It is true that the innovations of silicon valley are rooted in government (mostly military) sponsored research going back to at least WW2, but the idea that we

It is true that the innovations of silicon valley are rooted in government (mostly military) sponsored research going back to at least WW2, but the idea that we should therefore credit the government as some kind of "innovation driver" is totally wrong. Innovation comes from setting daring goals and tackling hard problems (the "Grand Challenge" theory of innovation). The government is incapable of directly innovating because it usually has no grand challenge, no overarching unity of vision. This is by design: the government exists to "make our collective lives better" but no one agrees on what this means so the government cannot do very much. On rare occasions, our collective opinion coagulates around a consensus that enables a kind of societal grand challenge from which remarkable innovations spring forth, for example WW2 and then the Space Race. Certainly the government (mostly the military) has an important role to play in such times. But that situation is the exception, not the rule, and we don't have that right now. At present, there are only three things that governments can do to spur innovation: increase military spending, fund academic research and remove legal and financial barriers to entry for new ventures.