Jacobsen v. Katzer Settled

The open source model train case has settled, obviating a second filed appeal to the US Court of Appeals.  The press has been describing the settlement as a big win for open source, which is not strictly accurate, but the result overall is a win for open source licensing.  The settlement involved a monetary payment, but there have now been other lawsuits that resulted in monetary settlements for violation of open source licenses.  And that means nothing in a true legal sense — though it underscores that corporate users of open source are fearful enough of enforcement actions that they will write a check to avoid them.

It is significant that this case involved a license other than GPL — the original Artistic License — which is generally considered permissive.  If a permissive license is enforceable, that lays substantial groundwork for enforceability of copyleft licenses.  Here, the actual harm to the copyright owner was at issue.  For Jacobsen to prevail, the court had to decide that there was sufficient harm to give rise to a remedy.  With a permissive license, the harm is, essentially, failure to deliver notices.  For a copyleft license, the conditions are much more significant, and thus violating them would give rise to greater harm.

The settlement was interesting in two ways.  It included an injunction by the district court against further infringing activities, and it was not sealed.  Most settlements are confidential, and an actual injunction by the court in a settlement is a bit unusual.  But the injunction would have carried more precedential weight if it had been won in court rather than agreed to in settlement.  (The District Court had previously declined to issue an injunction.)

The big win here was not the settlement itself, but the the opinion of the Federal Circuit in the first appeal, which was handed down last year — and secondarily, the order for summary judgment in December, to the extent it supported an award of damages for copyright infringement notwithstanding the royalty-free distribution of the software by the owner.

In fact, the conventional wisdom among those of us law geeks who live and breath open source legal issues is that Jacobsen v. Katzer was very nearly a case that made bad law.  It was plucked from the stormy sea of District Court by the Federal Circuit — an unlikely hero in the open source world.  But in a common law system, we don’t get to pick the cases that make law.  And a settlement means no more law gets made at all.  In other words, game over, but at least we beat one of the boss levels. 

MAEMO + MOBLIN = MEEGO

Intel and Nokia announced the merging of their Linux platforms, under the name MeeGo.  Intel’s Moblin was directed at netbooks, and Nokia’s Maemo was focused on tablets.  Cooperation on MeeGo signals the likely convergence of these device categories.  Meego has been placed under the stewardship of the Linux Foundation.  The Foundation’s Executive Director Jim Zemlin cites the need for MeeGo as driven by demand for always-on Internet connections and multi device compatibility.

Black Duck Obtains Patent on License Compliance Method

Black Duck announced that a patent issued on its license compliance methodology.  This move is likely to be unpopular with the open source community, given its opposition to software patents generally.  The patent application was filed seven years ago, which is a long lead time even in the glacial pace of US patent prosecution.  Black Duck’s Tim Yeaton blogged about the patent, floating a trial balloon about an IBM-style patent pledge.

SourceForge versus the Export Laws

https://www.eff.org/deeplinks/2011/09/stop-the-piecemeal-export-approachA news flurry has erupted over SourceForge’s policy denying access to users in nations sanctioned by US export law.  A blog states that The Obama administration insisted that Sourceforge adopt this policy — though the restriction in Sourceforge’s terms of service has already been in place for many years.  Sourceforge, of course, is caught in the cross-fire between US export regulation and the open source definition.  But statements that such a policy contravenes open source licenses are misguided — there are plenty of reasons why the rights granted under open source licenses may not be freely exerciseable: regulatory law, trademark law, patent law to name a few.  Those who are complaining about Sourceforge’s actions are probably unaware of the regulations that apply to software, and that they used to be far more egregious before they were revised in the early 1990s.  But the kerfuffle points up that the regulatory policy of the US is counterintuitive and awkward to apply to software.  It strains belief that any nefarious actor would be truly thwarted by a technical limitation on downloading code to a certain location.
The language of Sourceforge’s terms of use is standard in the on-line content industry, and if you read to the end, in almost every on line terms of service and end user license. 

 

Update 2/19/22: the ZDNet link above broke; here are alternatives:

Syria: Netizens Discuss SourceForge Ban

https://www.eff.org/deeplinks/2011/09/stop-the-piecemeal-export-approach

Bruce Perens on the Busybox Case

Bruce Perens, the original author of BusyBox, commented in his blog about the recent wave of enforcement cases from SFLC.  Perens was the original author of Busybox.  The utility has morphed over the years, and the current authors claim Perens no longer has any copyright interest in it because none of his original code is in the code base — which Perens disputes.  Perens is not party to the recent suits.

This points up a significant legal issue lurking in open source enforcement — the question of whether open source programs are joint works or multiple works of authorship.  Rights in joint works can be enforced only with all the authors’ participation, whereas a quilt of individual works can be enforced by individual authors separately.  This means serious potential problems for a defendant who must settle with a series of claimants rather than a single joint author group. 

Also, even if no original code exists in a modified work, it can still be a derivative work, as every copyright lawyer knows.  (Consider, for instance, a port into another language, or a rewrite using the identical structure, sequence, and organization.)

FSF generally takes the position that GPL covers an entire program as a single copyrightable work — if not, then its position on linking creating a single executable work would fall into question.  But this position may contradict the idea that a program is a combination of multiple works, each of whose author has a separate right of action.   This issue will likely become crucial one day in a GPL suit.

Please see my previous entry for the facts concerning the suits that were filed.