Clearly "let's cooperate and put trolls out of business" isn't happening right now, and other than the possibility that no one has thought of it before now (doubtful) there are two obvious reasons:
- For a given case where a troll sues multiple parties, each party is relying on the other defendants to cooperate, but the lowest risk alternative is to settle early and be done with it. The patent troll knows this, and will try to cut down the number of cooperators until all are forced to settle.
- Since most cases don't offer much resistance and often settle, patent trolls are able to fund an expansion of their operations, and they simultaneously gain credibility ("you'd better settle, Microsoft already did").
So the historical practice of settling with trolls works against companies who would think not to do so, and any group of companies that tries to band together in defense is in a situation where defection ruins the defense, and the incentive to defect starts high and increases with each additional defection.
As Mr. Crutchfield says, it's a shakedown that exploits a peculiarity of the patent system (namely, the combination of the ability to sell patents with the ability for NPEs to bring infringement suits). This is an emergent property of the legal system. The right approach is to change the system.
I doubt I'll have to convince many here that nothing of value would be lost if NPEs were no longer allowed to bring infringement suits for dormant patents.
One solution to the prisoner's dilemma is for everyone to get together and give up the ability to defect, say by incorporating it into corporate bylaws. Another is to drastically change the cost structure, such as by insurance or a mutual contract that holds anyone who settles liable for a large amount. If not done carefully, either of these could tie a party's hands against legitimate suits, but they are alternatives (after a fashion).
Game theory has analyzed the prisoner's dilemma to death. If there's any hope for a grassroots solution, it's there.
Perhaps a group of lawyers and actuaries should get some money together and create an anti-patent-troll company, viz., a company that sells patent troll insurance. They get a whole bunch of companies as clients, and contractually obligate them to litigate (via the insurance company) trollish patent suits. Centralize and incentivize.
The OP goes on to say that the victims of patent trolls band together to finance the review to make it cheaper than settling.
Yes, but the OP ignores the probability of failure. I'm claiming that even a small probability of failure would be sufficient to deter most businesses from pursuing the OP's advice because the costs of a defeat are so much greater than the costs of a settlement.
> The OP goes on to say that the victims of patent trolls band together to finance the review to make it cheaper than settling.
I'm not disputing that. The problem I'm pointing out does not have to do with financing the review, but with the fines that are due if you lose the case. Even if you band together to finance the review, if you lose, you have to pay the fines individually.
The USPTO is broken.
[1]http://fosspatents.blogspot.com/2011/05/worse-than-lodsys-ma...
[2]http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sec...