I have frequently expressed myself about the patent and business agreement between Novell and Microsoft. One and a half year later, this agreement continues to concern me and many others; last week however, we learned about another patent agreement, or rather a legal settlement on software patents involving Red Hat and two other companies that are sometimes referred as « patent trolls ».
As I commented on Novell’s patent agreement, I felt it would also be interesting, if not only fair, to comment on Red Hat’s patent settlement. NB: IANAL.
What happened in short:
Some time ago, some small and unknown companies sued Red Hat on the basis of alleged patent infringements. The issue at hand was theoretically a group of patents filed by FireStar Software covering some technology found in relational databases and was claimed to directly affect Hibernate and Jboss, two Red Hat solutions.
At the time the lawsuit started there was wild -but by no way foolish- speculations going on about the possible involvement of Microsoft behind that lawsuit. To this day, such an involvement has not been proven. Yet, it is known that Microsoft has so far made several public assertions on patents it would own on Linux and several other Free Software. The difference between these claims and the legal settlement at hand is of course that Microsoft has so far threatened, but never showed what it had in its satchel. Are they bluffing? Probably. But by threatening fear, uncertainty and doubt can be easily spread and disseminated, which is exactly the reason why some users would not adopt FOSS.
Back to the Red Hat’s patent settlement. The outcome of the settlement can be (simply) explained through two points:
Red Hat successfully demonstrated that the patent claims and the patents filed were invalid, and it demonstrated it through prior art. That means, there was a thorough research and subsequent evidences showed that technologies claimed to be patented had already been invented before the companies owning the patents had even filed the first application. In short, what the plaintiffs had claimed they had invented turn out to be bogus.
Red Hat negotiated not just the mere logical conclusion stemming from facts established in court, it went a big step further, and that’s what has made this settlement so special. Red Hat extended the immunity related to the litigious patents to both its customers, but also to the whole community (some might say the universe). Even better, this settlement not only covers Red Hat and Fedora platforms, it covers anybody using the problematic software (the software apps that were allegedly infringing on the patents) wether or not it is using Red Hat. The settlement also extends this protection to any version of the covered software apps past, present and future.
Now let’s take a step back and compare it to the patent agreement between Novell and Microsoft:
Novell did precisely not sign a patent agreement with Microsoft debunking any claims or myths related to FOSS infringing Microsoft’s “intellectual property”. It implicitly did just the contrary: Microsoft and Novell were teaming up to “protect” Novell customers against patent claims made by…Microsoft.
The agreement was only covering Novell customers (not even OpenSuse users) and was at the same time contradicting the GPL (v2). Red Hat’s settlement does not seem to conflict with any version of the GPL as it places no burden or extra deeds on users and developers of GPLv3 software (More on that later).
There was no prior art, no litigation, and perhaps as important as the rest, the Novell-Microsoft agreement involved money. Lots of it . On the other hand, Red Hat received to my knowledge no payment for the settlement and as a future outcome, no strong incentive to do business with the plaintiffs and have its existing customers sign some dubious “software patent insurance”.
So what made this patent settlement really historical? In my opinion, the extension of the effects of the settlement to the universe has made that settlement historical, and also provides the community at large with a strategical framework to deny any allegation of software patent infringement.
However, it should be noted that what this settlement was all about was not about denying the existence of software patents. Remember that software patents do exist in North America and other parts of the world, and that some large corporations out there are working hard at pushing them in the European legal framework as well (and they don’t care about what people think, because Die Partei had immer Recht. ) What this settlement did was to take the patent threat away from nontrivial Free and Open Source software applications by demonstrating prior art, essentially demonstrating how the whole filing proved ignorant of the state of the art or perhaps, as some will point out, how disingenuous the whole adventure was.
Last but not least, one might wonder how “compliant” or “compatible” the results of the settlement are in respect to the terms of the GPL v3. Groklaw has a pretty good post on this issue. Once again IANAL but here are some thoughts about this.
Contrary to what can be read here and there, the GPL v3 does not deny the existence of software patents. That would be stupid, as the GPL has been designed to live in a legal environment where such patent claims would be made against Free Software. What it does however is denying software patents the possibility to infringe on the users and developers’ freedom and ability to run, use, modify and redistribute software. If the software cannot be redistributed without some form of immunity in regard of software patents, then a poison pill option exists. Red Hat’s settlement does not apply here. What Red Hat did was “clean” patent busting: they demonstrated prior art first, and then essentially killed the patents by extending the immunity to those patents to anyone using the problematic software apps. That’s how historical it gets.
I am, just like many others, left not wholly satisfied by this. I am very happy of course, of the outcome of this lawsuit, but I know that the real issue at stake is software patents and that what will really put all these issues to rest, ultimately, shall be the end of software patents.