Without agreeing with him on this specific point, I do have to note that just because he sees the amount of litigation as a good thing does not mean he thinks that litigation itself is beneficial to the economy.
As a general rule, litigation is a sign of activity. The more economic activity you have, the more interactions and transactions you have, the more conflicts that arise, and the more litigation you have to resolve those conflicts. The litigation itself is of course overhead, but the fact that there are many conflicts can be a sign of an industry in which a lot of things are happening.
The litigation itself is of course overhead, but the fact that there are many conflicts can be a sign of an industry in which a lot of things are happening.
As I'm sure you are aware, not everyone thrives in conflict. Some would much rather design a system to prevent it. I can't help but feel that this is similar to claiming that there must be a lot of fun happening on the school playground, because a lot of bullies are beating up the kids who just want to get along (and, occasionally, other bullies).
Conflict is an unavoidable consequence of human interaction. There are various things you can do to make conflict resolution as quick and efficient as possible, but ultimately if you have X interactions between people, some fairly stable percentage of those interactions will result in conflict that requires litigation to mediate.
A more apropos playground analogy might be skinned knees. They are of course not a good thing in and of themselves, but if you have a bunch of kids running around and running into each other, you're going to get skinned knees. The more kids you have running around and the more active they are, the more trips to the nurse's office there are going to be. You can make the playground safer, tell kids to be more careful, etc, but at the end of the day kids getting playground injuries is correlated with kids playing.
Conflict may be unavoidable, but that doesn't mean it requires public resources to resolve, nor that it needs resolution at all. Even without patent disputes, there would be a "conflict" between android and iOS -- that doesn't mean we would need the courts to step in and figure out which os is better -- the conflict is kind of the point.
We could clearly today have either more or less protections: you can pretty easily imagine a system with less patents or one with more and stricter forms of legal protection. A system with more forms of protection would probably see more legal battles. A system with less protection would see more market battles with focus on winning over consumers vs juries. The system we have today isn't some divinely revealed sweet spot, so the attitude that "conflicts are inevitable and this is how we solve them, thus there's nothing we can do about it" is disingenuous.
I don't think the skinned knee analogy is quite apropos. Skinned knees are caused by gravity. Patent litigation is caused by the deliberate action of patent litigators.
I also disagree that conflict in the form of litigation is an unavoidable consequence of technology innovation. To a small developer, there's little effective difference between patent litigation and a bullet in the back of the head. They're both a death sentence, except the bullet would be over quicker. Instead of streamlining the process of executing small developers, courts and legislators should remove the actual threat.
I should note that, though I disagree with many of your points in many of your comments, I do value the alternate perspective you bring to patent-related discussions on HN.
The cost of compliance should be considered in determining whether a rule of law is ultimately to the net benefit of the economy, but if the net benefit is positive, the fact that compliance might put small players out of business shouldn't be a determinative objection. As an economic matter of fact, businesses that can't comply with generally beneficial regulation should go out of business.
So the debate should be about whether patent rights are in fact legitimate and beneficial regulations, taking into the cost of compliance but also taking into account ways in which those costs may be minimized through the structure of the law.
I think there are ways to structure the patent law, with bright-line defenses, so as to make defense of a suit extremely quick and cheap even by pro se litigants. One example would be requiring some sort of evidence of intentional copying as is required for copyright infringement.
A more apropos playground analogy might be skinned knees.
With the penalty that both the nurse and the injured person receives $1 from the person who caused the skinned knee. You can make the playground safer and tell kids to be more careful, but you'll still see Bruno falling on purpose and blaming Mary to the nurse.
You also have people who get shoved to the ground and just take it, letting the bullies get away.
People assume that there are more of the fakers than the pushovers, but most of the studies I've seen of the legal system show that there are way more people who fail to enforce their rights than there are people trying to game the system. E.g. only a small fraction of medical malpractice claims are clearly not meritorious. On the other hand, only a small fraction of those who have a legitimate malpractice actually sue. The same is true for environmental harms, etc.
As a general rule, litigation is a sign of activity. The more economic activity you have, the more interactions and transactions you have, the more conflicts that arise, and the more litigation you have to resolve those conflicts. The litigation itself is of course overhead, but the fact that there are many conflicts can be a sign of an industry in which a lot of things are happening.