Thanks Thomas for a sharp description of the academic world from an outsider (among others).
As you and many others have noted, the incredible amount of noise and hair-splitting is simply due to the "publish or perish" pressure, and also to the large number of researchers compared to the limited number of innovations of general interest.
I am nevertheless convinced that a better and stronger collaboration between industry and academia would be extremely beneficial to both the industry and the public (= Academia's employer in theory). Researchers have analysis and prototyping skills often badly needed in the industry. Researchers are often ignoring interesting industry problems (especially interesting since they are real).
I think the only way to achieve such a collaboration would be to have a significant number of individuals "bridging" these two very different universes by moving back and forth between the two. This number is not zero but it will unfortunately never be significant since building a career in the academic world is very painful and slow. So you cannot be late to enter the academic game, and besides some exceptional individuals most researchers cannot afford to leave it for a while. In brief I think the too inflexible academic universe is the main one to blame here.