Super excited that @KathiFisler just won the ICER Lasting Impact Award for her 2014 (solo!) paper, "The recurring rainfall problem", which reopened and finally made progress on a 30 year old open education question!
https://t.co/Z4v0WXkI0p
Ever wanted to try building a JIT from scratch? Undergraduates in our compilers course did last fall! We started from Ghuloum's “Incremental Approach”, and added a REPL. It turns out this forces a ton of useful decisions needed for JITs.
1/
@ShriramKMurthi@EneaszWrites@wfenza It is very difficult to accidentally measure the course - I agree. I try to make these types of questions a measure of knowledge transfer to the best of my ability!
@yawaramin@EneaszWrites@ShriramKMurthi@wfenza As a follow-up, students do this type of thing all the time in my class; they just do it more often on problem sets than exams (that is, they take the concepts from class (such as interfaces) and apply them to psets in ways I didn’t explicitly demo in lecture.
@yawaramin@EneaszWrites@ShriramKMurthi@wfenza I don’t think you understood (and perhaps that’s my fault). I might, e.g., ask students to design a “cons list” data structure on their final exam, despite never talking about ��cons” SPECIFICALLY in lecture (but we *did* talk about everything that allows you to make one!)
@EneaszWrites@ShriramKMurthi@wfenza This is what I do in my CS2 class. The first question on every final exam is about a data structure/problem that they haven't explicitly seen in class. I want to see how quickly they can pick up something new. (It uses concepts from class, of course, but isn't verbatim taught).