I've had the privilege of hearing Eric Roberts (the professor interviewed in the article) at various conferences over the years, and have always found him to be impressive. He's always thoughtful, and entertaining.
One idea mentioned in the article above is fascinating ... he tells his computing students that for every incident of cheating that happens in class (that he catches), he raises the weight of the final exam for everyone. The final exam starts at 15%, but increases by 5% for each incident. Thus, since most students despise exams, there's community pressure to stop cheating.
I'm not sure what I think about that. On the one hand, it seems to reinforce the "us vs. them" mentality of instructors vs. students. Of course, it's easy for me to criticize someone when my classes are around one-fifth of the size of his. On the other hand, it also does attack directly one of the bigger issues: cheating happens often because other students enable the cheater. Or, in the case of some schools like mine, the student body creates a culture in which cheating is often glorified, if not downright encouraged.
Interesting. I just looked up Roberts' course syllabus. There's another interesting twist. Cheating cases that are reported by students, rather than staff, don't incur the 5% class-wide penalty. (An aside: Stanford has a Honor Code, in which students are already obligated to report violations by other students of the Code.) So ... there's no penalty for people doing the right thing, either. Fascinating.
Reactions, anyone? (Particularly from students?)