Samba developer Andrew Tridgell reckons the open source community is poor at fighting patent attacks, but says things could improve with a change in strategy.
Samba developer and Australian open source legend Andrew Tridgell has had plenty of experience dealing with interesting legal issues while trying to get Linux servers communicating with Windows, but he now sees the exploitation of patents as the biggest threat to the open source movement.
"I believe that patents are unfortunately going to be more and more of a problem for the free software community," Tridgell said in a presentation at Linux.conf.au 2010 in Wellington. "I am concerned that patent attacks on the free software community are going to become more common in the future."
Tridgell's comments echo remarks by his Samba co-developer Jeremy Allison this week suggesting that Microsoft would actively use its patent portfolio to fight off netbook and mobile phone competitors.
To counter that threat, Tridgell offered some guidelines to how to construct responses to claims of patent violation (though he was careful to point out he was not a lawyer). He argued that attempts to displace patent claims by trying to invalidate patents by demonstrating 'prior art' were not well thought out. "The prior art defence is really hard to pull off. Prior art is not a panacea. It is very hard to kill all claims. You've got to knock them off completely. It's a massive amount of work
While many online forums spend time trying to find prior art examples when an open source project is threatened by a patent claim, people often don't look any further than the abstract, he noted. "The abstract is often very different to the actual claims. This is where Slashdot tends to get stuck." Righteous anger has its limitations as a practical strategy: "It's your job not to just blow your stack at the whole patent system."
Tridgell also suggested that the commonly circulated idea that people should resist reading patent documents so as to avoid were misguided, especially for non-commercial software, since any successful patent claim would likely bankrupt the project. "For most free software projects, certainly for the vast majority of projects in your average distribution, one death is enough."
The best defence is to identify the specific claims inherent in a patent, and demonstrate that a given piece of software doesn't actually perform the tasks specified in those claims. Proving that software doesn't infringe is a lower barrier than trying to prove the patent invalid, Tridgell said. "For a non-infringement defence, you only have to care about the independent claims. For prior art or invalidity, you must demolish every claim in the patent completely."
Co-ordinated efforts to help defend software will be the best long-term strategy, Tridgell said. We need to be the toughest, meanest target for patents. We have a technical community that's really good at the sort of logic and knowledge to fight against patents. If we can find a way to co-ordinate within that community, we can do something really interesting."