Tuesday, 5 March 2013

Paper C EQE 2013 - Impressions

Time was reduced from 6 hours to five hours. This seems to be made possible in the way I expected it to be done: less attacks (only five claims) and one piece of prior art less compared to most previous C exams. So far, I get the impression that most candidates managed in the given time.

For more impressions, please read further.

I welcome your impressions.

Jelle Hoekstra

Themes in this exam
  • Dealing with an attack where the primary document refers to a second document.
    Should it be a novelty attack or an inventive step attack?
  • Dealing with negative statements.
    Should it be seen as an unovercomeable teaching away or can it be ignored?
  • Dealing with lists
    When is it a single list situation, when a multiple list situation. Should it be a novelty attack or an inventive step attack? 
  • Selecting the closest prior art (usual theme in the C exam)
    If there are at least two candidate starting points for a problem-solution approach attcah, how do you choose?
  • Subject-matter added before filing Annex 1 (divisional)
    Is it added matter? What is the legal basis - art.76(1) or 123(2) (both mentioned in Art.100(c))? What does the Enlarged Board decide on matter added in a preceding divisional?
  • Dealing with definitions, spread out through the Annexes (usual theme in the C exam)
Complexity of this exam
Always difficult to judge. I personally found the exam doable and it looks like there is enough variation in the exam to give candidates a fair chance of passing.
Unlike last year, I did not see many solutions proposed by candidates that are a clear fail.

I expect a high pass rate for this years paper C.

Jelle Hoekstra


  1. I felt really pushed for time especially when doing the analysis of the claims/prior art. The time pressure seemed greater than when I did past papers (I did C2010, C2011 and C2012 in the 'old' time of 6 hours and didn't feel as rushed). So I looked at the length of the papers:
    C2010 - 25 pages
    C2011 - 25 pages
    C2012 - 24 pages
    C2013 - 24 pages
    Although there was one less claim and one less piece of prior art, my impression is that there was no difference in the amount of material present in the paper.
    I also wonder whether it was originally written as a longer paper. To me it felt like there was sufficient affitional information for an attack on an extra (and non-existant) claim.

  2. I think Liz is correct. Quite the same number of pages as in previous years. I also expect a somehow higher pass rate than last year but there is - as always - plenty of possibilities to screw up although I feel that the added matter attack was clear cut and most of us got it right. As usual, it all depends on the marking and whether an alternative attack to A123 is required (i'd say no).

  3. Interesting observation. I did indeed also have the impression that both the analysis of Annex 1 and of the prior art may have taken a very similar amount of time as in previous exams.
    We train our candidates to minimize analysis and integrate it at least partly with performing attacks. In such an approach less attacks should result in some reduction in analysis time.
    Candidates performing a full analysis before doing any attacks may not have saved that much time and maybe ran into time problems.
    Interesting to hear other opinions.

  4. I agree - I thought the time pressure was greater for Paper C than it was for the new-format Paper D. This surprised me (I fully expected it to be the other way around).

  5. You never know with paper C. When the examiner's repor comes out, it might very well be something in line with "The focus of this years exam was time management. Therefore, very few marks were awarded to candidates attacking only claims 1-4 based on inventive step and novelty. Instead, candidates were expected to perform four different IS attacks on the subject matter of claim 5 and possible fallback positions in the description."

    1. I do not agree. The focus in the exam is never time management. The committee tries to prepare a paper that they believe can be reasonably done in time by a prepared candidate. So, as long as you keep on the right track, the committee believes you could get to the end.
      Candidates that do not get to the end will loose marks, but there is never a punishment on it; the exam is also never designed to do that.
      It is more likely that marks for the last attack are tuned down if a lot of candidates did run into time problems.

    2. I was a bit ironic, trying to paraphrase the Examiner's Report from EQE C 2007...

      For "ordinary" years, the Examiners' Report is mostly reasonable.

    3. If the marks are tuned down for the later attacks that would worry me (whatever the reasoning). I started with claims 4 and 5 and then went back to claims 1, 3 and finally claim 2. I'm sure other candidates also work through the claims non-sequentially.

    4. Marks are only exceptionally adjusted, is my impression, but details are rarely given. I do not see any reason for an adjustment in this exam. The committee usually allows marks for, possibly not foreseen, reasonable attacks. From paper DII it sometimes looks like that marks are tuned down if an issue was handled badly, with hindsight may have been to difficult in view of already other complexities in the paper, and would cause too many candidates to fail.
      I have the impression that candidates with a good answer on a tuned down issue can score bonus marks up to the original intended level.

  6. For me personally, the amount of time available would have sufficed, if I had come to a conclusion a bit sooner regarding the closest prior art for claims 2 and 3. Claims 1, 4 and 5 seemed straight forward but claims 2 and 3 were a real struggle on selecting the CPA. Nevertheless, in case for claim 5 further attacks (beyond Art. 100(c)EPC) were expected, the time available was certainly not sufficient.

    1. Thanks, like to hear those things. Can trigger me to change our training

    2. This is also how I thought. It was not clear if an inventive step attack was required for claim 5 in addition to A.100(c) attack (I still do not know). If it was needed then the time was too little.

  7. I assumed that due to the time management side attacks are not expected anymore. So I also agree with Richard because I also started with claim 5 and 4. Then deal with 1 and loose efficient time with 2 and 3. Hardly finished writing claim 3, I have serious doubt about my handwriting for claim 3 if it is legible or not. So no time to think about additional attacks according to me.

  8. I think that I may have managed to propose a solution that is a clear fail: in my notes (novelty matrix) I noted that the lists did not appear to be an explicit disclosure for claim 4. However, I managed to gloss over this when attacking claim 4 and started with a clearly wrong novelty attack.

    I managed A5 novelty and A3+A2 IS for claim 1; and coninued A3+A2 for claim 2; and A5+A2 for claim 3. Although I am not sure that my argumentation was as strong as I would have liked.

    However, I realised my error with claim 4 and tried to correct it with IS attacks, but did not really explain that I had identified the issue and I ran out of time. The first time in any paper or past paper that I have effectively panicked, becuase I was shocked at the simplicity of my error.

    Running out of time for claim 4 also meant my IS attack for claim 5 merely referred to claim 4's attack and identified the ratio feature in A4, without properly applying the PSA (and I only attacked claim 5 as granted, not amended to paragraph 16's range).

    Given the options in claim 4, and the simplicity of the issue of the lists, then I would deduct a lot of marks for my attempts at claims 4 and 5.

    I suspect that the reason why you do not see many solutions that are clearly a fail is because this year it became very clear very quickly after the exam what the desired solutions are and people have probably realised themselves this year that they have failed and decided not to post their clearly wrong solutions.

    1. Thanks for your reply. Appreciate that. The discussions so far have been:
      - a bit on claim 1 - was it novelty or inventive step (I do not know how being wrong will be punished)
      - a lot of issues on closest prior art for claims 2 and 3. Most people have got at least one right
      - discussions have been a bit silent on claim 4. I regard the attack on claim 4 as being very tricky. The two options are easy to see, but how to deal with the lists? Think that marks will have been lost here even if the right attack is chosen.
      - Claim 5: think that only added matter should be attacked and most candidates probably saw that

    2. I think it depends on your background. I sat the chemistry papers for A and B and I therefore I thought dealing with lists was straightforward (hence I started with the attacks on claim 4).
      I made my mistake on claim 2 (which I attacked using A5+A2). I had done the correct analysis, I just panicked and by the time I realised I had made the wrong choice as CPA there was not enough time to change my answer.
      Also I got the added matter attack right, but I don't feel that I argued it as well as I might have done.
      I won't be surprised if I have to re-sit next year.

  9. Jelle, thank you for all of your comments. I have not attended any DeltaPatents courses (yet), but I had come to the conclusion myself that the new format of paper C does not lend itself so well to some exam techniques. I think that I wasted too much time filling in a detailed matrix, whereas the reduced number of claims, and particularly the reduced number of annexes may enable a more straightforward comparison.

    I think that it should be possible to identify a couple of potential CPA's for each claim, and potential sources of the missing feature. Thereafter it is relatively easy to directly compare a couple of annexes to provide very good attacks.

    Alhoewel ik duidelijk engelstallig ben (Brits) woonde ik lang genoeg in Nederland om toch (ooit) aardig Nederlands te kunnen. Mogelijk dat een cursus in Nederland de zilveren rand van de wolk van mijn mijn slecht prestatie in een paper C herexamen wordt.

  10. Hi, thanks for your comments.

    Indeed we've already realized long time ago from our re-sitters courses that for some candidates a wrong choice in analysis methodology result in wasting time. Through the years we've developed another approach which avoids a detailed analysis for aspects not used in an attack. We've tried and tested this approach for several years on many papers and many trainees. It works. A first version was last year turned into our main analysis methodology and we've dropped other approaches. Last season, I've fine-tuned the new approach and will also show this in the training for the coming years.

    Van harte welkom op onze cursussen. We spreken wel Engels of Dunglish :-)

  11. Somewhat off-topic, but still much on-topic...

    Do you have any indications on when the results of the main examination will be published this year?


    1. No, I do not know. Expect the same as last year (or a bit earlier); so around mid July.