Yesternight, I went memory-road-tripping through my blog archives. One of the things I realized, apart from how amazingly enthusiastic I was for the blogging form back in 2007, was how much I reviled my sophomore-year university physics classes. At the time, they were unpleasant; in retrospect, they were deleterious.

The worst was the relativity class, which had most of the interesting stuff dug out, and hot air pumped in to fill it out to a semester. Also a waste of time was the first semester of quantum mechanics—we had three, and the latter two were great. That is actually a topic where a three-fold division could make sense. One could have, for example, a semester on the basic formalism, a semester on symmetries and exactly solvable systems and then a term covering approximation methods. But that’s not what we did. Instead, the first term was “intuition building,” which translates to “let’s plod through differential equations over and over and over again, because learning any oh-so-much-*harder* math would be too much for your young brains.” Mixed in with that was a rather bog-standard “physicist’s history of physics,” which was as usual too inevitably misleading to be worth bothering with. The time period in question is fascinating to me, as is our professional mythologizing about it. The textbook cardboard we got couldn’t do the subject justice.

And another thing: ditch the requirement of going to the Math Dept for a differential equations class. That class sucked. Again, pretty much inevitably. I’d say it was doomed from the get-go, *i.e.,* failure was ensured by the choice of topics and level of coverage. (All I remember from that semester was the professor’s claim that the Laplace transform takes you into a world where everything is yellow. I still don’t know what that means. Everything I was *supposed* to learn in that class, I picked up elsewhere and elsewhen.) A much better alternative would be an in-house class on mathematical methods. It’d slot neatly into the same place in the curriculum, even. And there are good books to teach out of! Also, for the love of Gauss, don’t rely on Matlab. Real programming languages exist and are at least as easy to introduce. If you’re going to be gung-ho about Technologically Enhancing the Active Learning in your classrooms, you might as well teach some skills which physicists could actually use.

(Said in a tone of voice suggesting that one might eventually have dinosaurs on one’s dinosaur tour.)