Actually, I would argue even your approach does not really represent the real utility of a university. My list includes 1. failed to start a startup about hard

Actually, I would argue even your approach does not really represent the real utility of a university. My list includes 1. failed to start a startup about hardware tutorials in local languages 2. failed at productionizing a propeller display, they are pretty popular popular now. 3. failed at productionizing a gas sensors with high resolution The list of failures go on and on actually. But the real utility I had was the nice cozy university bubble where failure of projects isn't really ... accounted for.