It's funny, but quite a while ago when I was briefly considering going into teaching computer science, I thought it might be interesting to give homework and/or exam problems that have students program under various unusual restrictions.
-
It's funny, but quite a while ago when I was briefly considering going into teaching computer science, I thought it might be interesting to give homework and/or exam problems that have students program under various unusual restrictions. For example, accomplish some task without a `for` loop, or something. Basically try to encourage getting to know *all* of the tools a language offers.
-
Xandra Granade 🏳️⚧️replied to Xandra Granade 🏳️⚧️ last edited by
I feel like that idea looks a lot better now that AI coding "assistants" have pushed so many homogenized ideas of what programming can and should look like. Encouraging students to go outside of that kind of local optima, even to a worse solution, seems like it can only have the effect of pushing back on that homogenization.
-
Xandra Granade 🏳️⚧️replied to Xandra Granade 🏳️⚧️ last edited by
Tangentially, another idea I was always curious about was to incorporate some degree of self-evaluation into homework and exams. Basically, give students a number of "confidence points" that they can assign to increase the weight of a given problem. That way, they can assign more to problems that they're confident in the solution to, also avoiding penalizing students for not knowing some particular thing by rote.