the editor's annotations)
  • Indexed EPO-EPC Guidelines (Fireball Patents) => EPC implementation by EPO, case law (1 Nov 19)
  • National Law and Validation (Fireball Patents) => National Law tables (implementation by EPC states) and validation agreements (31 Oct 20)
  • EPO as a PCT Authority (Fireball Patents) => Euro-PCT and EPO's PCT implementation (31 Oct 20)
  • PCT.App (IP.Appify) => PCT and priority part of Paris Convention legal provisions (Sep 20 -  standard from Amazon, but printed with color mark-up made using Print on Demand option)
  • Complete PCT Applicant's Guide (Fireball Patents) => PCT implementation by all offices (1 Sep 20)
  • PCT References - Part 1 & 2 (Fireball Patents) => Most relevant PCT annexes (implementation by PCT states) and overviews (31 Oct 20)

To find things quickly:
  • During study, you need to learn what each book contains and where/how to find it. For example:
    • EPC.App is arranged according to Articles in numerical order, together with the relevant Rules. It has a Detailed Contents with the Articles, Rules and other contents like G-Decisions - this is the best tool to quickly scan through the contents if you are not sure where to look. There are also an Index of Legal Provisions to directly find specific Rules, and an Index of Decisions to directly find G, J , R, T & W case law. In general, the Articles and Rules of the EPC are arranged in chronological order as an application proceeds from filing to grant.
    • EPO-EPC Guidelines are arranged in approximately chronological Parts as an application proceeds from filing to grant. Substantive issues, such as amendment and patentability, are arranged per subject. The Detailed Table of Contents is the best tool to quickly scan through the contents. In most other cases, you will be following a specific chapter or section link from another reference, such as OJEPO or EPC.App
    • PCT.App is arranged according subjects in approximately chronological order, mirroring the structure of the WIPO Introduction to International and National Phases. It has a Detailed Contents with the Articles, Rules and other contents like Admin Instructions - this is the best tool to quickly scan through the contents if you are not sure where to look. There are also an Index of Legal Provisions to directly find specific Article and Rules. In general, the Articles of the PCT are arranged in chronological order as an application proceeds from filing to Chapter I entry, and from examination to Chapter II entry.
    • PCT Applicants' Guides (Introduction to International and National Phases) are arranged in approximately chronological order as an application proceeds from filing through search, publication, examination to national entry. The Detailed Table of Contents is the best tool to quickly scan through the contents. In some cases, you will be following a specific chapter or section link from another reference, such as PCT Newsletter or PCT.App.
  • To decide on which reference to use during the exam, determine: 
    • EPC, PCT or Euro-PCT?
    • More likely to be found under legal provisions or implementation?
    • If more than one, which will be quickest to find?

My approach to D1
  • I copied each question completely into the answer sheet, and only used occasional highlighting (this can waste a lot of time and does not generate marks). I left it in my answer just to have a copy immediately after the exam.
    • CTRL-C and CTL-V (Win 10) were sufficient to properly copy/paste all D1 questions except Q.4 (needed to manually remove formatting and tidy up lines)
  • I use 3 minutes per mark as a rule of thumb to quickly calculate how much time to spend. Traditionally, the D Exam was 330 minutes (3.3 mins/mark). This year, an extra 30 mins was given to compensate for the format (3.6 mins/mark). It is normal to run out of time answering D1 questions.
  • There are no negative marks and no requirement to fully complete everything you start - you want to generate enough to get 55 marks or higher. Get the marks first for the things you know well - if necessary, skip a part of a question until the end.
  • There is no time to read all the questions first, so I read each question in order as if I am going to answer it completely. If the question seems likely to take a long time to answer (or I have no idea), I skip it and go back to it at the end.
    • it is very difficult under pressure to read, understand and apply something you read for the first time (or it seems like the first time)
    • even if you think know the issues or can find them, you don't want to waste too much time 
  • I identified EPC, PCT and/or Euro-PCT parts while reading
  • I copied the last part of the questions ("the explicit questions") as a rough framework for my answer to make sure I didn't forget anything.
  • I made a rough vertical timeline by copying phrases and putting them in chronological order (no need to make it nice - just understandable during the exam). I avoided editing by sticking close to the exact wording, although I did reduce some long sentences. E.g.: 
    • -- EP-A1 was filed in 2018  A in English - X and Y, and claims invention X only
    • --  A.94(3) issued by the EPO on 5 October 2020
    • -- received by applicant A on 8 October 2020
    • -- No response has been filed
    • -- January 2021, he filed a European divisional application EP-A2 based on EP-A1 in French. claiming and describing invention Y
  • For some issues I know well, I have an idea how to form the answer and select the argumentation to use. For others, it can be less clear how to proceed, so I just to start working on pieces (like calculating time limits) until a bigger picture emerges. In most cases, the overall answer to the question about whether something is possible or not is added at the end.
  • Main part of answer includes identifying the correct law (legal provisions, case law and/or implementation) by citation, applying it step-by-step to the facts given and any other facts you can derive, and giving a conclusion as an answer (yes/no/yes if.../no).
  • Be conscious of where you are in the procedure - certain conditions must have already been fulfilled (and thus are less relevant to discuss). Focus should be on the next few weeks and major actions required in the future. Assume that the case takes place on the day of the exam - here 2 Mar 2021. 
  • At the end, if time, I check whether there are still facts from question that I have not used - these may need to be included somehow. But I am always trying to get to the next question, and often decide not to address these if I have already answered the question asked.

Saturday, April 10, 2021

Lessons learned from e-EQE and looking to the future

Updated 18 Apr 2021: see also my post commenting on the epi discussion paper.
Updated 12 Apr 2021.
"It is only when the tide goes out that you see who's been swimming naked"
I already posted these comments on LinkedIn - this post gives a little more background. There was an enormous effort behind the EQE scenes, and a lot of people working many extra hours, but this could not prevent cracks appearing. 
On reflection, I think a few long-standing EQE issues may have played a role in creating the problems:
  • No effective communication with most candidates. Email is fine for sending standard messages, but there is little chance to request additional information. It is not always clear whether candidates actually receive them and understand the intention behind the emails.  A lot of candidates are surprised by changes in exam formats as not everyone prepares using a tutor, or with other students to discuss things.  
  • Secrecy and little transparency about every aspect of the exam and organization. Seen particularly by candidates who fail - it is very difficult to figure out where marks were missed, why the official answer was preferred, what alternatives were available etc. But this year, it was promised that the exams had been adapted to take into account the new format, but it was refused to provide any details as this was related to the content of the exams.
  • Just enough volunteers to make and mark the exams. No one expects perfection -  limited resources means that not everything can be addressed. But it is a professional exam, so it should be possible with sufficient preparation to have a reasonable expectation of passing. There is a minimum level that should always be achieved.
  • Few real limits for each paper in what can be asked. A real problem for many candidates is the habit of regularly introducing new things not announced before or tested before. It is unheard of in any other professional exam not to have an official detailed syllabus of what you need to know, both for the legal parts and the practical subjects. The REE is too vague to be useful, so candidates prepare based on unofficial study guides (like CEIPI or EQE Guide for Preparation) and doing old exams. There is still no clear distinction between legal subjects tested at Pre-Exam level and Main Exam level.
    And it is ridiculous that every year candidates need to have 2000 pages from the WIPO in case a question is asked about a PCT state (in many years, the WIPO and National Law are not needed at all).
  • Few real limits for each paper in length and style. Since 2017, an extra 30 mins was added to each Main Exam so that non-native speakers have more time. But there is no clear way in which this is enforced, and it is clear that the time has been eaten up by the exams. 
  • No effective way to assist those with En/Fr/Ge as a 2nd language. Non-native speakers have a clear disadvantage. The patent attorney terminology manuals are very useful, but reading/typing exams can still be slower. And giving everyone an extra 30 mins has now lost its effect.
Normally, these are compensated by a lot of flexibility in the marking, and that will be repeated this year. So, anyone well-prepared who generated a reasonable answer has a good chance (as in all other years). But moving forward, some suggestions of possible improvements:
  • Make e-EQE improvements highly visible, so that candidates know what to expect (platform + content) and can properly prepare. 
  • Using the new possibilities, better define the content to be tested and make sure it is relevant to daily work. 
  • Have a proper official syllabus for each exam - articles and rules, sections of Guidelines, case law, OJEPO's. 
  • Limit detailed PCT questions outside Europe to IP5 countries, and provide resources electronically.
  • Remove overlap in testing and preparation (candidates & committees). For example:
    • Pre-Exam: limit to only only legal questions, covering procedural and substantive law. Drop Claims Analysis (covered extensively in A, B, C). Use a mix of multiple choice and "D1"-style questions to test D1 level of knowledge. This is main test of legal knowledge.
    • Allow ME: A and ME: B to be taken anytime - either with Pre-Exam or Main Exam. These are practically oriented, relating to skills which are being developed early (drafting and amendment).
    • ME: A can be cut in half => reduce subject matter, focus mainly on independent claims,  provide less prior art.
    • ME: B is ridiculous with client giving you claims. Go back to old format, cut in half => reduced subject matter, only include a few claims, provide less prior art.
    • ME: D - limit to current D2 (D1 is now in Pre-Exam). But also include some basic trademarks/designs and basic trade secrets. Test legal concepts, and not substantive patent law (or greatly simplified substantive law).
      Expand basic foreign knowledge (JP, US) to include CN, KR (IP5). Also define clearly what you need to know. Possibly also test something about those foreign national patent systems. Knowledge of DE, FR and UK systems would also be useful.
    • ME: C has much repetition in attacks. Cut in half => fewer claims, provide a lot fewer documents. There should also be a proper infringement part in EQE - it makes most sense to add it here as technical understanding is important.
Perhaps also look even wider, putting the EQE in a broader perspective. A lot of effort is put into only passing candidates who are fit to practice, but there are many other people who interact with the EPO at different levels:
  • Permanent education required for qualified attorneys. For most, a lot of the knowledge is lost within a few months of taking their last exam. I know there were some attempts in the past to introduce this, and I know that no system is perfect, but something must be better than nothing. This would also address the issue of the varying knowledge levels of grandfathers \ grandmothers.
  • European formalities officer qualification. These are the people who have the most up-to-date procedural knowledge in the offices, but the levels of knowledge vary wildly. 
  • European patent agent qualification. Similar to US, where a basic knowledge of procedural and substantive law needed for those interacting with the patent office. This would be useful for people who do not wish to become patent attorneys (yet), or work in a company's patent department, or work in a lawyer's office.