118

I run a lab in an applied computational field (e.g. computational biology/chemistry/physics etc.)

One of my postdocs (“A”) has been working on a problem for the last two years. They have made good progress, but the accuracy of their solution is not quite where we want it.

A new postdoc (“B”) joined my lab a month ago. Recently, B told me that they had given some thought to A’s problem and realized it was amenable to a specific computational method that neither A nor I were familiar with. B then showed me that, over the course of an afternoon, they had adapted said method to solve A’s problem. B’s approach is far superior to anything A and I had come up with in every possible way—accuracy, speed, interpretability, etc.

A does not yet know what B has done; B came to me first because they did not know how to break the news to A. To tell you the truth, neither do I. Knowing A’s personality, I think they will be extremely upset about this, and I have no idea how to navigate this delicate situation.

(To give an idea of how much better B’s method is: imagine spending months trying to come up with a good numerical approximation to a fiendish differential equation or integral, only to be told that there’s an obscure closed form analytic solution, obtained using techniques from a completely unfamiliar subfield of mathematics.)

Nik
  • 4,748
  • 13
  • 50
ScoopedPostdoc
  • 943
  • 2
  • 3
  • 8
  • Comments have been moved to chat; please do not continue the discussion here. Before posting a comment below this one, please review the purposes of comments. Comments that do not request clarification or suggest improvements usually belong as an answer, on [meta], or in [chat]. Comments continuing discussion may be removed. – Massimo Ortolano Apr 01 '23 at 11:37
  • 14
    Could you clarify if B looked at A's approaches/attempts at a solution at all? Or did B merely hear the initial problem statement and then come up with a solution without additional info? I ask this because it seems quite possible that B's recognition that A's current approach wasn't working was itself important to the solution - otherwise B might've tried going on the same (or a different) path, and reached their solution later/never. If B might've benefited from A's work indirectly (even if it was just realizing "A's approach probably won't work"), a joint paper really makes sense. – user541686 Apr 02 '23 at 06:56

11 Answers11

137

This is a tough one.

I am a pure mathematician. Early in my career I was in a situation akin to the one of Postdoc B. (Except I was still a master student and the person in the role the postdoc A already had a permanent position. He worked on a problem for a year but had only an incorrect proof. One more thing to add here: I have learned a lot of research-level math from A as an undergraduate student.) Here is how my advisor (who was also a former advisor of A) handled the situation:

He first talked to me and said something along the following lines:

"Moishe, you are brilliant. We all know this. In your future math life you will prove many remarkable theorems. [At the time I felt embarrassed by all this undeserved flattery. In retrospect, my advisor used the right approach here by “smoothing my feathers” first.] You solved easily a problem that A was working for a year. What do you think of publishing the result as a joint paper (with A)? To you it probably will not mean much, but to A it will mean a lot."

He was right, I said yes and published the result together with A. I did not resent my advisor's advice; I also do not regret following it.

Will this strategy work in your situation? I do not know, but, I think, it is worth trying.

Nik
  • 4,748
  • 13
  • 50
Moishe Kohan
  • 4,716
  • 3
  • 16
  • 24
  • 1
    Comments have been moved to chat; please do not continue the discussion here. Before posting a comment below this one, please review the purposes of comments. Comments that do not request clarification or suggest improvements usually belong as an answer, on [meta], or in [chat]. Comments continuing discussion may be removed. – cag51 Apr 03 '23 at 01:15
62

I wish all the world were as generous as the situation and solution described in the answer by Moishe Kohan.

However, before you do anything, you need to verify the solution of B to be sure that it holds up. But, assuming it does, there is no real solution but to "bite the bullet" and deal with A. It should be your task, not B's to break the news to A. You will probably also have some responsibility to A to get them to a success point.

But talk to B first to explore their position.

I don't think that A has an automatic authorship position here, given your description, but I also don't think it would be exactly a "gift" to include them.

But, if B insists on sole authorship, as is probably valid, then you need to sit down with A to break the news. You can then explore how much is salvageable from the work already done by A.

And, don't try to pressure B on a joint publication. (Caveat, a somewhat different scenario might call for a very different solution.)

One option in a joint paper is to describe both approaches and why the one is superior. That would make joint authorship valid by anyone's (I hope) definition.

Buffy
  • 363,966
  • 84
  • 956
  • 1,406
  • 101
    A has worked on the problem for a year. They almost certainly have a lot of insight, tests, and comparison data that would be very useful for a paper. It would almost certainly clear the bar for "intellectual contribution" to work on a paper together. The paper might be better by describing the failed attempts because it highlights the quality of the solution described in the paper. – Wolfgang Bangerth Mar 31 '23 at 17:10
  • 1
    @WolfgangBangerth Of course! A has certainly done enough work on the problem irrespective of the method used to ultimately solve it to warrant a co-first authorship. However, the publication would still be a methods paper first and foremost, and the method showcased therein would certainly be B's. – ScoopedPostdoc Mar 31 '23 at 17:23
  • 3
    It would be B's method, but the readers won't know this. For all anyone can infer from the paper, it's a group effort. – Wolfgang Bangerth Mar 31 '23 at 18:25
  • 49
    If B insists on sole authorship, B must not include any conceptual or writing contribution from A in their paper. B learned about the problem from A, and it would be easy to accidentally include an insight from A in the paper that B assumes is trivial or well-known, but is actually novel, and an intellectual contribution warranting co-authorship. Given A has worked on this problem for so long, this is quite likely. The safest option for B is to offer co-authorship. This also will lead to a faster publication because A is probably more familiar with the literature than B. – WetlabStudent Apr 01 '23 at 03:20
  • 3
    @WetlabStudent, the OP has described it as independent work. I based the answer on that, not assuming otherwise. But, yes, a sole author publication can't include IP from others. Learning of a problem, however is not sufficient. You have a lot of speculation in your comment, actually. – Buffy Apr 01 '23 at 11:24
  • @WolfgangBangerth : indeed, and in such a situation it might well happen that B's solution only works within certain limits or given certain assumptions, while A's work can still be used to study how the system works outside of those constraints. – vsz Apr 01 '23 at 11:30
  • 5
    @Wolfgang Bangerth It would be B's method, but the readers won't know this. For all anyone can infer from the paper, it's a group effort. Excuse me, can we try to give A some credit for self-respect here ? Most people don't want to wake up every morning and feel like a phony. They want to be known for their own work, not that of others. – Trunk Apr 01 '23 at 14:20
  • 5
    Everyone here seems to assume a pure maths approach where authorship order doesn't matter but the OP talks about an applied field and in most applied fields order does matter. In this case B would have nothing to lose from having A as co-author since the contribution of the first author isn't watered down by having an extra author. The OP could explain this to B and ask if they would be happy with this arrangement, also explaining the advantages of having A as co-author (e.g. that they can more easily than B write the introduction and literature review, etc) – Luca Citi Apr 02 '23 at 22:07
  • @buffy actually the OP has described nothing about the interaction between A and B at all. Both your answer and my comment are the same level of speculative, but at least my comment is a conditional statement: if X then make sure Y. Since the OP has said nothing about whether Y is true, it is fair to recommend they make sure they have done their due diligence to make sure Y is the case. It is also speculative to assume B had not benefited at all from intellectual knowledge from A, even if indirectly through the supervisor. All I am doing is making the assumptions explicit. – WetlabStudent Apr 03 '23 at 00:57
  • 5
    Given that A have not solved the problem for 2 years, B's solution should be ran through every failed test from A's experiments to confirm its validity. At minimum, A would know a lot about why his results didn't work, and this is extremely non-trivial knowledge. – Nelson Apr 03 '23 at 01:58
  • 2
    @Trunk Surely failures can also be contributions. Edison is said to have invented 10000 ways to not make a lightbulb before he invented 1 way to. – user253751 Apr 03 '23 at 18:37
  • 5
    Another way of framing it is that everyone got lucky. B got lucky, because a bit of knowledge they had happened to be the key to a lock someone else had constructed, and they're getting a free paper out of it; A got lucky in that they finally have an outlet for all the knowledge and tests and reading they now have, and can stop banging their head against a wall; and you got lucky, in having the good taste to recruit a new postdoc, B, from (by the sound of it) a slightly different background, who managed to unblock a logjam. Drinks all round, when you knock out the paper! – Norman Gray Apr 03 '23 at 19:11
  • A can't say B's idea was his. B's idea - or should I say, his application to A's problem of an existing concept - could not have occurred if A's approach hadn't failed to attain adequate accuracy. A joint paper showcasing how good the B approach is versus that of B (and of his supervisor, ScoopedPostdoc) would as it stands now put A in a bad light. Someone here suggested that an exploration of the strengths & limitations of A's approach in various situations might give body to A's work. I also think that this may be a possibility. Without domain detail we can't say. Ask a math expert. – Trunk Apr 04 '23 at 15:33
  • @Norman Gray Let's try and stay sober for the moment, please. A got unlucky, pure and simple. Firstly, he took an approach that turned out to be unfruitful. Secondly, the discovery of a classic solution to his problem seems set to make his 2 years work come to naught. Were I HoD and ScoopedPostdoc to come to me for advice, I would - after thoroughly scolding SP for his impatience - call in the best applied math expert on the campus to advise on the scope of A's approach. Sure he'll say why wasn't I called in beforehand. But he would be able to help A salvage something from his 2 years work. – Trunk Apr 04 '23 at 15:45
31

Dilemma appreciated. I'm a sw developer (40 years) who has worked with many scientists & done a lot of collaborative design in different fields, with oft-touchy people.

Further to the excellent answer by Moishe Kohan is the question of framing - would B have thought about A's problem in the same manner if they didn't know:

  1. A had explored a lot of approaches
  2. a nearly-acceptable answer had been found but accuracy of their solution is not quite where we want it?

That context of A's work as a preface to B's insight is further reason for a joint paper.

Andy Dent
  • 411
  • 3
  • 4
  • I am not sure A's work is a preface. The solution by B is analytical so it does not really matter what A did before (this is not to lower A's work, which could be excellent in the computational world, but is is orthogonal to B's work) – WoJ Apr 02 '23 at 11:59
  • 1
    @WoJ Analytical? OP writes B showed the problem is "amenable to a specific computational method that neither A nor I were familiar with". It does seem to be a bit of an assumption to claim the two approaches are orthogonal. – Anyon Apr 02 '23 at 16:51
  • @Anyon, ah, you are right. I was misled by " (...) only to be told that there’s an obscure closed form analytic* solution, obtained using techniques from a completely unfamiliar subfield of mathematics"* – WoJ Apr 02 '23 at 16:53
7

This scenario reminded me of a Billy Corgan (of Smashing Pumpkins) interview. Essentially, he says he would have been more successful and faced less drama if he had shared his writing credits.

Based on this logic, if "A" contributed to "B's" idea, B will have more success AND an ally by sharing the credit with A.

Billy Corgan Interview: https://youtu.be/GLvqzWSRwnA?t=307

The interview contains some swearing.

Soren
  • 87
  • 5
5

You are wondering how to best deal with A's problem, but you need to realize that this is not primarily A's problem but yours.

Why? You are "running a lab". That means science. Now in your lab, a problem met a solution. That is the best thing that can happen in science. But it sells itself as a problem to you.

You want this to happen, and you want to encourage that to happen. Of course, you should try to wrap your head around what you can do in future to make this happen sooner rather than later. That's a long-term strategy. Part of it is making sure that you are up to date on the state of art at a name-dropping level so that you can offer suggestions to your post-docs what material to get acquainted with to be sure they are not obsoleted by prior or ongoing work (and if there is ongoing work, try maintaining the contacts to have a good idea whether there is a way to join it while it happens).

The problem you now have is to sell to A that they have likely wasted time and work regarding the goal of making science progress. It would not be honest to try burying B's work or rather discovery, and you need to figure out what B feels they should be getting out of it: they way you relate this, it seems like they are quite aware about the difference in actual invested work compared to what A did and are comparatively flexible regarding how to on from here.

Now it very much depends on A's priorities how to best serve their interest. Essentially there are the options:

a) need to get something published, no matter what. That is likely best served by finishing what they started. If turning the method found by B into a formal and tested and verified paper on the same problem space takes longer than finishing A's work, there is not even a conflict. It's just that A's paper is doomed to early obsolescence, but that would be independent of whether B is involved with applying the published third-party method to this particular problem or not.

b) need to establish lasting credits particular to that topic. In that case it is essential for A to get up to speed with B's discovery and publish/copublish about it in a reasonable time frame. Whether this is better served by first publishing work, methology and verification of his current approach and then basing a second work on this, or by binning his current work and restarting (and trying to secure funding, deadlines etc for it which you should support to the utmost of your ability) immediately is A's decision.

Being able to run a serious in-depth analysis of "best prior art competently applied" to "new method" is very very valuable for advancing the state of science and B cannot deliver that because he does not have the 2-year experience of A.

So again: this is a stroke of luck for science, and it is the thing what running a lab is for. It is your task to make both A and B see this as the stroke of luck it is and make sure that it turns into a net positive for both of them in order to make the progress of science line up with the well-being of your post-docs. Since this will involve replanning (particularly for A) and throwing large pieces of one's scientific fortune into an unplanned pairing with a colleague, it is your task to ensure a solid and trustworthy basis for the collaboration of the two. How much time and effort you will need to invest yourself very much depends on the two's ability to progress to coordination where both are satisfied with the outcome. Particularly regarding the actual workload and involvement and payback of B, there is a wide range of what can feel satisfactory to them, more so than with A, given their upfront work.

If they have problems trusting one another, it is your job to provide a common point of trust by nailing down agreements and making clear what developments you are willing to accept or not in the course of work submitted under the auspices of your lab.

Good luck in making this work! It's rare to get a chance to actively align the interests of people and scientific progress, and you should do your best not to waste it.

3

My gut feeling in computational mathematics is that every method has place to shine, given the right circumstances.

Method X may work very well in situation ABC, but might not generalize beyond that setting (in theory or practice). Method Y might not work as well in situation ABC, but may apply to situations DEF, GHI, and JKL much easier than X.

If your example is anywhere close to the scientific setting

(To give an idea of how much better B’s method is: imagine spending months trying to come up with a good numerical approximation to a fiendish differential equation or integral, only to be told that there’s an obscure closed form analytic solution, obtained using techniques from a completely unfamiliar subfield of mathematics.)

then would guess that B's method works in specific setting but A's method may work still once you change the specific problem a little bit (stuff like: nonlinearity, wild material coefficients, etc., can work wonders to derail good methods).

Ambicion
  • 5,541
  • 5
  • 31
  • 53
  • There's truth in this. But what is your recommendation ? Two separate papers, a joint paper or what ? – Trunk Apr 03 '23 at 12:19
0

This answer is to this question, but moved into an industrial environment rather than academic.

It seems that the "delicate situation", at least as referenced by the other answers, relates to who gets to publish the results, and (I presume) get a career boost. I re-read the question, and "publishing" is never mentioned.

Even within a lab, many problems are solved and advances are made that aren't directly published. A lab itself is a small industrial culture which solves many problems, some big, some small, some publishable, some patentable, and some that are keep secret until they need to be revealed. I would claim that given the OP's problem statement, this may have much in common with an industrial setting.

In an industrial setting, person A may have been working on a problem for a year and made good, but not complete, progress. Person B has a new idea for how to solve it.

My job as the manager is to put persons A and B together to talk about the problem, and the progress B has made.

If the problem was important, chances are that A was my best pick for someone to work on it. If I had a preferred outcome, it would be that A continues his work with B's ideas, and finishes the problem. If the problem was not important, it could mean that A isn't important to the organization. I should think about that.

Several outcomes are possible. No matter what, I would publicly note A's long-term progress and B's breakthrough.

I would also look for other problems to slide past B.

It is important that I be part of their first meeting to judge the vibe, and then follow up with guiding an outcome based on the responses of the individuals. I would also keep an eye on interactions between them after the meeting, looking for signs of collaboration or obstruction.

cmm
  • 431
  • 4
  • 10
0

Given the fact that B had the presence of mind to consult you before approaching A, I would echo the sentiments of several posters here to impress upon B how the terrain mapping of A framed the problem in an efficient way. Knowing where not to go has huge value and often comes at a high cost of time and effort.

Gearoid Murphy
  • 101
  • 2
0

This is in fact very good news, and is one of the best reasons why we discuss our research with our peers. Yes, it is a situation that must be navigated with care, but really this is a win-win. Clearly the three of you should share authorship of this result, as it was cultivated in your laboratory through an ideas-sharing process. Culture of authorship order differs by field and should be navigated accordingly; you need only to be sure that everyone gets credit, including the credit for working at this problem for more than a year.

rhombidodecahedron
  • 1,401
  • 13
  • 24
-6

A career math's challenge is like a romantic partner: if they get unexpectedly wed to a competitor, jealousy can ensue, even revenge. If you hint the news to him and he starts to splutter and even cry, don't unleash the full force of the news. Use subtilty: Scientists are temperamental, unstable, you potentially stand in the way of causing A to become depressed, feel like a failure, belittled in his field and workplace. You and B can conspire to announce it gently to him over a period of months and years, not necessarily all in one shocking destabilizing revelation... Say that news has come of a potential solution for the problem, using said analytic system, and let A find the solution for himself and feel positively about it.

Solving the maths problem using broad general knowledge is a very laudable feat (is that OK english?) You can agree with B to not break A's heart and belittle him if you feel that it is the professional way of dealing with it, depending on A's mental stability.

Once he is over the entire quandary, a few months down the line, perhaps years, you can start to break the news to him... "Oh! it was student B that told me that said analytic method could solve the mystery". Some months later, say that B even "started to demonstrate the solution, so you told him not to give A chance to solve his beloved maths problem". see how he reacts, if he starts to cry, leave it.

Various tactics can be used to lessen the danger of emotional destabilization.

Academia and depression are often closely bound, and there is no need to raise the risk of a student's depression. Perhaps now is a good time to research instances of madness and nervous breakdowns of scientists in relation to academic theories.

cag51
  • 67,924
  • 25
  • 181
  • 247
bandybabboon
  • 142
  • 7
  • 1
    Announcing this over months or years does not seem practical; B presumably wants to publish now. Moreover, I agree that a certain degree of tact is called for, but I see no need to "conspire" to avoid a "nervous breakdown"; A can presumably handle their disappointment appropriately. – cag51 Apr 03 '23 at 18:23
  • scientists can be spectrum autistic (i.e. Elon ) and hyper nervous in a weird social combo which lends me to consider that mental health is not enough valued in favored answers, I know folk that have had major depressions from phd's and sometimes it put them off that career choice completely and they change fields. – bandybabboon Apr 03 '23 at 20:23
-42

To begin with, this is not a problem with A's sensitive personality and how to delicately explain to him that he's been scooped.

The problem here is you.

Yes. You. And your selfish way of (mis)managing a group of researchers.

You should have lain down the law to all new researchers that they do not second guess the work others in the group are doing without their consent. Not even as an academic exercise. This does not preclude researchers exchanging perspectives: it just makes it happen in a decent and consensual way.

If the solution to this problem was so important then you should have - with A's consent - had a brainstorming session with the whole group, perhaps a few guys from the app math school too.

You've now got a discovery that appears valuable to the world but which neither you nor A can claim credit for. And B will belatedly realize that he shouldn't put his name to it either.

Gather A and B and talk to your Head of Department and tell him/her the whole truth - not least in acknowledging your foolish omissions that led to this.

If you decry responsibility on this you will have more messes ahead of you.

Trunk
  • 4,418
  • 9
  • 28
  • Comments have been moved to chat; please do not continue the discussion here. Before posting a comment below this one, please review the purposes of comments. Comments that do not request clarification or suggest improvements usually belong as an answer, on [meta], or in [chat]. Comments continuing discussion may be removed. – cag51 Apr 02 '23 at 15:14
  • 7
    Can you clarify what you mean by "second guessing" here? On the face of it, it sounds like you're saying that postdoc B should have been prohibited from thinking about any problem that any other researcher is studying (unless they've specifically been given permission to think about it), but of course that would be absurd. – Sneftel Apr 03 '23 at 16:24
  • @Sneftel I said it's proper to wait till someone asks for help. Researchers like to work things out for themselves. Nobody - researcher or otherwise - likes imposed "help" or "advice". A responsible supervisor should be mindful of this. – Trunk Apr 04 '23 at 14:30