Saving Open Source

The cause of free and open-source software dodged a bullet earlier this year.


October 01, 2008
URL:http://www.drdobbs.com/windows/saving-open-source/210604978

The cause of free and open-source software (FOSS) dodged a bullet in August of this year. On August 13, the U.S. Court of Appeals for the Federal Circuit, known familiarly as "CAFC" among court watchers, made known its decision in a closely watched case that has immediate and far-reaching implications for anyone who writes or uses free or open-source software. Although Jacobsen v Katzer specifically upheld the Artistic License, the 12-page decision explicitly referred to Creative Commons and other copyleft licenses.

This case was the first real test in any U.S. court of the legal teeth behind open-source licenses. This alone makes Jacobsen v Katzer both important and interesting, but as with most legal cases, the story behind the story is also interesting. And here the case behind the case, and the software at issue, evoke themes that date to the earliest roots of free and open-source software back in the 1950s.

Jacobsen v Katzer

In 2006, Bob Jacobsen sued Matthew Katzer and his company Kamind Associates, alleging that Katzer had violated licensing terms for Jacobsen's Decoder Pro, a utility for controlling model railroad signals and a portion of his Java Model Railroad Interface (JMRI). Decoder Pro was released under the Artistic License, an open-source license. Kamind had, Jacobsen claimed, used the Decoder Pro software in his own Decoder Commander utility without documenting the changes or attributing the licensed software to Jacobsen, both of which would be violations of the license.

Jacobsen brought suit in the U.S. District Court for Northern California and sought, among other remedies, monetary damages and an injunction to stop Katzer from using the code.

The case seemed like a no-brainer: The code in the two utilities spoke for itself, Jacobsen's code was clearly covered by the Artistic License, and Katzer had clearly not followed the license requirements in releasing his utility.

So it was something of a shock, not just to Jacobsen but to the FOSS community generally, when the court handed down its decision. The court refused to issue the injunction and drastically limited the kind of monetary damage Jacobsen could recover, if any, for Katzer's violations.

The linchpin of the court's decision had to do with a legalistic nicety, the distinction between a covenant and a condition. Violating a condition of a license based on copyright violates the copyright, but violating a separate covenant that is added above and beyond any conditions of the copyright does not. The court decided that, because the Artistic License was extremely broad in its scope, not placing very many restrictions on the rights of a licensee to create derivative works from the software, the copyleft language in the license constituted a covenant outside the copyright itself, and the only remedy open to Jacobsen for violation of the covenant was recovery of monetary damages he had suffered from the violation.

Having to prove monetary harm from violation of software provided for free is not all that easy, and the court was saying that this was Jacobsen's only claim. Katzer had not violated Jacobsen's copyright.

The decision was a serious blow to free and open-source software. The whole idea of copyleft licensing is that a violation of the license is a violation of the copyright. The court's interpretation, making the defining conditions of a copyleft license merely extra covenants, would severely weaken the force of licenses, whether the GPL or Creative Commons or some other model of FOSS licensing.

On Appeal

Jacobsen appealed the ruling and the Court of Appeals for the Federal Circuit (CAFC) reversed the lower court. Disaster averted, was the general sense in the FOSS community. But it's worth taking a look at the details.

The CAFC found that the conditions of the Artistic License were in fact conditions on the scope of the license. The court pointed in particular to the phrase "provided that" in the license, a phrase that is generally regarded in law as imposing a condition. And the court concluded that Katzer was liable for both copyright infringement and violation of contract, implying that Jacobsen was entitled to both injunctive and monetary damages under copyright law.

The fact that no money changes hands in a copyleft license didn't mean that there were no economic implications of the violation of the license, the court said. And, interestingly, the court cited Lawrence Lessig's Creative Commons license (creativecommons.org) in its opinion, clearly implying that its decision was not limited to this one license.

The CAFC sent the case back to the lower court for resolution under this guidance.

The implications of the CAFC's reversal are pretty significant. This makes it clear that free and open-source software creators can seek statutory damages without having to prove monetary harm, can get an injunction against an infringer, may be able to recover attorney's fees, and may have other rights as well. All of this would have gone away if the lower court's decision had been upheld. It's a big deal.

This is not the end of the story. Various people have raised questions regarding the scope of the decision and the jurisdiction of the court. The CAFC typically deals with patent law, not copyright; this is a decision in a U.S. court and not binding on other countries; the decision hinges on the language of the license and might not apply to all copyleft licenses; etc. Any speculation here falls under the heading of opinions about opinions not yet rendered, but basically, in the view of Lawrence Lessig and others who have been following the case closely, this decision is likely to set the precedent for other courts and cases and jurisdictions.

As for the fact that the decision hinges on the specific language of the license, that's a positive thing for those crafting copyleft licenses. A Federal court has now articulated just what language needs to be in your contract to make it enforcable under copyright law.

This is an important decision not just for those who create free and open-source software, of course. It also has implications for anyone on the other end. If you've been deliberately remaining ignorant of whether software you use infringes copyleft licenses, this ruling is your wake-up call.

The Case Behind the Case

But why did this case end up in a patent court instead of a copyright court? The reason is that, in a strictly legal sense, a charge of patent infringement was the trigger for the legal actions that led to this case being heard. And curiously, it was Jacobsen who was accused of violating Katzer's patent, although it was Jacobsen who filed the suit against Katzer. The law is indeed a curious thing.

In a sense, nobody really knows what the law is until a case is brought and a judge gets to rule on specific facts. And a case gets brought when some individual chooses to do so, meaning that the detailed facts in a particular case have a lot to do with what elements of the law come to the judge's attention and therefore how the nuances of interpretation play out. But free and open-source software was waiting for its day in court, and this particular case got it there.

Jacobsen works at Lawrence Livermore National Laboratory and is a model train enthusiast. He is the principle developer of the JMRI Project (jmri.sourceforge.net), an open-source project devoted to his Java Model Railroad Interface software. Jacobsen knew he had a problem when he started getting letters from Katzer alleging violation of Katzer's patent, culminating in a request for something over $200,000.

Jacobsen looked at Katzer's software and decided he'd better strike first. He file a preemptive suit against Katzer, claiming that Katzer's patent was fraudulently obtained, invalid, and unenforceable; and that in fact Katzer had improperly incorporated part of Jacobsen's JMRI software in the code for which he claimed a patent. He also charged Katzer with cybersquatting and a few other offenses. And he made an online plea to other developers to help him nail down the prior art to debunk Katzer's patent. Open-source lawyers and developers jumped on board, helping him make his case.

And Jacobsen v Katzer became the test case for the enforceability of copyleft licenses.

Time To Railroad?

It is highly appropriate that the software at the center of the underlying case was code for controlling model trains. Because model train control software is, in a way, where FOSS began.

As Steven Levy recorded in his 1984 book Hackers, the first true hackers were members of the Tech Model Railroad Club (TMRC) at MIT in the late '50s, and the first memorable articulation of the principles on which the FOSS movement are based was the Hacker Ethic espoused by the members of TMRC.

The model railroaders—specifically, members of the Signals and Power Subcommittee of the TMRC—spent hours on their backs under the monster table that supported a room-sized world of HO-gauge tracks and scale models of roads and buildings and trees. They were managing the spaghetti wiring that supported the multiple controllers and trains that ran on those tracks.

When the first class in computer science was offered at MIT, taught by some crazy guy named John McCarthy, the TMRCers already had a complex system that they were trying to manage. Many famous hackers came out of the TMRC, as did, you could argue, the philosophy behind free software.

Interestingly, the control system under the table in the model railroad room would seem to have dealt right from the start with the problem in Jacobsen v Katzer, in that it managed signals from multiple controllers distributed to different sections of track. Complex model train systems have been revisiting that problem for forty years. What this means in terms of prior art in Jacobsen v Katzer is a matter for others to decide.

Credit where credit is due: It would be easy to think that the case was simply a matter of a more savvy court correcting a misguided one, but in fact a lot of work went on behind the scenes to ensure that the decision ultimately came out the way it should—and did. Credit goes of course to Bob Jacobsen for bringing the case and standing up for FOSS. But credit also goes to the lawyers and consultants who offered time and expertise, often for free: Jacobsen's lawyer, Victoria Hall; the authors of the amici brief from Creative Commons, Chris Ridder and Anthony Falzone; Mark Radcliffe, Lawrence Lessig, Karen Copenhaver, Allison Randal, Roberta Cairney, Larry Rosen, Scott Peterson, David Gross, and Steve Chiari; plus the Stanford Center for Internet and Society, the Linux Foundation, Perl Foundation, HP, OSI, and others.

An individual identified as GrayHairPgmr on Lessig's blog put the ruling in this perspective:

In my not so humble opinion, software IS a service, the code is only one delivery vehicle. SaaS is redundant and proprietary software is an oxymoron. This ruling begins a process of building a meaningful legal framework for software commerce.

It's a big deal.

Terms of Service | Privacy Statement | Copyright © 2024 UBM Tech, All rights reserved.