Thursday, July 5, 2018

EQE2018 Examiner Reports with expected solutions

For those who want to practice the papers, or to analyse their own performance, the EQE 2018 compendium is now almost complete:

Examiners Reports show the expected solutions and some of the alternatives that were accepted:

For anyone who unfortunately failed, a marking sheet will be provided with the official results letter, showing the number of marks per question awarded by each marker. 

Candidate's Solutions are usually added after the summer (one in each language). These are usually someone who scored about 80 points. In most cases, they are made by a native speaker who wrote a lot, so don't consider these to be typical. But they can be a good source of terminology and phrasing.


  1. Hi Pete, what do you think of this statement for Paper B:

    "In the following example the definition by a functional feature even leads to a lack of inventive step: “the cover layer being configured to soften and flow”. In this regard the glass layer disclosed in D2 is inherently “configured to soften and flow”; in this case there is a lack of inventive step over the prior art D1 and D2 (-10 marks)."



  2. It states in D2 that the glass ruptures due to the energy. The comment on p.11 of the Examiners Report seems to be reading too much into D2. I guess theoretically glass could soften and flow - according to Google at 1400 to 1600 degrees C. I have dealt with similar cases in practice, and this would normally be seen as a hindsight reasoning, although it would depend on the kind of temperatures reached when the fuse blows (which we are not told). They expect you on the exam to only use general knowledge, which is that glass is a solid which melts when heated. If you consider it like that, then their statement is correct. But the Examiner's Report also explains the differences (p18 & p19) as supporting the fact that D2 cannot be the closest prior art, which seems slightly contradictory.

    1. IMHO, for inherency they need the statement "ALL glasses soften and flow upon heating" to be the common general knowledge. This seems very unreasonable to me.

      The statement that "ALL glasses soften and flow upon heating" may also be factually incorrect. Firstly, some glasses harden rather than soften in certain temperature ranges (e.g. fused silica glass: Secondly, one can suspect that some glass can decompose, like dry ice, into gas, and thus stop being glass, instead of flowing.

      And what do you think of the Art. 123(2) EPC objection (intermediate generalization) to the same feature?

  3. I am not sure - cover layer is only mentioned as "cover layer made of epoxy resin" in paragraphs [008], [009], [011], in the table 2. Based on that, it would be extension. But the claims are also part of the application as filed, and claim 3 is only directed to a cover layer. Claim 4 is dependent (mentions epoxy resin). If claim 4 had referred to claim 3, then that would be sufficient support for epoxy resin being an option (but that was not the case - it incorrectly referred to claim 2). I think they are correct in this case.

    1. I am not sure which part of the intermediate generalization test is not fulfilled. Is it possible to perform the test without deciding on the level of the skilled person?

      I had an argument that the level of skilled person in this art is clear from D3, par. 2, which states "The expansion layer 305 is formed of a material which greatly expands when subjected to heating", but does not specify even one possible material. That is, the skilled person has such level that he would be able to find material by the function.

      In the application, the function of epoxy is to soften and flow into the gap. When the gap is filled, it does not exist anymore, and without the gap, there is no metal reflow (see par [002] "Metal reflow is the phenomenon whereby the metal of the fuse track of a blown fuse flows back into the gap between the electrodes ...").

    2. Intermediate generalisation is based on the technical disclosure of the application as filed - what does the skilled person understand using common general knowledge. D3 is a just An A.54(2) document, so you cannot use it to interpret the technical disclosure of the application.

    3. I did not mean that whole D3 is a part of the common technical knowledge. Rather, I meant that the level of the skilled person must be somehow estimated, and in D3 finding a material by its expansion property is treated as common general knowledge. That is, while we have not received a handbook with a table of expansion coefficients for different materials as "A" publication from the Examiner, we can conclude from D3 that such a handbook exists, and materials, which greatly expand, can be found in it. By analogy, a handbook for materials which soften and flow upon heating exists as well.

      All right, I admit that this is not likely to bring me any marks :) Thank you for the discussion!

    4. If the "cover layer made of epoxy resin" had been in the claim, then I think it would have been clear. By putting just "cover layer" in a claim, this would tend to suggest that a broader protection is sought than just with "epoxy resin". And a broader claim may be supported by a single example if the skilled person can extend the teaching.
      Unfortunately on the exam, you don't have time to think about this too much - write the claim, write a defence and carry on. If your claim 2 is their claim 1, and the rest is done, you should still be able to pass.

    5. I thought that I would pass, since my claim 1 seems to be rather close to yours. Indeed, you wrote in comments to one of the previous posts that you put SRI in claim 1 instead of quality and added the cover layer configured to soften and flow. Also, it followed from your comments that you did not add epoxy. One difference between our claims may be in the AlCu feature: I added the range of 10-20%, while your comments have not mentioned AlCu yet. Another difference may be in the SRI feature: I formulated it as SRI of not more than ABOUT 5 (because 5- corresponds to 62+ and we needed to cover 60), while your exact formulation is not known to me.

      However, I got 35 marks, and it seems from the Report that I could lose between 10 and all 20 points for claim 1 for "claiming the result", "lack of inventive step", and "violations of Article 123(2)". Also, I surely lost 4-5 for not "improving" the dependent claims dependencies. I think that the Examination Board is surely incorrect on the inherency issue and that it may be incorrect also on other issues. So, I am considering filing an appeal. This is pretty difficult, but I have already passed Paper C of 2017 this way (got additional 4 marks, and 50 in total).

  4. I put in SRI less than or equal to 5. I did not use about as that is inherently unclear before the EPO. The Examiners Report states that no range has been disclosed, but I don't agree. The range in quality was given and I used the tables as a way to convert. I assumed that was why they gave you the values in the tables. I don't agree with SRI being a result to be achieved - the tables give several examples showing how the percentage Cu may be modified.
    I put AlCu in my claim 2 with 10-20%.

    With regard to the cover layer that must be of epoxy resin, it seems like they are ignoring the contribution of the claims as filed (claim 3 and claim 4) to the scope of interpretation. This is something I see in practice in real office actions - they just look at the description as filed.

    1. I tend to disagree in part: F-IV, 4.7 actually allows using "about", despite that it sets some conditions. Indeed, it states "Particular attention is required whenever the word "about" ...[is] used. Such a word may be applied, for example, to a particular value (e.g. "about 200°C"), ...". However, I have not analyzed these conditions in the present case.

      I referred to both claims 3 and 4 in my Paper B. The Examiner's Report does not mention them in the context of the intermediate generalization objection. So indeed they seem to have ignored them.

      Of course, I disagree with the Examiners' conclusion that "no range is disclosed", and that "the SRI is the result to be achieved".

      Also, I do not see a legal basis for 4 marks available for "improving" the dependencies of dependent claims. This is not required by the general rules of practice, and there is no instruction from the client, as this would not broaden the protection.

      However, I can only appeal, and most likely lose time and money without achieving anything, while the potential benefit is only some correction.

  5. I will have to check previous exams - I don't recall improving the dependencies being awarded points before. This does not change the scope of protection, but does improve the strength of the claims as fall-back positions. In real life, I don't really spend much time on this (only if there is an error), but maybe there is a good reason to do it? I know there used to be some countries that only allowed amendments to dependent claims during invalidity case (not allowing something from the description to be added or allowing claims to be split), but I am not sure if that is still the case.

    Regarding appeal, it only makes sense if think you have a good case, and you will have at least 45 points after any correction. Even then, the decision will probably not made until after the next EQE. The chances of winning on substantive issues are quite small. There are two stages - evaluation by the Examination Board to see if they wish to revise their decision (interlocutory revision). If not it is sent to the Disciplinary BoA - it is this second part which takes a long time.

    I will do the paper again in the next couple of weeks - it is possible I missed some comment or misread something. I advise you do the same, and try and figure out why you did not see the other solutions and why you did not follow the directions which seemed to have been followed by most of the other candidates. This will help prepare for next year's paper.

    1. I saw the Examiners' solution very well and spent between 40 and 50 minutes choosing between the two approaches. Of course, it was clear that "our" approach was riskier, but at the same time I did not see any likely non-resolvable issue with "our" claim, and the Examiners' claim 1 seemed to me too narrow to be useful. Importantly, the client did not refer to epoxy as needed for the effect, but in contrast discussed AlCu at length. Also, the client knew that the features could be imported from a dependent claim to claim 1, as he used the feature from claim 3, but not the epoxy from claim 4. Of course, the amendments to the client's version were needed, and I thought that I would rather do what could be too difficult for the client: import features from the description.

      I also thought that it was my responsibility to defend the broad version in such circumstances. I mean the following arithmetic. If I were an examiner, I would give almost zero marks for the extra-narrow claim with epoxy if some broader claim was patentable (not necessarily "our", but for example the variation with the feature "for high-sensitivity electronic components" in the preamble instead of the SRI feature), Im contrast, if I were an examiner, I would give just slightly below maximum for a non-patentable, but easily correctable (before and after grant) broad claim. So, I estimated the potential losses as much smaller if I would choose "our" broad claim, and chose it.

    2. Re appeal: I appealed last years decision on A, because I got 0 points for my main claim although it was the solution of the examiners report. It was forwarded to the appeal board. I received a first response from the appeal board with pretty far-fetched case law why my claim is unclear and thus not novel (allegedly, it was unclear whether I claimed the dish washer composition comprising the pouches or not because I used "verpackt in" in German corresponding to "packaged in"). They did not care that the English solution was using the corresponding term in ENglish. Only after their response the German translation was published, which used the identical term! I filed a response and since then no reaction. So I had to take this years exam and passed. In my opinion, candidates are deprived of their fundamental right to appeal a decision by the EPO due to deliberately slowing down this process.

      So, unfortunately, the likelihood that your appeal goes through is almost 0.

    3. Have you requested accelerated proceedings?

      This spring there were five decisions by the Appeal Board in favor of candidates for Paper C - 2017. See D 25/17, it is pretty interesting.

      I know that chances are small. Also, I doubt that some limited correction would be useful to me. Before I also hoped to improve the system by appealing. However, now I think that it will be just getting worse. The quality control seems to be a about zero even at the task formulation stage. See, for example, how many unclear questions have been asked this year in the pre-EQE, or the pains of interpreting part 2 of paper D in the delta patent blog.

    4. Thanks for pointing out the appeals - interesting that with regard to closest prior art in the C Paper of 2017 that they agreed (based on the arguments in D 14/17). Also in the D 25/17, they accepted that "selling" may be ambiguous.

      Don't appeal to try and improve the system - the best way is to pass the exam and volunteer to help in making the exams. Appeal because you think you are correct, and you have the time and resources to prepare the case and go to oral proceedings. And if you would regret not filing an appeal.

      You will still need to fully prepare for next year's exam, so consider it a bonus. Don't focus on the appeal. Winning on substantive issues ("my solution should have been awarded more points") is not possible, because the Board does not look at that. However, if you were awarded zero points for something, then you might have a chance at an extra couple of points.

      Also be honest with yourself - there may have been some arguments or approaches that they did not accept, but what about all the other points you missed.

      There are a lot of checks and quality control, but it is impossible to make an exam which is unambiguous for several hundred people with different technical backgrounds (and experiences) in three languages. I have talked to many people making the exams over the years, and they are all trying to make the best exam possible. To be honest, I think the best way to improve the system is for more patent attorneys to be involved in the commissions, and in particular from countries who do not do the exam in their native language.