What is it like to do scientific research today?

[Marin]

Hi all!

I'm doing my bachelor in physics (have presumably one more year left) and I already started concerning about which area of physics I want to specialise in, since I have to do a Bachelor Thesis in a semester from now.

I also don't know if I want to stay at the university after finishing my entire education (I'd maybe try a PhD) or try working elsewhere (of course, I don't have the slightest clue where and what). Unfortunately, both issues turn out to be at least as tough a choice as an important one, so I'd be glad to hear some advices from you :)

Here's my situation:

I like the subject and very much like solving puzzles so I'm ready to do what it takes to get an answer to a problem. I'm also doing lots of math courses parallel to physics, to help understanding the underlying technical details :) but maths is also interesting enough on its own. What I like most is applying mathematics to physics providing new insights and a deeper understanding. I wouldn't bet on theories that cannot be proven experimentally in, say, my lifetime either. [I'm aware this sounds egoistic but otherwise I would try doing pure maths]Although I don't have any deep contact with condensed matter theory and plasma physics yet, I find effects like superfluidity, superconductivity, Bose-Einstein condensates and similar very interesting and would consider trying to go into these areas. Recently I have heard that research in these fields consists primarily of doing computer-based simulations or some other numerics which was kind of embarrassing since it's not what I actually like most.

So what do you say? Does theoretical research in most areas of physics today consist of simulating stuff and doing programming or are there sufficiently many (in order to find a job/position, to publish regularly etc.) areas (which?), where one can try improving the given theories? How does this look like for the areas I mentioned above?I'll be glad to welcome anyone opinion :)

Thanks a lot

 

[someperson05]

I can only say what I've noticed so far as a undergraduate myself. (It makes sense in my head.)

There is a growing field of computational science, thanks to the ever increasing computational power available. Computational science is, from my understanding, a discipline designed to learning how to design simulations and explore science from that aspect. I would think, that given the current state of physics where one has to wait years for something like the LHC to be built, simulations are the best that most theoretical models have.

Computational science has been sort of illustrated as a new way of gaining scientific knowledge.
Before theory -> experimentation.
Now you have sort of this mid ground, theory -> simulation -> experimentation.
I'm sure its hard with the sufficiently complicated theories to really know what the hell you are looking for, hence the simulations. It can give you results to look for in experimentation. Or that is my understanding.

And you say the simulations wouldn't help with improving a theory.
Of course it could.
So you come up with some theory, but given its complicated structure and formulation, its hard for even the best theoreotical physicist to test the theory in his mind. He needs some sort of playground, some sort of laboratory. That is where the computer simulation comes in. He can explore if his theory does what he thinks it does. Apply it to different situations, watch the outcomes, and then manipulate his theory after seeing the theoreotical implications.

Again, this is a undergraduate opinion, from someone who was once strongly interested in pursuing computational science.

EDIT:

"It is also different from theory and experiment which are the traditional forms of science and engineering. The scientific computing approach is to gain understanding, mainly through the analysis of mathematical models implemented on computers." From the wiki page for computational science.

 

[Marin]

I see there's no way around computational science but to what extent are theoretical physicists doing it (in 20%, 50%, 80% of their everyday research/work?)?

Is it the case that they're doing simulations more than using mathematics while developing their new theories? How does this balance look like in the different areas of research?

 

[someperson05]

Again, I have no actual experience, only a perception, but as no one else has responded I am trying my best.

I would think the simulations are not quite as important as the mathematics. The simulations in fact would rely on the mathematical model to be built in the first place. So I would think it would be something like
build model -> simulate to see what happens with model -> refine and improve model -> rinse and repeat
I would think that in today's world where collaboration is really important, and that people are specializing more and more, you as a theoretical physicist probably would collaborate with computer scientists who would be developing the simulation. You would probably only need some rudimentary understanding of what they are doing.
So back to one of your original concerns, I wouldn't think you would be doing so much of the programming yourself.

I think simulation offers theoretical physicists a new way to see what their models are saying. In my opinion, it would be cool to see a model you've developed come alive.