So, Andreessen is talking about "ideas" -- hmm .... His ideas seem to be (A) some large changes in the economy and society from (B) some exploitations of large

So, Andreessen is talking about "ideas" -- hmm .... His ideas seem to be (A) some large changes in the economy and society from (B) some exploitations of largely existing computer technology to meet some want/need previously unnoticed or infeasible to meet. But, even for just (A) and (B), there is potentially MUCH more potential in ideas that Andreessen seems to ignore. An example was Xerox: Copying paper documents was important. The main means was carbon paper. Xerox did quite a lot of engineering research based on some early research, IIRC, at Battelle. The result was one of the biggest business success stories of all time. Andreessen doesn't discuss research ideas -- how to have them, pursue them, apply them, evaluate them, etc.