I started my PhD two years ago on a specific problem. The problem is to study a given computer science structure from one specific perspective. My current advisor is an expert in that perspective and I got some results and publications with him.

However, during the last 6 months, I started to worry about my PhD thesis as there are no more room for novel results in that direction (several people tackled it recently). There are still some room for minor results and my advisor urged me to tackle them for my PhD thesis.

I am no longer interested in pursuing that direction and shifted towards the other perspectives of the problem as I believe there are still major results to come out of them. My advisor has no objection on pursuing these directions by myself (at the end he will put his name in whatever comes out of it!). Recently, I took a class with another professor and tackled one of the important directions of the problem as a class project and had incomplete but interesting results. The class professor was very happy with my work and advised me really well during the class semester to start tackling the problem. After couple of months of working, it seems I can really get something nice out of this.

I started to believe that, for my PhD thesis, the best thing is to have the class professor as a co-advisor. This will allow me to put in action and test rigorously several ideas I have in the other direction and make my PhD thesis more valuable.

I spoke with my current supervisor about the co-supervision but he insisted to make it as a collaboration instead of co-supervision. He has good relation with the other professor. And if it matters, I have my own external funding with no financial obligation from the current supervisor.

What should I do in such case?

EDIT on the collaborator vs co-advisor thing: Because I had a specific problem and had a class with him (which makes it easy to work on the other direction) and the results were almost ready couple of months later, there was a collaboration. My advisor is not willing to collaborate again and urging me to do the minor results instead of pursuing the other direction. The reason I look for co-supervision is

  1. Consult him in many scientific results I am not good at (neither my current advisor).
  2. No commitment in advance for collaborations/ papers.
  3. co-advisor may suggest a new perspective to the problem instead of accepting/rejecting what me and my advisor suggest.
  • 8
    What's the difference between collaboration and co-supervision? I'm not sure why it matters to your advisor, but then again I'm not sure why it matters to you. If you do good work, write interesting papers, and get good letters of recommendation from both faculty members, then I'm not convinced it matters who is listed as the advisor. But maybe it carries some extra importance in your situation. Commented Nov 7, 2014 at 21:33
  • 3
    @seteropere but why would whatever tag you put onto your collaboration with the other professor matter for (1) or (2)? I mentor a few people who I am not formally advising - why would I care as long as we get joint papers out of the deal? It's not as if it "co-supervising students" has any particular bragging rights. And why would your ability to work on a specific interesting problem depend on whether the other professor is a co-supervisor or a collaborator?
    – xLeitix
    Commented Nov 7, 2014 at 22:34
  • 2
    @seteropere: Perhaps you could clarify what you think this other researcher would do as a co-supervisor that he would not do as a collaborator.
    – Jim Conant
    Commented Nov 8, 2014 at 1:47
  • 10
    Brown's Rule for finishing the Ph.D.: Listen closely to what your supervisor tells you to do and do it. You can do the other research later.
    – Bob Brown
    Commented Nov 8, 2014 at 3:03
  • 2
    My advisor is not willing to collaborate againIt's your PhD. You are free to collaborate with anyone you want. If your advisor doesn't like it, find a new advisor.
    – JeffE
    Commented Nov 8, 2014 at 20:37

1 Answer 1


I think @BobBrown summarized everything that needs to be said.

Brown's Rule for finishing the Ph.D.: Listen closely to what your supervisor tells you to do and do it. You can do the other research later.

In detail:

  • You have some publications from your cooperation with your main supervisor. Excellent.

  • Publish another paper with your new collaborator, if you have not done already.

  • Then do another paper (those minor results) with your advisor. Finish everything there is to be done on this direction fast.

  • Finish PhD.

  • Do whatever you want afterwards.

This is by far the easiest route to your PhD.

  • I edited your answer to include the quoted comment. If you feel it is irrelevant, please roll it back to the previous version.
    – enthu
    Commented Nov 8, 2014 at 8:42
  • 2
    @seteropere Why not the best? You can work on the "other" idea on your free time and work on your supervisor idea on official time. Then you are happy, he is happy
    – Alexandros
    Commented Nov 8, 2014 at 18:15
  • 3
    Downvoted. PhD students don't have "supervisors"; they have advisors.
    – JeffE
    Commented Nov 8, 2014 at 20:38
  • 3
    @JeffE I do not think it is the word that actually bothers you. Why do you downvote? The OP has produced solid results by both professors. Why don't you agree the OP wrapping things up on their first research direction, prepare the PHD thesis and presentation instead of wandering towards unknown waters that may pull her back a year. If she wants the other direction better, she can always try it later or on her free (Google equivalent of 20%) time.
    – Alexandros
    Commented Nov 8, 2014 at 21:01
  • 5
    I do not think it is the word that actually bothers you — It is not only the word that bothers me; it is also the meaning. Yes, of course PhD students need advice on how best to achieve that goal, and listening to their advisors is important, but the blanket rule "and then do it" is too simple. PhD students are learning to be independent researchers. They cannot learn to be independent by following orders.
    – JeffE
    Commented Nov 8, 2014 at 23:39

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .