But lets be clear, this is the case everywhere. Geeks have the same clan politics. The difference is rarely the politics, but where you sit in the pecking order, that changes your perspective.
I actually think the problem with academia in CS today is it's too applied. They should NOT be focused on today's problems. The Cloud is not something one should do research on in academia. Industry will tackle the cloud from a million different angles. But what's after the cloud? What after touch and Kinect? Precise mind control and HUD in glasses/contacts really should be the mainstream in academic research today, but it's actually a fringe.
I think academia should be way out there more than it is. The geeks will protest that they do nothing of relevance, but that's how it should be. :-)