we waste billions on "defense".

While I agree that the USA does has a large defense budget, waste can be an amibiguous term. The new fastest supercomputer at Livermore is designed to model those weapons you speak of. I'm not sure that driving that technology is a waste.

http://www.top500.org/lists/2012/06

trying to show the kind of immediate causal connection (we funded X million dollars that created Y new industries) that are required for these kinds of cost benefit analyses are difficult to come up with.

Maybe, but i also think that you can't just say funding "big science" creates new industries, there should be more quantative statements. One of the most repeated posts on this entire website must be "show me the evidence," I think that is all that governments are saying.

Just because you think it may be "hard" or "difficult" to show isn't going to be a good enough reason not to do it, and, clearly, teams of scientists are (starting) to do it.

How about this question, which is one governments must answer. What should the science budget be? Given that scientists could spend many times more money than any governemnt has, how do you decide how much to give them?

Or to paraphrase a question in an early post:

The public are mainly concerned with: health, cheap clean energy, and security from terrorism. Why shouldn't tax money go exclusively to those areas?