OK. This is a long slog, but I skimmed and got the gist.
There must be a technical solution to the cheating problem.
Many years ago I taught an intro to programming course. For the grading I built a randomized online test - students would have one hour to answer a number of programming questions using a terminal in the computer lab, and would get their grade as the test progressed with the final grade after the last question. It worked both at freeing up my time for grading, and eliminating cheating.
Automated randomized testing in all types of classes (not just an intro class where this task is much simpler) would go a long way towards solving the cheating problem.
Heh, I wonder if the quizzes just shuffled the questions between students, how many would get it wrong because their option A) was other people's option B)
No. It wasn't like that. I built software that would generate unique completely random test questions. The software would eval() the students response and would give 100% if correct, 0 if not.
There must be a technical solution to the cheating problem.
Many years ago I taught an intro to programming course. For the grading I built a randomized online test - students would have one hour to answer a number of programming questions using a terminal in the computer lab, and would get their grade as the test progressed with the final grade after the last question. It worked both at freeing up my time for grading, and eliminating cheating.
Automated randomized testing in all types of classes (not just an intro class where this task is much simpler) would go a long way towards solving the cheating problem.
There, a problem for all you to solve.