" Suggestions to change completely the assignments from year to year are appealing on the first sight but they cause others types of problems: It is very diffic

" Suggestions to change completely the assignments from year to year are appealing on the first sight but they cause others types of problems: It is very difficult to know in advance if an assignment is going to be too easy, too hard, or too ambiguous. Even small-scale testing with TA's and other faculty does not help. You need to "test" the new assignment by giving it to students. If it is a good one, you want to keep it. " I've often heard that from professors in the US, but in my university in France we had a rule forbidding professors from reusing exams and things were fine... Now one thing though, exams were made so that the average is around 10 or 12 out of 20... In cases where the gaussian curve was centered too low, the teachers would decide to grade out of 22 or 24 points... I have to say though that the idea of using public project, competitions and peer reviewed projects is a very good idea and worked well when I was a student...