Tuesday, July 25, 2017

Sleeping Beauties in Theoretical Physics: 26 Surprising Insights by Thanu Padmanahhan:

Thanu Padmanahhan's book, Sleeping Beauties in Theoretical Physics is a great find.  It's reasonably accessible for a technical tome on theoretical physics, it has an interesting framework that makes the connections between chapters on very different subjects coherent, and it covers a number of interesting topics, including ones that I'd thought I'd have references to in other books but my knowledge of them must have come from papers.

The feature that Padmanabhan uses to categorize physics is similar to the one used by Griffiths in his Introduction to Electrodynamics:  a cube has eight vertices, and each of these vertices is a type of physical theory.  Each of the eight vertices has one of three Boolean values: gravitation, relativity, and quantum.  If the effect is accounted for, the value is on, if not, it's off.  The vertex where they are all off is Newtonian Mechanics (without gravity) and the far vertex is a Theory of Everything (or somesuch, he calls it something else). Turning gravity off is like setting the gravitational constant equal to zero, turning relativity off is like setting the speed of light to infinity, and turning quantum mechanics off is like setting Planck's constant equal to zero.

So, I could make a list of the different vertices:

(0,0,0) Newtonian Mechanics
(1,0,0) Newtonian Gravity
(0,1,0) Special Relativity
(0,0,1) Qunatum Mechanics
(1,1,0) General Relativity
(1,0,1) Gravitational Quantum Mechanics
(0,1,1) Quantum Field Theory
(1,1,1) Theory of Everything


The most interesting statement in the introductory chapter is that GQM is a theoretically unpopular zone, and very poorly developed, despite being one step away from the TOE and possibly holding clues to it (or, I assume it would, even if it's boring to the high-powered mathematical physics wizards).

Sleeping Beauties touches on most of these topics  (perhaps being short on the TOEs), showing up with some interesting takes on different phenomena, many of which I knew (rainbows and mirages, Thomas precession), many of which I didn't realize (the connection between Thomas precession and the Foucault pendulum), and many I had never touched on in my life (gravitational bending of the electric field).  The twenty-six chapters have at least twenty-five topics and are each worthy of attention, especially those where he's making connections either between the vertices of his scheme or between subjects that seem disparate (but really aren't).

But, it's not for people who avoid mathematics.

The very first chapter of content, "The Emergence of Classical Physics," is an example of this.  It tries to show how minimization principles in classical physics are explained by the limit of quantum mechanics.  This uses Wigner functions to show that the action is the phase of the quantum mechanical wave function, and so classical particles follow trajectories that follow paths where the wave function's phases isn't cancelled out by neighboring paths (just like in Feynman's QED: The Strange Theory of Light and Matter). Obviously, if you're not up on the quantum mechanics or the calculus of variations, you might not see why this is so awesome.  Speaking of QED,  chapter 17, "If Quantum Mechanics is the Paraxial Optics, Then..." does something similar by applying the results of chapter 16's investigation of the transformation from wave optics to ray optics to discuss how quantum particles move in quantum field theory.  Including why you must include trajectories for your particles that go backward in time.  It's all the same process.

That is probably the most obvious recurring theme in the book: by understanding how a theory can be derived from a more fundamental theory, mysterious things like minimization principles can be made to make sense.  That is, if you know how to turn off one of the switches, then you can better understand why things occur in the lower level theories -- and where those theories might lead you astray.

I therefore heartily recommend this book, and hope you both read and like it, but it does require an investment in mathematics to realize its physical returns.

["I therefore heartily recommend..." Looks like I've been writing too many recommendation letters, again.]

Monday, July 17, 2017

Screwing Up on Quantum Computing

Well, I screwed that up.

In the next Physics Frontiers podcast, recorded about six months ago (approx. 1/27), Randy and I talked about photonics and quantum computing. The papers we read were very interesting and we had a good time talking about them, although we got a little bit confused (as my memory serves me, I believe that happens in the next episode as well). During the episode I went off on three digressions that I probably shouldn't have: science fiction, quantum computing, and economics and politics. I cut out all of the economics and politics bits (again). I left in the quantum computing, and that's what this is about.

I first read a journal article and had discussions with reasonably technical people about quantum computing in the 90's, before 1998 and after 1995 based on the people I was talking to. The whole idea that someone could construct an algorithm for a computer of a type that didn't exist yet really intrigued me at that point, and since one of my majors was "theoretical mathematics" (I think that's what the track name for all that abstract algebra and number theory stuff was at that school), figuring things about like how to factor large numbers very quickly was interesting to me. The applications to cryptography, not so much. But, the state of the art at that time was a sheet of paper and a proof, so my marginal interest in the subject didn't really go anywhere.

I've occasionally looked in on quantum computing from time to time since then. I've read a couple of technical books on the subject, for example, and I'll read articles about it when I actually have time to crack open one or two the Science and Nature magazines that clutter up my office. So, although I'm certainly not an expert (I'm not an expert about anything we talk about on Physics Frontiers), I certainly didn't think anything really astounding had happened. I was pretty sure a couple of people had put together a small number of qubits in one or the other of the ultracold settings.

I had no idea that Google and IBM had them up and running.

That's what my phone told me today, about lunchtime.

I think that I had heard that quantum computer simulators existed, and so some of my comments were just thoughtless ("people who design algorithms for computers that don't exist and then mathematically prove they'll run correctly" or something like that), but the fact that IBM has a 17 bit quantum processor up and running and they're giving people beta access with an SDK for Python makes some of my comments laughable. And of course, Google's up to the same thing.

It's amazing how fast these things are developing.

Oh well. The episode's already short, so I won't edit anything more out. You can just hear me babble from the past about things in the future.

Now, how can I get into that IBM beta?