The analogy is to teaching algebra through endless, dry repetition of problem solving. Some of you won't get what the big deal is; you're the ones who are good at 'crossword puzzles'.
The analogy is to teaching algebra through endless, dry repetition of problem solving. Some of you won't get what the big deal is; you're the ones who are good at 'crossword puzzles'.
He taught math by talking about the history of the problems we were trying to solve, and how people went about solving them. It was the concepts that mattered; the equations were secondary. He would put a problem on the board, and ask the class to solve it. He would then give us pieces of the puzzle, until we had enough that we could figure it out ourselves. I found the class participation to be key. With most math classes in college, the instructor works through 3 or 4 problems for a given topic on the board, and the students spend the entire class furiously copying down what he writes. It's pointless. I get that it's driven by the need to cover a lot of material, but I'd rather the class work through two examples then the instructor work through 4. Also, regular quizzes were a key part of Irani's class that made retention much, much easier. Instead of the typical weekly homework, two midterms and a final, I would much prefer weekly quizzes, 3 (or 4) midterms, and a final worth a much smaller chunk of the grade.
Last semester, I took linear algebra. We had weekly assignments that I did with a tutor (bad decision), two midterms, and a final worth 45%. I went into the final with ~40% in the class, crammed for it, and ended up with a mid-60 in the course. It was pointless, and I don't feel like I learned much. I would much prefer the system Irani set up, but he's not a university professor, and he doesn't have a PhD. He taught himself calculus with an inch-thick pocket book he picked up from a second hand store.