Doing my PhD in a multidisciplinary project -- I need advice please

[Michelle]

I started my PhD in physics at a position associated with a project. In the project I'm the only one working in my field and the only one working full-time that knows how to program (a computer scientist is also working on the project but only 50% and has another supervisor in his field). I entered the PhD as computational scientist/physicist, and I was assured there were also science behind and scientific contributions I could make.

I have no supervisor, since my boss at the project knows nothing about software, programming, mathematics or physics, and the supervisor I have at the university is only for paper work. My boss help is quite vague and he always comes to us with ideas that he has not thought for more than a day, then changes them and think is really easy to program them.

One of my colleagues (also phd student) has made several unpleseant comments about how unimportant is my job for the project and is not speaking to me anymore since I told him to stop it the last time he was out of the line with me (months ago). My boss does not care and has done anything about it. So the environment is not the best.

After two years I have realized that I have not done any meaningful work towards my thesis. I have "only" programmed a software for "real scientists TM" to use, and all my questions regarding science have been relegated to my free time, since the software was required to meet project deadlines. I have been the only PhD student in my group required to make project deliverables entirely by myself, and thanks to that I didn't even got to go home for Christmas (which now I really regret).

After the last one, I wanted to give my thesis a push, so I took one of my bosses ideas (suggested by him to me) and decided to implement it. I made the skeleton for a possible paper, and started to work in an algorithm. Explained what I was doing to my boss, and he said ok. But just two weeks after he came with a new idea, and has told me to drop it for now and do it in my spare time. Of course this new idea involves again heavy monkey coding instead of an algorithmic contribution I can use for my thesis. For several reasons involving that I'm messed up if not, I have to comply.

This has upset me a lot. I took a few days sick because I just don't want to go to work. I really want to quit. I have no support in making my PhD, I feel they are taking advange of me by paying me the salary of a PhD student and have me do the job of a software engineer and I have no way of defending myself.

I am over 30 now, and I'm afraid I would not be able to find another PhD. Also all the time invested in the software would be gone, and how do I explain to a future employer what I have been doing the last 2 years? I also think it is my fault for not investing more time, being more productive or more intelligent to take on everything.

Is this normal? What's my best move? Change to computer science (although I would have the same problem of not making meaningful contributions to computer science)? Could I use the software in a thesis in Computer Science? Quit without a plan B is not an option, as I need the salary.

Is the PhD not for me? My colleague keeps telling me that there are worse supervisors, but then makes me wonder maybe I'm the problem. If I change the PhD, I'm afraid this will happen again if I'm the problem. Also I really like the field and what I'm doing, even the routing programming, it's just is not the only thing I want to do. Honestly I would have been happy just to expend a few more weeks working on the algorithm, polishing it and producing input for an experiment.

Any other computational scientist feeling the same?

<< Mentor Note -- post edited to remove profanities >>

 

[Dr. Courtney]

It seems like an unusual situation. Where's the funding coming from? Who is the PI on the grant? They are likely the person really in charge with whom you need to have an honest discussion of your frustrations and possible paths forward.

You may be caught in a situation where they needed a computer programmer but only wanted to pay for a grad student, so they created a phantom PhD opportunity that is too poorly defined to realize. But I'd dig deeper before drawing this conclusion and jumping ship.

 

[Michelle]

It seems like an unusual situation. Where's the funding coming from? Who is the PI on the grant? They are likely the person really in charge with whom you need to have an honest discussion of your frustrations and possible paths forward.

You may be caught in a situation where they needed a computer programmer but only wanted to pay for a grad student, so they created a phantom PhD opportunity that is too poorly defined to realize. But I'd dig deeper before drawing this conclusion and jumping ship.

Thanks for the reply. The funding is from the EU and the PI is my boss, the project got +2 million euros. I have already told him that a software is usually not enough for a phd in physics (is somewhat a sophisticated software but it's not physics itself and even in terms of software I wouldn't know how to highlight as research contribution), but he doesn't seem to care, he says that if I can publish the software I can use it for the thesis, but I think he misses my point on doing research. Last software in my field that was published was almost 9 years ago, and what was published was an algorithm + experimental results, software was just a tool created so other researchers could use it.

Also, my boss has no experience in the field of the project, he is a young researcher stablishing his group, and has not publish anything there yet.

 

[Dr. Courtney]

Thanks for the reply. The funding is from the EU and the PI is my boss, the project got +2 million euros. I have already told him that a software is usually not enough for a phd in physics (is somewhat a sophisticated software but it's not physics itself and even in terms of software I wouldn't know how to highlight as research contribution), but he doesn't seem to care, he says that if I can publish the software I can use it for the thesis, but I think he misses my point on doing research. Last software in my field that was published was almost 9 years ago, and what was published was an algorithm + experimental results, software was just a tool created so other researchers could use it.

Also, my boss has no experience in the field of the project, he is a young researcher stablishing his group, and has not publish anything there yet.

Has your formal thesis proposal been approved by a committee?

"Publishing" software means different things to different people. Is posting it to an open source site good enough, or are you going to need a peer-reviewed journal article? If you need a peer-reviewed journal article, what tier of journal are you looking at? Publishing is easy. The higher tier journals are harder. But you are in a tight spot if you'll never know if it's good enough until you get the reports from peer reviewers or if your thesis committee says, "That journal is not good enough."

You need answers to these questions.

 

[Michelle]

Has your formal thesis proposal been approved by a committee?

"Publishing" software means different things to different people. Is posting it to an open source site good enough, or are you going to need a peer-reviewed journal article? If you need a peer-reviewed journal article, what tier of journal are you looking at? Publishing is easy. The higher tier journals are harder. But you are in a tight spot if you'll never know if it's good enough until you get the reports from peer reviewers or if your thesis committee says, "That journal is not good enough."

You need answers to these questions.

The approval has not yet happened, I wrote the thesis proposal with the help of my thesis supervisor at the univeristy. I send it to my boss but got no comments (I also left a copy at his desk). The problem is that there is a difference between what I want to do (my proposal) and what I end up having to do because my boss asks me. This proposal should be aproved/rejected within a month.

I want a peer-reviewed publication (that's why I need the experiments, which my boss has to approve and pay), so does my boss. I don't care about impact, it would be my first paper and if it's published and read I would be happy, but my boss aims for huge impact factor (I'm talking nature or science), that's why he disregarded the other idea I was working in. I think this is highly unrealistic, and I would like to focus on a nice contribution, finishing it into a paper, then we see where it lands. The software would be also open source by project requirements, but right now the software contirbution is that it was done, it's what they are calling "enabling technologies" in software from what I read.

For my thesis I don't really need anything specific in order to pass it, at least there are no such requirements at my university. But in general people usually go for 3 papers as first author, although 1 or 2 decent ones are enough from what I have heard.

Thanks for your help.

 

[Orodruin]

Just to say that you can publish work related to physics software tools. It is done quite often and I have done so myself. You would not be looking at a Nature or Science paper, but probably at some journal such as Computer Physics Communications. It is quite broad in terms of the physics. My CRC paper was written regarding a plugin adding Markov Chain Monte Carlo capabilities to an existing software that is used to simulate long baseline neutrino experiments.

 

[Michelle]

Just to say that you can publish work related to physics software tools. It is done quite often and I have done so myself. You would not be looking at a Nature or Science paper, but probably at some journal such as Computer Physics Communications. It is quite broad in terms of the physics. My CRC paper was written regarding a plugin adding Markov Chain Monte Carlo capabilities to an existing software that is used to simulate long baseline neutrino experiments.

Thanks! I didn't know that journal, I will save it. My work would be within the scope of "Computer Programs in Physics Papers".

In terms of future employment, in case i don't get to do science, how are these software contributions regarded for future post-docs? (i.e. implementations of already existing computational methods, but in a framework that allows the user to do something new).

 

[f95toli]

I have no supervisor, since my boss at the project knows nothing about software, programming, mathematics or physics, and the supervisor I have at the university is only for paper work. My boss help is quite vague and he always comes to us with ideas that he has not thought for more than a day, then changes them and think is really easy to program them.

Where are you in the world? Since you are working on an EU funded project I would assume it is somewhere in Europe.
You mention your "supervisor at the university" and from what I understand from your description he/she is not your day-to-day supervisor? Does that mean that the person supervising you is your industrial supervisor (or something similar)?
Regardless, there should be some sort of university/department level oversight over all the PhD projects and they should be able to step in if you are not making sufficient progress towards your thesis. Are you not required to e.g. submit annual progress reports?

It is worth keeping in mind that the department/university does have an interest in you successfully getting your PhD; unsuccessful PhD projects do not look good and in some countries (e.g. here in the UK) there can even be a financial penalty from the funding agencies (the university only gets fully reimbursed once the student graduates).
My point is that you should not assume that the university/department will automatically side with your supervisor if you raise your concerns with them

 

[Michelle]

Where are you in the world? Since you are working on an EU funded project I would assume it is somewhere in Europe.
You mention your "supervisor at the university" and from what I understand from your description he/she is not your day-to-day supervisor? Does that mean that the person supervising you is your industrial supervisor (or something similar)?
Regardless, there should be some sort of university/department level oversight over all the PhD projects and they should be able to step in if you are not making sufficient progress towards your thesis. Are you not required to e.g. submit annual progress reports?

It is worth keeping in mind that the department/university does have an interest in you successfully getting your PhD; unsuccessful PhD projects do not look good and in some countries (e.g. here in the UK) there can even be a financial penalty from the funding agencies (the university only gets fully reimbursed once the student graduates).
My point is that you should not assume that the university/department will automatically side with your supervisor if you raise your concerns with them

Germany. I have an industrial supervisor, and someone in the university who just signs the paperwork and is not really close to my field, but it was as close as I could find. Thanks, I will try to talk with my supervisor at the university without giving too much detail, just in case.