“Innovation and its Discontents” by Adam Jaffe and Josh Lerner

There is much ammunition in this book for what I already suspected. There is a good deal of theory and practice of patents. Many famous patent disputes are explained more cogently than the contemporaneous press reports. The book suggests that patents provide a viable business model for drug research. Clearly sometimes patents do encourage innovation. The hardest part is generally not the invention but the commercialization of the invention.

Some pro-patent points he has missed so far:

Regarding the Rambus vs. Infineon case I thought that there were severe penalties in claiming that the inventor invented something that he did not. I thought the penalty was much more than merely voiding the patent.
I have no nexus on patents so I note here an interesting paper by James Bessen and Eric Maskin: Sequential Innovation, Patents, and Imitation
On that paper I note: Regarding licensing patents for further innovation. Is the decision to invest in further innovation made before or after the license agreement? There are problems either way. A low transaction cost license mechanism would help, but I don't know any. Perhaps most of the cost of innovation is not in having the innovative idea, but in proving its efficacy. This does not much change the dilemma however.