HT to PatentlyO.
This video from Red Hat was apparently meant to promote Red Hat’s efforts to rid the world of software patents. The video begins with the title “Liberating Innovation” and the first speaker is Rob Tiller, Red Hat’s Assistant General Counsel. Tiller opens the discussion by talking about encouraging innovation in the context of what he calls “free and open source software” (FOSS for short). Presumably, Tiller and Red Hat believe that the FOSS model is the best way to encourage innovation in the software space.
However, they obviously feel differently about encouraging innovation in creative expression. Originally, I planned to record my own video response to Red Hat’s position, interspersing my reactions following the various points made during the video. However, I have not done so out of the fear that creating such a work would be considered an “Adaptation” which is expressly prohibited by the license agreement that accompanies the video. More on that later…
First, a few reactions:
1. Rob Tiller – uses the word “free” 4 times.
Truthfully, this is not incredibly offensive, as he uses the word free in the phrase “free and open source software” which has become somewhat of a term of art. However, in my conversations with economists in the past, I have often gotten into trouble by using the economist version of the “f-word,” so it is important to clarify what the “free” in FOSS means. Tiller is probably aware of a case called Wallace v. IBM, where a software developer attempted to sue Red Hat (among other companies), alleging that the FOSS model constituted an illegal price fixing arrangement (seriously). Judge Easterbrook disposed of the case very easily, and in the process explained how (1) the FOSS model exists because of rather than in spite of the copyright system, and (2) that while no money changes hands, consideration is exchanged that makes the open source license (in this case the GPL version) an enforceable contract.
People may make and distribute derivative works if and only if they come under the same license terms as the original work. Thus the GPL propagates from user to user and revision to revision: neither the original author, nor any creator of a revised or improved version, may charge for the software or allow any successor to charge. Copyright law, usually the basis of limiting reproduction in order to collect a fee, ensures that open-source software remains free: any attempt to sell a derivative work will violate the copyright laws, even if the improver has not accepted the GPL.
Wallace v. IBM, 467 F.3d 1104 (7th Cir. 2006).
In other words, in the absence of copyright, the FOSS model wouldn’t exist at all. While it is true that any software developer could still write code and give it away, the developer would have no assurances that subsequent developers would “return the favor.” One need look no further than the explicit terms of the GPL v.2 license agreement that is used by Red Hat to verify this. For example, while the agreement generally grants the right to make and distribute derivative works, it also specifically conditions that grant:
4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License.
* * *
5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License.
The “prohibited by law” language of Section 5 is referring, of course, to copyright. In other words, if you create derivative works and distribute them under terms that are inconsistent with the GPL license, then you do so without permission of the original copyright holder, and are exposed to liability for copyright infringement.
2. Tiller goes on to claim that “patents are a hinderance to innovation with free and open source software.”
How could this be so? First, for FOSS participants, the newer versions of GPL expressly require the contributor to grant licenses under any patents that embody the software (and this is arguably implied in the earlier GPL licenses). In addition, the contributor is obligated to “clear” any known patent issues on terms that provide downstream protection, or else they are not allowed to distribute their work.
Conversely, patents held by non-FOSS participants present no more or less of a hinderance than copyrights held by those same persons. That is, if a FOSS developer copied source code from a non-FOSS participant into a Red Hat product, the non-FOSS participant would have a valid copyright infringement claim against Red Hat. A non-FOSS patent holder would essentially be in the same position as the copyright holder. That is, if Red Hat software infringes patents owned by non-FOSS participants, Red Hat is similarly vulnerable to potential infringement claims. Thus, patents present no more of a hinderance to FOSS than do copyrights, and, as discussed above, FOSS could not exist without copyright protection. Red Hat can’t reasonably expect to control the behavior of the entire world simply because they have adopted the FOSS model.
The simple truth is, Red Hat is free to innovate as it sees fit. Participation in FOSS is an individual choice, and Red Hat’s choice doesn’t put in any better or worse position vis-à-vis third party infringement claims than does a proprietary development model.
3. Professor Michael Meurer claims that the “successful innovators” are the ones that are targeted with infringement suits.
This is undoubtedly true. However, I might ask Professor Meurer why he would be suggesting that the software industry is unique in this regard.
4. Jan Wildeboer—who carries the title “Open Source Affairs Evangelist”—is clearly the most intelligent person in the video.
Jan suggests that patenting makes little, if no sense for software companies. First, he greatly understates the current lag time between filing and issuance as 3-4 years. In some cases, it might be 3-4 years before your application is even examined. In any event, there is no doubt that this encompasses several complete product cycles, and the life of the patent extends another 16-17 years (or more, depending on term extension) beyond an already expired product.
There is nothing untrue about this statement, but this is also not unique to the software industry. However, this should simply prompt software companies to re-evaluate their patenting strategy. Instead of narrowly focusing patent applications on specific software products and features, energy should instead focus on broader innovative concepts that survive from one product cycle to the next.
Alternatively, software companies could lobby for policy changes. If software companies think a 20 year patent term is too long, they should take a look at drug patents, where drug companies almost universally think that a 20 year term is too short. In fact, Congress specifically passed procedures that allow drug companies to extend their patent term based on delays at the FDA. Hypothetically, software companies could lobby for a system where applicants could elect an alternate patenting process that was examined on a much faster timeline, but perhaps provided a shortened term. In any event, those types of questions (or even eliminating patents on software altogether) are policy questions that are properly directed at Congress.
5. Tiller complains that it is not possible to know the boundaries of the various patent claims.
Join the club. Once again, there is nothing unique to software companies in this respect. I actually do have thoughts on how to fix this problem in general, but more on that later…
6. Tiller states that Red Hat will not assert any patents against users of free and open source software.
In fact, Red Hat’s patent pledge (which constitutes an unenforceable, naked promise as far as I can tell) states that Red Hat will refrain from enforcing their patents against software that is licensed under a list of various approved open source licenses. In other words, any competitor to Red Hat that happens to license their software under one of the approved licenses (which are common and available for anyone in the public to use) gets a free pass on infringing Red Hat’s patents. This is worth mentioning because Red Hat’s patents are assets of the company. They do not belong to Rob Tiller, nor do they belong to Michael Cunningham or even Jim Whitehurst. Rather, the assets belong to the shareholders. It may only be a matter of time before somebody asks whether, and under what conditions such a decision falls within the protection of the business judgment rule.
In addition, Red Hat’s software, with a few exceptions, is licensed under an EULA that provides: “Red Hat, Inc. (“Red Hat”) grants to you (“User”) a perpetual, worldwide license to the Programs pursuant to the GNU General Public License v.2.” As noted above, the GNU Version 2 specifically allows for derivative works: “You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work.”
This discussion brings us back to the claimed copyright and license terms of the video itself. As noted above, the video is licensed under a Creative Commons Attribution-NoDerivs 3.0 license, which specifically provides that “you have no rights to make Adaptations.” For the avoidance of doubt, the “human readable” version of the license specifically states: “You may not alter, transform, or build upon this work.” (emphasis added). This irony should not be lost on Red Hat’s shareholders: Red Hat protects this video with a stronger license than it uses to protect the source code for its commercial products.
Oh, and in case you were wondering:
This work is licensed under a Creative Commons Attribution 3.0 United States License.
Rob, feel free to transform away …
- FOSS can implement patented standards (fosspatents.blogspot.com)