Consider the concept of opportunity cost.

There are two possible situations:

  1. I program everything, do the evaluation with a benchmark, tell nobody, write everything and publish alone.

  2. I program everything, do the evaluation with a benchmark, write something and publish with four other authors, that mostly write the paper.

What should I do?

To answer your possible questions:

  • They are not going to program, no matter what, so that doesn't help to save any time.
  • Publishing without telling them anything may be perceived in a suspicious way, specially by external reviewers and considering there may be past papers in which I wasn't alone.
  • It may also be perceived as if I'm not a good team player, which worries me, as finding a new job may be harder due to this.
  • The order of the authors is alphabetical, and I'm not lucky in that sense.
  • They are more reputed than me.

PD: the original idea for the paper either comes from me or it comes from someone else. The implementation is always my task. At this very moment I could refine one of those ideas and improve it (the previous version is submitted but not reviewed). I'm very confused about how to proceed.

PD: I am a postdoc now but I am a PhD in a different institution (I haven't defended my thesis, it's nearly finished).

  • In your field, does the author ordering count, i.e. is it commonly understood that the first author typically did most of the work? Why would you have 4 coauthors, that seems like an awful lot if you indeed do "everything"? What were their scientific contributions? I am sure that the more reputable scientists can help you write a better paper and focus on what is important, and as such it would help you. Also if the idea was pitched to you by someone else and you take it and publish an implementation without the other person, that is somewhat nasty. Especially if you work in the same group.
    – alarge
    Jun 5 '14 at 17:22
  • @amlrg in general the author ordering is by contribution, but in this research group (and I guess some others) the tradition is using alphabetical order. In my previous research group this was the case for an associated subgroup and it changed 5 years ago (it's hard to explain). Their scientific contributions vary with the paper, in some cases formal proofs, in some cases writing and structure, in some cases the original idea, or a combination of the previous. Nothing that I couldn't do on my own. Everything (even put together) is less time consuming than programming, IMHO.
    – Trylks
    Jun 5 '14 at 17:41
  • @amlrg They have suggestions about how to write things better, but their style contradicts in some points the one of my previous supervisor; I think I've already learned the basics and I'd like to have my own style writing. About ideas pitched, it's not taking the idea from the person, but from a previous publication (once it's published) and doing the new version and its publication alone. I guess it's still nasty but it's less nasty than what you said.
    – Trylks
    Jun 5 '14 at 17:44
  • Related note, Emanuel Derman wrote "He was a much better known (and a better) physicist than I was, and with his name on the paper, it would have been more widely read, and perhaps have led to further collaboration. But my pride stood in the way. Later that year, when another physicist carried out a similar analysis and received broader attention, Chris pointed out to me that I would have been better off if I had written the article with him. He was right. Years later, working on financial models at Goldman Sachs, I became much less compulsive about such strict authorial cost accounting."
    – alarge
    Jun 5 '14 at 20:34
  • To be fair, though, my previous comment is about the research scene of the old, before internet, and perhaps also with less of the "publish or perish" mentality. Nevertheless, I think both approaches have their pros and cons. I'd personally probably include the others as coauthors, certainly if they have had any input (some consider funding as entitlement to authorship, but I think that is acknowledgements-stuff). Is it common in the journals of your field to write who did what on the article (eg PNAS does this)? Not that this would appear in your CV where you just list publications.
    – alarge
    Jun 5 '14 at 20:43

I had discussion with several well-known professors with experience (30+ years in research) about this sort of thing, how did they list authorship in their early papers where they were the less well known authors. I started publishing only several years ago, so I will give their opinion (which I agree with I should add). Their response was the same across the board: the person who did most of the work they ought simply list themselves as first author when writing the paper and if the coauthors ask them about it, then they just discuss their concern with their coauthors honestly. If your coauthors are real scientists (regardless of whether they are more well known than you), and if your claim is true, they will agree with you.

If your coauthors are administrators mostly, and it's convention in your field and location, that they are added to your paper and listed ahead of you, consider working elsewhere, otherwise the problem cannot be solved in that environment.

There is another option: publish with them according to whatever convention is appropriate, and if you did not get listed as first author when you did actually most of the work, then at the same time also extend the work (>40% different material) on your own and publish a second paper in another journal as single author, covering both new and old material. (This is how you also find out who the major contributing author was on a particular work, where there are both team and individual papers at once.)

The first solution minimizes opportunity cost; the last solution requires more time and hassle for you, but you don't forgo anything academically, only some time writing a second paper and doing a bit more work.

  • I like the individual and team approach. Do you think I could put a paper on arxiv or figshare about my contribution as a sole author even if it overlaps with the team paper? It would be like a pre-print, but it would actually be a pre-collaboration. Should I make a new question specifically for this?
    – Trylks
    Jun 9 '14 at 9:56
  • You can certainly post such a paper on archive, and make a comment in the paper regarding the extent of your contribution to the team project. In fact, quite a lot of people, some of them very well known scientists do this anyway, one paper as a team and one paper by the main author covering their personal contribution to the team work (so of course there is overlap in the papers), their motivation (goal in doing the work), plus additional comments and results discussion in the individual paper to make it a paper worth reading on its own independent from the team paper.
    – user15282
    Jun 10 '14 at 22:56

My answer is straight foward.

If the original idea was from somebody else and the implementation is yours, publish the paper with the people who had the original idea. Of course, you'll be one of the authors.

If both the idea and the implementation are yours, publish the paper as you are the sole author. Pure and simple.

  • The problem with this is that they could claim the idea was theirs but you decided to go on your own and do everything without them. It's the word of one person against another, in a situation where one person has a tenured position and a reputed career, while the other one has a contract that finishes in a few months and a thesis defense pending. I agree it's pure and simple, but this could lead to horrible consequences.
    – Trylks
    Jun 9 '14 at 9:53
  • @Trylks The intention of this answer is to make things pure and simple. If you have dispute with "them" about who had the original idea, you need to resolve it with them before you publish the paper. If you believe you have the original idea, publish it yourself. Please note the word "original" I use in my answer and this comment.
    – Nobody
    Jun 9 '14 at 11:10
  • Well, I have an idea about how to improve something that is already done, I haven't shared it with anyone and nobody has commented anything on this, they are happy with the current version and have another idea for the improvement that will not work at all, IMHO. Therefore there is no possible dispute, however I can foresee at least raised eyebrows if I publish alone with no prior notice. In the worst case scenario I'd be considered as non-collaborative, I'd be isolated from everything but programming and my contract would not be renewed. No recommendation letters for me either...
    – Trylks
    Jun 9 '14 at 11:59
  • @Trylks I think the issue is to what extent the improvement is. If the improvement is not too far away from the original idea, they should be on the author list. Again, I think the key is who had the original idea. IMHO, you should not worry about the consequence if you do everything right and ethical.
    – Nobody
    Jun 9 '14 at 12:43
  • I don't think the closeness to the original idea is an issue (with the agreement from the reviewers) if the original idea has already been published with the full author list, and is (of course) cited. But I'm worried that things don't seem to be as they should in many cases.
    – Trylks
    Jun 9 '14 at 13:24

I can relate to your question. I've also works in projects where I'm the person doing all the programming/implementation. Actually, in some of those cases the people in question don't even want to work on the paper, so I guess you are doing better with your people.

I think that one issue to keep in mind is that an active collaboration, assuming the other people are in there with you, is helpful, because it helps keep one focused. I don't know if you have experience working on a project alone, but it can get pretty lonely. And working on an applied research project which requires both writing a paper and an implementation is a lot of work, usually over a long period of time, and it can be hard to keep focus. If you have actual reasonable collaborators (even if they don't do programming) as opposed to people who just want to put their name on your paper, they can definitely help to keep you focused, so in theory in can be an easier, faster, more pleasant process.

I do think that if both (a) much of the ideas come from you, and (b) you are doing the entire implementation, then you should be first/main author, so you should try to find some way to be recognized as such. Some journals now require a section where the author contributions are listed. Even if they don't require it, perhaps you could add one? Just a thought. This could help to emphasize that the bulk of the work is yours, if that is indeed the case.

  • Thank you for your suggestions. I have been working alone during my whole PhD and I'm used to that, actually I like it. But I don't want to close opportunities to working with other people, I want to be a good team player. If I could I would like to be good at everything (perfect), so I do my best.
    – Trylks
    Jun 6 '14 at 17:45
  • @Trylks Well, maybe you like it for now, but I predict you will grow tired of working alone. Also, it is simply difficult to succeed in modern academia without good collaborators, especially in applied areas, which yours seems to be. Or putting it differently, it helps greatly to have good collaborators. In the long term it is not healthy to be isolated in research, as in life. Jun 6 '14 at 19:07
  • I agree good collaborators are important. But bad collaborators may have a negative impact and being alone may be better (or less bad) than this bad collaboration. I agree that being alone is not good, and one of the reasons why I moved here was exactly that. I like loneliness for the freedom, I don't like that you have to be the expert in everything, any mistake will only be detected by the reviewers and specially I have to implement everything, but I got used to that. Now I still have to implement everything, but 4 authors appear out of nowhere, thus I think I'd better move somewhere else.
    – Trylks
    Jun 9 '14 at 9:38
  • I short, collaborators may as well be a help as an hindrance.
    – Trylks
    Jun 9 '14 at 9:50
  • @Trylks: Yes, you seem to have a good handle on the situation. Remember that working alone is much slower unless you are super-efficient and super-motivated, and in the long term you will not be able to keep up with everyone else without help. But yes, bad collaborators are useless. In some ways they may be worse than useless, because they will expect to be consulted about things despite not doing any work. Ridiculous, but it has happened to me. Jun 9 '14 at 9:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.