Dorky Poll: Optics or Thermo?

Thermo. That was an easy one.

Thermo.

As a biologist who's already taken some thermo and stat mech, I'm going to go with classical optics. It would both improve my photography, microscopy, and spectroscopy technique, and would be infinitely more practical in real life than knowing the miniscule amount of extra thermo.

Thermo here too (but I teach the course, so I'm biased!). Doubly so if thermo includes some statistical mechanics, as it often does.

Classical Optics.

As a slight aside (since we're talking heat and classical optics), there's an article in Nature (found via Ars Technica) on femtosecond plasma holography that you might find interesting, Chad. The diagrams are nice and explanatory, though I've forgotten everything I ever learned about holography, so...

Thermo, especially if you include some statistical mechanics as a previous commenter said. A strong grounding in these will help you in just about any other area of physics (and probably chemistry as well). Also, I think that thermo is harder than optics, so the chance to learn it all without having to do the work is pretty attractive.

Thermo.

It's more-- much more-- directly applicable to what I do on a day to day basis, and as the years go by, I ahve to take more and more responsibility for things that affect my Stuff, but the analysis for which I have not done myself.

In short, it would help me sanity check my junior mechanical engineers.

I was once faced with a similar choice. When I was a secondary-ed science major I had to take a physics elective. The suggested elective was Optics. I choose Intro. to Modern Physics. The class covered thermodynamics, quantum mechanics and special relatity. Not very in depth, mind you, it was a sophmore level course. I figured modern physics would be much more interesting than mirrors and lenses. But I've always wondered - how much more is there to optics than Snell's law?

Classical optics.

I already know plenty about both subjects, but I despise thermo with every fiber of my being, and I like optics quite a bit.

Both subjects strike me as quite boring. But I would choose thermo, because if I knew everything there was to know about it (I am reading this as not limited to current knowledge), chances are I would get some clues about the answers to some interesting questions, i.e. black hole entropy, quantum gravity, etc.

Optics, no question about it. If nothing else, it has a much more immediate connection to anybody who's ever picked up a camera or binoculars, which, judging from the sales figures of digital compacts and cellphones with cameras, is just about everyone. Please?

Besides, thermodynamics? "It'll even out over time" - what else do you really need to know?

Thermo.

Real-world optics lessons = bad photographs.
Real-world thermo lessons = burned skin.

Thermo

Optics.

I had to take Thermo, and I hated it. I did not have to take optics, and I am an astronomer.

Not much new in classical optics since lasers, holograms, helical polarization, and negative optical index of refraction.

Whereas, on the other hand, thermodynamics has Nonequilibrium stuff that won a Nobel prize, and dissipative structures (even if Prigogine's Nobel was for something that didn't work out), and black holes, and other weird and wonderful things.

I say, "Thermo!" Of course, I may only be a Boltzmann brain in a universe at heat death...

Optics. It's what I do for a living these days.

I just finished reading "Why the Sky is Blue" (http://www.amazon.com/Why-Sky-Blue-Discovering-Color/dp/0691124531) and it had me thinking about this very topic.

A single point scatterer, e.g. an air molecule, will scatter isotropically in all directions perpendicular to the incident polarization.

This result let us understand the polarization of sunlight and its blue color.

However, two point scatterers will create an interference pattern and N (large N) uniformly spaced scatterers will interfere so that scattering will occur only in certain special directions. Finally, an inÂfinite system of regularly and closely spaced scatterers will not scatter at all.

i.e. Uniform systems do not scatter light.

Why is the sky not transparent then?

Fluctuations in density of the air. (e.g., a variation
of the mass density of a gas occur due to thermodynamics / Brownian motion)

So if you want to know why the sky is blue you need optics plus thermo.

So to answer the question:

Which would you rather know more about, Classical Optics or Thermodynamics?

I want to know whatever is necessary to solve the problem.

Disciplines only exist curriculum organization purposes.

/experimentalist

I was going to say thermodynamics, because it is more interesting, but I think I would choose optics since I think it would be more profitable.

This entirely depends on my motivations. i disliked thermo and felt like I never understood anything. It seemed to be random definitions of vague quantities by taking the partial derivative of something with respect to something else. (Let's see, the partial derivative of S with respect to V, and we'll call it, "Entalolopy", for which we'll use the letter R.) So I might pick thermo, because this magical being represents the only way I could ever understand thermo. And I actually do already have a decent understanding of classical optics.

On the other hand, I like classical optics a whole lot more.

Thermo, absolutely. You can never know enough about limits of the universe. It informs chemistry, biology, geology and information theory. Without classical thermodynamics and statisical mechanics, you can't really understand any energy-limited process, which is to say: all of them.

/experimental-environmental

Thermo, but done as Stat Mech as information theory!!!!!

Thermo (including the statistical description of course) because it is basic.

[And in my case because I've studied both, but thermo has advanced and seems to open up to more cool (or is it hot :-P) effects as Vos Post mentions.]

Thermo can include optics can't it? I think Stat Mech is physics, Thermo is a discipline unto itself. IMHO Stat mech is more fundamental. I mean can't I find a Gibbs free energy for an EM field, and use that to do optics? Or do stat mech for a quantized field, and then go all eikonal. I mean you can do thermo with various energy functions, treat it all as empirical, , consider entropy as a liquid that flows, and go on with no use for Newton, Maxwell, Schrodinger, etc....

Now I'm talking purely in terms of their order in the "tradtitional" reductionist attitude of a theorist. Carnot and thin lens are both on the MCAT, engineers need Carnot but maybe not lenses, so at many schools taht have a large engineering base, thin lens is not in the calc based course, but it flips a bit if there is a separate calc (or non calc) based course the premeds and other squishy science types take. We have no large egr school, tons of premeds, but we do a stat mech based version of thermo as befits a liberal arts approach to doing something more fundamental. I didn't make that decision but I agree with it.

Having said all this, heat engines and thin lenses arfe pretty hard to get excited about! At least for me.

Most definitely Non-equilibrium Thermo and Stat Mech. I could then solve the Navier-Stokes equations and collect a Million $.

Thermo. Although optics is more interesting and quick to yield results in experiments (mirrors, prisms, lasters) and easier to measure, it is sort of limited compared to Thermo. On the other hand measuring and calculating heat flow is significantly harder compared to light rays bouncing up and down. Also there is that pesky radiation thing. For engineers (which I am one) thermo provides much more interesting things to work on (heat exchangers, engines, flows, pipes, resistance). For all matters, basic thermo laws are equivalent to electric laws and currents in calculation altough it is easier to prototype electric/electronic circuits.

It's not sexy stuff but it is interesting to learn. Optics? Sexy but you run out of things to do soon.

Signed: A mechanical engineer working in IT

I would rather know about classical optics (it is full of tasty morsels like the Airy function, complex indices of refraction, and other quantum-related - but not quantum based - wave topics), but I would rather teach thermo.

Why? Because I truly hated thermo when I took it, and know that some of my grads heading into mechanical engineering hate it also. Sounds odd, but (if you are serious about teaching) you tend to do a better job teaching topics you hate and struggled to make fit into a sensible logical structure. A thermo course starting with quantum stat mech rather than "classical" thermo with its array of partial derivatives comes to mind. Looks like I agree with Perry on that aspect of "thermo".

I'd mention to the ME in IT that you could spend an entire semester just dealing with all of the mathematics needed to get close to understanding the supernumerary arcs in a rainbow, but the real money is in optics improvements needed for military and civilian applications in robotics and the like.

Thermo thermo thermo... There's no competition. Optics is very interesting and potentially useful for a lot of people, but thermo is *basic*. My thermo is terrible. Oh, the shame!

Thermo! Even if you don't include stat mech, thermodynamics is so incredibly broad in its applicability. Also, the mathematical structure is quite interesting in own right.

Thermo for sure, for the same reasons that Josh cited.

Optics. I think I know enough thermo to get by, but I'd love to be able to design (and more fully understand) telescopes and cameras.