2004-08-02 OSRM: Linux infringes 283 Patents, We Offer Insurance
--> [ deutsch | News | Insurance | Munich 04-07-30 ]
Media Echo
Sydney Morning Herald: 283 patents potentially infringed by Linux
Informationweek: Study Identifies Potential Patent Threats Against Linux Users
EWeek: Open-Source Insurance Provider Finds Patent Risks in Linux
Outline
An insurance venture called "Open Source Risk Managment" (OSRM) estimates that Linux-based systems fall afoul of 283 not-yet-litigated patents, among them 27 from Microsoft, and offer an insurance for covering the litigation risk.
Dan Ravicher claims to have conducted the investigation but does not disclose the patents and how they are infringed because by doing so he would expose people to a heightened risk of knowing and wilful infringment.
Apart from the confidentiality, there are however some more reasons to doubt about the figures:
- it seems impossible for 1 person or a small group to investigate the mass of relevant patents and relevant software code and single out any precise number of patents.
- the campaign seems to serve a clear PR purpose, e.g. that of telling people that
- there is a big risk
- the risk is managable, we can handle it
- the risk is due to patents which can be challenged in court with prior art, corroborated patents are not infringed (lawyer's dream: litigation services are the key to solving all problems)
As the Forbes article points out, Ravicher has made some questionable public statements before.
Yet there is some reason to believe that the OSRM people are seriously committed to building a systematic defense strategy for the free software community in the US.
Perhaps the figure of 283 is simply a conservative estimate of the number of patents of which an ordinary GNU/Linux system (and also an MS Windows system) probably falls afoul.
In a recent rough report for the Munich Municipal administration, FFII came up with a figure of 50 probably infringed patents granted by the EPO, but this figure was based on very insufficient browsing of both the patents and the concerned software. Especially in kernel-related fields such as memory managment the situation looks completely unoverseeable and we would especially like to hear how OSRM arrives at their estimate that the Linux kernel is not falling afoul of any one of hundreds or thousands of patents in this field or that all these patents problems can be solved by an insurance.
An insurance will not cover operational losses or the losses associated with not being able to use a patented technique. There is also a risk that an initiative such as OSRM, just like the meanwhile defunct priorart.org initiative which was associated with some of the same "community leaders", will just enlist volunteer labor from the free software community in order to help the patent office clean up its mess of poorly examined patents and thereby strengthen the credibility of the more dangerous ones.
See our criticism of priorart.org.
