A lot of universities that came early to computing knew a lot about administration because they couldn't buy in management of all the equipment. This is why most universities I know about have two essentially disconnected networks: the main one with all the university web apps, campus workstations, and student emails, and the one run by the computer science department, which does all the same things but only for those students and occasionally other important systems. For example, I've seen where the CS admins maintain the HPC systems, even though it's mostly the other sciences using it. There are a few exceptions where, when the universities needed administration, they expanded what they already had, but most appear to have taken a more basic approach.
Biting the hand that feeds IT
