Saturday, April 3, 2010


It occurred to me recently that it might be nice to log what I'm doing in the world of physics from time to time. When I was teaching I had interesting stories from work almost every day but very little time to actually write them down. Now that I'm in grad school, the opposite is true: I have ample time but a dearth of interestingness.

This past year has consisted mostly of me going to class and working problem sets. It's been interesting for me personally, but it doesn't really make for dynamic writing. "Isn't graduate-level physics where all the exciting stuff happens though?" Yes and no. Most of it is so steeped in mathematical formalism that it becomes interesting more as a set of tools than a collection of ideas. There are still interesting ideas out there, but they're rarely the focus of the required reading or problems.

The more I learn about physics, the less I enjoy digging into its implicit philosophical quandaries. Maybe I'm too much of a pragmatist, but as I do more and more physics, I realize exactly what physics is: a set of mathematical approximations describing physical phenomena. The descriptions can be awesome, but in my mind there's a difference the predictions themselves, and the way we obtain them.

What is a Green's function? Does it have physical meaning? Not as far as I can tell, but still it pops up again and again throughout all branches of fundamental physics. It's a really useful mathematical artifact that solves a wide variety of problems, so we use it. When it's useful to assign a physical meaning to something, we do it. When it's too abstract or convoluted, we don't. "Shut up and calculate" is a pretty useful way to view things sometimes.

I like problem solving. I also like big ideas, but I'm happy to put them aside most of the time to get things done. Still, in order for me to stay motivated, I need to know that there's something cool waiting at the end once all the work is done. This is why I'm probably going to end up in astrophysics - even if there's no less grinding than in any other branch of physics, it's to describe something that's (sometimes literally) orders of magnitude bigger and more awesome than, say, surface science or condensed matter.

I'm currently trying to arrange some summer research with a professor who applies large-scale fluid dynamics simulations to various problems in astrophysics. In preparation for a summer job, I've done/am doing a couple of things:

  1. Over spring break I went to a seminar on high-performance (parallel processor) computing. While I've done computational physics in the past, it was all relatively small scale. It's not too hard to re-envision writing code that will run in parallel, but it takes a different kind of thinking. While we didn't really do any useful examples, it was a nice introduction to MPI, the standard interface for parallel computing. Also, writing scripts and working in five linux terminals simultaneously always makes me feel like a hacker.

  2. I've been reading Thompson's An Introduction to Astrophysical Fluid Dynamics to get some of the basic physics down. Fluid dynamics is one of those things that it's possible to never have to do when you're learning physics. I touched upon it only once as an undergrad during a partial differential equations class. It's not part of the "standard" curriculum, and often gets overlooked, but it's fairly common sense stuff a lot of the time, and it seems like vector calculus was made for it. I'm currently working my way through material on accretion and shocks, which isn't intuitive yet, but is becoming so, slowly.

Hopefully as I start doing (as opposed to learning) more physics, I'll have more to report.

No comments: