3

So I am a master's degree student in math at a local university. I am very dissatisfied and surprised with the type of work that my adviser asked me to do. I am unsure if I have bad attitude towards work or what is happening is already inappropriate.

The task I was given is extremely routine; it was mind-numbing. It involves a LOT of symbolic manipulations that follows a few set of rules and I apply them repeatedly. To give you guys an idea of how much computation by hand, I have done; I have used more than 30 pages and I write really small (around 0.5cm height and 0.3cm width) with very few white spaces(less than 0.5cm) so this may take other peoples a 100 pages or even much much more. One can imagine the high likelihood of making a lot of computational error due to sheer size of the number of computations. I have checked my computations a few times already, and I am still finding errors here and there. I have asked my adviser, if we can use software to do this computations and he refused without giving any reason why other than being embarrassed(I did not ask why.). Obviously, we have to make sure my work is a 100% correct so we could conjecture correctly, but I have no idea how we'll manage to do this manually.

Hence, I am really finding it difficult to be not cynical about my adviser. I feel like it's a stalling strategy, but perhaps not?

edit

@Anyon, in response to "Can you at least check your pen and paper results using software or is he against that too?"

I have asked if I could allot time to study and use software on my own and he said we should do it near the end, but that means any errors may, of course, negate results; which I don't think makes any sense because it seems very inefficient. I don't understand how he plans to proceed without any computer checks. He also said that he'll check my computations, although he did not say if he'll do it by hand or by computer. He also mentioned that, he had use software during his dissertation and have now forgotten how to use it which means he can't help me with software.

update I have succeeded in getting my adviser to reveal his motives. Well, he has no motives. As everyone below who answered have guessed, my adviser, simply, does not want to use software because there are technicalities in the mathematics that he does not know how to implement correctly to the software, neither do I. Wolfgang Bangerth is correct, there was a serious breakdown in communication. I am happy this is settled. Thank you to everyone who've answered.

  • 6
    What reason did you advisor give for not using software? As a computer scientist, my first reaction to boredom is always automation, but there might be reasons it is not a good idea in the particular calculations that you are doing. – jakebeal Jul 20 '19 at 12:31
  • @jakebeal: The calculations were like high school algebra minus commutativity and a couple of rewriting rules, i.e. replace this symbol with that symbol. He said it would damage his reputation. Not sure if he was joking or not. – TheLast Cipher Jul 20 '19 at 12:48
  • also, he did not mention why it would damage his reputation. – TheLast Cipher Jul 20 '19 at 12:54
  • 4
    This is extremely bewildering. Your advisor's response doesn't seem to make any sense. – knzhou Jul 20 '19 at 17:54
  • 5
    I feel like we are missing some information. – Anonymous Physicist Jul 21 '19 at 7:44
12

Since everyone here seems to scratch their heads about what might be going on, my first reaction is that there is a breakdown of communication between you and your adviser. It clearly doesn't make sense to do 100-page long computations on paper if they can be done by a computer. So the reaction of your adviser doesn't seem to make any sense -- but what stands out to me is that you don't seem to talk to your adviser about the situation.

Talk to him about the issue. Have a conversation, understand his motives, discuss why he wants you to do these calculations by hand and why he doesn't seem to trust the software. He may have good reasons, but you'll never find out if you don't talk about it. At this point, it's all speculation when you could have a conversation to actually understand.

  • 1
    I have talked to him, but like you said there might be a breakdown in communication. I'll try again, though, and be more explicit about everything and try to get him to clarify his motives, especially the almost endless computations I did. Thanks! Will update you guys. – TheLast Cipher Jul 21 '19 at 13:57
  • You were absolutely right. I have now updated my post. Thank you! – TheLast Cipher Jul 23 '19 at 6:40
  • Ah, very glad to hear! I think half of the posts on this forum essentially boil down to the answer "Have an honest talk with your adviser." – Wolfgang Bangerth Jul 24 '19 at 2:33
5

I have asked my adviser, if we can use software to do this computations and he refused without giving any reason why other than being embarrassed(I did not ask why.).

He said it would damage his reputation. Not sure if he was joking or not.

He also mentioned that, he had use software during his dissertation and have now forgotten how to use it which means he can't help me with software.

This doesn't sound like an adviser trying to stall you. Instead I'd guess that he is essentially saying "I don't know the software, so I won't trust results derived from it and only it, nor can I help you use it. Hence I'm also not willing to put my name on/near results derived only by software, as there could be errors I can't check, which could damage my reputation."

The algebra steps you described seem like they could be implemented in e.g. Mathematica reasonably easily, especially if you already know the basics of the software package. However, there's certainly something to be said for doing it by hand too - if both methods arrive at the same answer that tends to make you more confident in your results. There's also something to be said for trying to reach analytical result before deriving the same result using computer algebra, as you'd be less likely to fool yourself into mistakes by working towards checking a specific result. A result that may be incorrect, by the way, if you've made any coding mistakes/ Such checks along the way (not just the end) can be a valuable tool, as can any sanity checks of the resulting expressions. Maybe proposing doing so that would be a decent compromise.

  • Is it possible that he does not want to do the computations himself so he's letting me do it? Also, since he does not know how to use any computer algebra software, then he'd rather not let me use it; because for the reasons you have just mentioned? – TheLast Cipher Jul 21 '19 at 12:07
  • I also tried asking for a different problem and the current one is the only thing he has for anyone right now. – TheLast Cipher Jul 21 '19 at 12:15
  • @TheLastCipher Is it possible? Yes, but it's also possible that he thought this would be a good project for a student, and thus opted not to do the calculations for more altruistic reasons. It's also entirely possible that this project turned out to be more involved than he expected, or even that he knows about a subtle way to reduce the number of calculation steps needed (like a symmetry argument). I don't think we can tell from this question, so you should probably just discuss this with your adviser. But I wouldn't go into that meeting suspecting stalling or malice. – Anyon Jul 21 '19 at 14:17
  • Thank you for the advice. I guess it is just in my personality to expect the worst; hence, the suspicion which I am trying hard to fend off. – TheLast Cipher Jul 21 '19 at 14:20
2

I doubt that your advisors motives can be learned except from him. There are too many options, but I doubt that simply stalling you is one of them. No one gains from that, I suspect. I can only speculate on why this is happening.

Perhaps the advisor is simply distrustful of computer programming for this sort of thing. An error in coding can throw you off worse than errors in human computation, perhaps. Maybe he has been burned before.

Or perhaps he thinks that you don't have the computer skills to do it correctly and thinks the time required to obtain those skills and develop the correct program will slow down the path to the end.

Perhaps the advisor thinks you need some sort of discipline in computations (or in general) to make sure that you are learning lessons he thinks are important of vital. Maybe this isn't the most likely scenario, unless you have given him reason to think you need that discipline.

If you want a way out, and have the computing skills, you might develop a small program for a small part of the work, using something like Test Driven Development in which you can show the advisor that the codes are correct, through the tests, and are producing the correct results. That might change his mind if it is the first scenario and might also find errors that you are making with hand calculation.

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.