I have to wonder at this. How did the actuaries at Lloyd's members actually figure out the risks here? What dataset did they use? Do pharmaceutical companies sue hospitals over patent infringements rather than each other? Did they base their calculations on the lonely two suits brought about by SCO Group against Daimler-Chrysler and Autozone?
And if the patent risks in the Linux base claimed by OSRM's Daniel Egger are real, why aren't those patent holders performing their fiduciary responsibilities and at the very least suing the Linux distribution companies like Red Hat and Novell, regardless of whether the patent holders are traditional vendors or merely these dreaded patent holding companies we keep hearing about? Or why aren't they chasing the well known large customers? You think it would already be a veritable gold mine in a multi-billion dollar marketplace, and if they don't act soon, surely they risk losing control of those same patents. This assumes of course that these software patents are strong defensible patents.
Wikipedia has a fascinating history of insurance. The bit on patents for insurance is also fascinating as well as the controversy over whether its good to issue such business method patents in the U.S. for new insurance types. (I wonder if OSRM applied for this patent compounding their ignominious behaviour.)
Insurance as a business would appear to need to have three factors:
- a dire consequence (pirates, fire, earthquake, lawyers)
- a probability greater than zero but less than 1.0 that the consequence will occur
- a banker willing to collect dues from all to cover the costs of the few (investing the proceeds in the mean time of course)
We haven't seen the data on the dire consequences so how can they calculate the risk?
Of course there's lots of data available on vendors suing customers over user and CPU counting issues for proprietary software. Some vendors even unleash the dreaded BSA on customers for costly audits. Here's a situation where free and open source software shines, because there is no license counting. Maybe the Microsoft executive had it backwards all those years ago when he claimed open source was a cancer. You see, if you have cancer, insurance companies won't sell you life insurance. Maybe the risk of proprietary software litigation and audits over user counting is so high you can't buy insurance against that eventuality.
Either way, until real risk demonstrates itself in the market place, I can't imagine buying open source insurance from OSRM. At the very least, there have got to be better business risk mitigation strategies in which I can invest money. And on the positive side, there have got to be better business opportunities around open source on which I can spend money.