Monday, April 4, 2016

Why I don't worry about getting scooped

Some witty colleagues might say it is because I am not doing anything worth scooping!

Some scientists live in a constant fear of being scooped, i.e. not being the first to publish their latest research result. This can lead to people being very secretive about what they are working on and/or being in an incredible rush to publish. Some groups even require members to sign confidentiality agreements about talking to outsiders.
I always find it strange and disappointing when I ask someone what they are working on and then at some point they say, "I can't say anymore because I don't want to be scooped."
In most cases this strikes me as both egotistical and unrealistic. Most of what we are working on is not so important that others are going to drop everything they are currently doing, steal our idea, work out all the details, and rush to publish...
Sorry to disillusion you.

I often write on this blog about things I am currently working on, long before I have a preprint.
Someone once suggested to me that this was a bit "gutsy" or risky.
To be honest, I would be flattered if someone read a post here and "stole my ideas".

I also don't worry about getting scooped because I have had several concrete "scooped" experiences that showed me that in the long term it did not really matter, particularly in terms of publications.
The stories below show how things are all a bit random.

1. With some mathematics colleagues I wrote a paper about the exact solution to the BCS Hamiltonian for superconducting nanoparticles. We put the paper on the arXiv and amazingly there was a similar posting of a longer work the same day, just after ours in the daily listing. We sent our paper to PRL, but a referee claimed that we had been "beaten" by the other paper! Our paper ended up in PRB Rapid Communications, the other in PRB.

2. While a postdoc with me, Ben Powell did some nice calculations showing how a RVB solution to a Hubbard-Heisenberg model relevant to organic charge transfer salts gave a d-wave superconducting state. I thought this pretty exciting and that we should write the results up for PRL. However, even before we started writing two preprints with similar results appeared on the arXiv. I suggested we write the paper as quick as possible, put it on the arXiv, and send it to PRL anyway. We did and our paper did make it into PRL, along with the other two!

3. Another postdoc completed a draft paper with me, that I then sent to a colleague for comment. He informed me that he read the abstract and realised his postdoc had similar results. He said he would not read more and requested that we both submit to the arXiv on the same day and then send the two papers to PRL, acknowledging each others existence. I agreed. Neither paper made it into PRL. Ours ended up in PRB Rapid Communications. For some reason the other paper was delayed for more than a year.

4. Another postdoc visited his home country and gave talks about our work. Later it was brought to my attention that someone who he had talked to wrote a similar paper, with no acknowledgement of our preprint. When challenged the author claimed it was all different and there was no need to acknowledge our paper.

5. I once gave a talk at which I discussed work that had just been published. A year later someone in the audience published a very similar paper with no acknowledgement of the published work I had described.

So chill out!
It is probably not worth worrying out.
Get some perspective. Remember that in academia the stakes are low.

The cartoon by John R. McKiernan is from here and features in this blog post.

Do you have any experiences of being scooped?
Did it really matter in the end?


  1. I always find such experiences to be very deflating, regardless of whether I am the "scooper" or the "scoopee". If the result I came up with was just derived by someone else, or was just about to be derived by someone else, then it feels like I am adding nothing to the scientific enterprise. I tend to view such moments as a sign that I need to do a better job of choosing which problem to work on.

    1. Thanks for the comment. I am sorry you can only see the negative side of such experiences. The positive is that you are picking problems that are of interest and importance to others. Someone getting the same result is also encouraging; it increases the chance that you actually have a valid result. For this reason, I think it is very important for science that we do have people working in similar problems to test reproducibility.

  2. The consequences of getting scooped may have little impact for established scientists, but for undergraduate and graduate students they can have big impacts. Some work that I worked on for two years as an undergraduate got scooped and was subsequently rendered unpublishable. That is time that could have been spent on other projects that do yield publications which can make a big difference in terms of getting fellowships and so on. When applying to NSF fellowships, for example, my proposals all got excellent marks and my experience was praised but I lost points for having not been able to produce a publication.

    1. Thanks for the comment. I agree it is important to acknowledge that the consequences have a bigger impact on junior than senior people. But, my experience with the stories I recounted was that I don't think there was much negative impact on the junior people involved.

      I can't say anything about your particular experience. But, I do find it puzzling that in many situations people think there are no "salvageable" publications after being scooped. I find the results can be reworked into a slightly different paper, possibly at a "lesser" journal. Furthermore, as I commented above I think it is important for the reproducibility of science for people to be publishing "similar" results.

    2. Oh I absolutely agree. The experience did deflate my confidence in the whole process because I had figured there was still value in putting out my own work in the spirit of reproducibility. Journal editors however disagreed and it didn't go out for review. The experiments were simple low-hanging fruit and there was relatively little that could have been done to put a new spin on the work.

  3. I think this touches on a fundamental problem in the current climate in which (at least our corner of) science is done.
    Reproducibility should be one of the most important things. Both within study and between different groups. I.e. publishing something second should have (almost) the same assigned value as publishing it first.

    I may be overly sceptic, but I am not really comfortable with the result of many papers until I see it reproduced (and my experience shows my scepticism is often warranted).

    If (at least the first) reproduction is valued to a similar level as the original publication, being scooped would become a non-issue.

    1. Thanks for these helpful observations. I agree completely!

  4. Addition: that also means that citation behaviour may have to change: don't only cite the first paper. Cite the second one as well.
    (and to avoid overly long numbered lists, journals should allow multiple papers being cited within a single reference.)

  5. My whole life consists of seeing other people, usually American, develop ideas that I had years before, while I waste away here in Queensland, isolated and ignored.