There's plenty of open source projects that derived from an university thesis, project, collaboration. Redis and Ceph quickly come to mind -- and somewhat the o

There's plenty of open source projects that derived from an university thesis, project, collaboration. Redis and Ceph quickly come to mind -- and somewhat the original Linux kernel, too. But they where very good ideas of a sole student that gained traction, not really "encouraged" by the university itself. After 10 years as an educator myself, I think the main problem is laziness in evaluation methods. It'd be hard, time consuming but totally doable to assess the competency of people contributing to one or various open source projects, even if each one works on a different problem. But if I make everyone build a small operating system or a database, I can evaluate, assess and score all of them with the same method, at the same time. I've seen that laziness taken to the extreme by old tired professors, and it ain't pretty.