Back to article
Bruce Perens: A Big Change for Open SourceBy Bruce Perens
October 2, 2008
About the Author Bruce Perens is the creator of the Open Source Definition, the manifesto of Open Source and the criterion for Open Source software licensing. Perens represented Open Source at the United Nations World Summit on the Information Society, at the request of the United Nations Development Program.
An appeals court has erased most of the doubt around Open Source licensing, permanently, in a decision that was extremely favorable toward projects like GNU, Creative Commons, Wikipedia, and Linux. The man who prompted that decision could be described as the worst enemy a Free Software project could have. This is the story of how our community was able to benefit from that enemy.
For a decade there'd been questions: Are Open Source licenses enforceable at all? Are their terms, calling for a patent detente or disclosure of source code, legal?
Are they contracts, which require agreement by all parties to be valid, or licenses, which are binding even if you don't agree to then? What legal penalties can a Free Software developer employ: only token damages, or much more?
The court's ruling makes the answers to these clear. Did such weighty questions come up in cases involving IBM, Sun, HP, or Red Hat? No, this is the quirky world of Free Software: it was a court case about model trains.
The reason for so many questions about Open Source licenses was simple: there weren't any court cases about them, so nobody could say with any confidence how a judge would rule. The few cases that did start up never reached a verdict, because the parties settled their dispute and kept the details of their agreement secret.
The one high-publicity case we've ever had, SCO's self-destructive pursuit of Linux users and IBM, established the originality of Linux, but didn't concern Free Software licensing. So, we had waited 10 years for the magic lawsuit that would establish the legal solidity of Open Source licensing, and hadn't gotten it.
Enter the two opponents: on the left, Bob Jacobsen: by day on the staff of a government nuclear research lab, by night a model train hobbyist. Jacobsen built what might be the ultimate nerd product: "Java Model Railroad Interface" or "JMRI," computer software for controlling model trains. Jacobsen gave JMRI to the world as Free Software, never expecting to make a cent from the project but only asking to share the software he created with other train hobbyists.
On the right: Matthew Katzer, owner of a company that sells model train software, who has filed patents that essentially cover all use of computers to control model trains. Katzer has brought and later withdrawn a few lawsuits against other model train hobbyists, who in turn allege that the technology Katzer claims to have invented recently is not his, and has actually existed since the 1960's.
Jacobsen alleges that in 2002, Katzer filed for patents on features that were already available in Jacobsen's JMRI, and that Katzer didn't tell the patent office about the "prior art," evidence that other people made the invention before Katzer did. Oddities of U.S. patent law allow applicants to claim that they created an invention long before they file the patent application, and Katzer's 2002 application claims to be the continuation of a 1998 patent application - thus side-stepping the pre-existence of Jacobsen's work. Unfortunately, the U.S. Patent Office does little to verify that people actually made their invention when they say they did, and doesn't do a thorough check for "prior art" either.
Once Katzer's patent was granted, he started sending bills to Jacobsen, asking for $200,000 and threatening to sue. Jacobsen could have waited for the inevitable lawsuit, but felt that turning the tables would work better. He brought suit against Katzer, asking the court to decide that Katzer's patent was not valid.
After he filed the suit, Jacobsen found that Katzer's commercial product copies some of Jacobsen's JMRI (which Katzer has admitted in court, according to the finding of the appeals court) and that's where this story gets even more complicated. By putting Jacobsen's JMRI software in his product, Katzer bound himself to the terms in JMRI's Open Source license, which prohibit Katzer from asserting his patents against the developers of JMRI - Jacobsen and his friends.
This, folks, is one big reason why Free Software developers use licenses. Open Source projects give their work away for free, and their developers can't spend millions in court when someone attacks them with a questionable patent. They especially can't do that while the patent aggressor makes the Open Source project's own work into part of the patent aggressor's commercial products.
Jacobsen amended his case against Katzer to include a claim that Katzer was infringing the copyright on Jacobsen's software by using it in his product without honoring the license terms or giving any attribution that the software was Jacobsen's.