So How do I get Research Level good?

[aLearner]

So I've been wondering, how much of a subject do you study before you begin research on the subject? And how do you know when you're ready? Say that I'm working on general relativity. How do I:

1) Get really good on the subject?
2) Know when I'm ready to do research on the subject?

Also, any advice on mastering a subject in physics? Do I just work out all the problems in the book? Or is there more to it than just that?

-aLearner

 

[mfb]

Visit lectures, work on problems in exercises/books, try to find some research group and ask them.

 

[Choppy]

Usually you start out with research at the senior undergraduate level. Here you have a fresh foundation which is largely based on textbook and laboratory exercises. In most cases research at this level is almost completely directed by your mentor who tells you what papers to read, what code to program, what measurements to make, what variables to graph, etc.

Next you move on to the beginning graduate level. Here, the work is still directed by a mentor, but you have more freedom in what you want to do. You're expected to start reading on your own, start planning out the research long term and make your own (supervised) decisions with respect to how the research is carrier out.

You then move on to the senior graduate level where you are (or at least should be) given the freedom to explore as you see fit (at least within the confines of your project). You design and implement the projects yourself. Your mentor moves into the role of a close peer with whom you can speak regularly for advice.

After that you should be at a level where you are reading independently and coming up with your own ideas for projects. You won't always have the opportunity to explore them of course as you generally have to go where the work is for post-docs. But at this point you should at least be capable of independent research if given the opportunity.

So it's not really a question of passing a specific threshold, rather, the ability to perform research tends to be a progression towards intellectual independence.

 

[negru]

It depends on the topic and the type of research you want to do, and I'm not sure about GR research specifically (not sure if there's anything going on there besides numerical stuff, if you mean just classical GR), but I'd say that the equivalent of a related course is more than enough to get you started.

Do you want to do theory, numerical simulations, experiment? Do you mean quantum gravity? Do you have any specific advisor in mind? Or do you want to work mostly on your own?

In any case, my advice is to pick the relevant research papers, and work down from there, not the other way. Look up words you don't know, read relevant references, etc. Most of what is taught in normal courses (including advanced graduate courses) or normal textbooks is completely irrelevant to current research. For example I work on quantum field theory stuff, and 90% of anything in say Peskin and Schroeder is just useless.

You don't have to be really good on the subject. You don't need to know how to solve every exercise in the textbook to start doing research. At most, you should go through a course or a textbook just to know where to find the results you need, when and if you need them. Though honestly, wikipedia is much better at that anyway.

 

[aLearner]

Wow I never knew that negru. Well, quantum gravity does sound interesting, but I'm just a sophomore undergrad who is interested in theory. School bores me, I prefer studying by my self at my own pace. But it is a lot easier to get help at school, which is why I prefer self studying at school on my own free time. Hence, I was wondering about research. Right now I'm focusing on general relativity, and trying to get a feel of the subject. Then I'll probably move to the quantum field theory, where I will get a feel of the quantum level. And then, when I feel ready, I will join others at combining the two. But I never knew I didn't need to know everything to do research. But I guess, if I want to do good research work, I would need to do everything