I've taught the modern physics course for almost three years now, and we've updated the syllabus so that we include topics from elementary particle physics in the relativity section of the modern physics course. I'll tinker with how we teach it so that we emphasize conservation of energy momentum. So that part of the syllabus is something I'm looking forward to.
What really excites me though, is the opportunity to teach the complex (variables) methods course. My students are beginning graduate students with back subjects. They come from universities with weak undergraduate physics programs, and were required by the department to take complex methods. I'm planning to push them to the limit, and I spent two days preparing the syllabus.
Note that the syllabus is a detailed, day-by-day schedule, with listed course objectives and suggested problems. I'll be using Arfken as the main text, because I like the problem collection, and because I'd like them to learn the art of reading books like Arfken's.
It's an extensive set of topics-- I looked at an old course outline (used by Professor Y), and it goes from Chapter 6 of Arfken to Chapter 14. (With the Chapter on Sturm-Liouville theory removed). I've removed as much as I could, without violating the course description; even then, what is left is substantial. What I did was to look at what I've already solved, and then select a subset ( a third?). Even then, the grand total is 149 problems on the suggested problems list.
What I plan to do though, is to assign a subset from the subset of 149. The rate of submission will be about five a week, and I'll solve some of the others in class and during out-of-class consultations. (They should consult as a group.) Since I only have five students, I plan to work with them so that at the end of the course, they will have a good grasp of methods.
I'm hoping that after the effort I put into this class, these students will make the next round of graduate courses more difficult to renormalize.