In the past grad students gave us : -> Relational databases -> Polynomial time deterministic algorithms for testing primality -> Shannon's masters thesis etc.

In the past grad students gave us : -> Relational databases -> Polynomial time deterministic algorithms for testing primality -> Shannon's masters thesis etc. etc. I still don't believe that Google itself could have produced so excellent a product outside an environment that a top graduate school offers. The concept of advancing the boundaries of the field is noble - the idea of making $$ out of it should hardly be the motivation IMO.