Comments on GPLv3

by Lawrence Rosen[1]

Many long months ago I committed to help Eben Moglen and his colleagues at the Free Software Foundation and the Software Freedom Law Center write a better license than GPLv2. They have done so and I applaud their efforts. GPLv3 is clearer about its purposes and effects than its predecessor. It is a legal document intended to be honored and obeyed, and so I am glad that we can now know more precisely what is permitted and what is prohibited under its terms.

FSF’s software and certainly much other software will soon be available for free under the terms of GPLv3. In some respects, GPLv3 is like all other licenses imposed upon the world by the owners of intellectual property. There is not a public domain commons of software that runs commerce worldwide; it is software that is available by the goodwill of its owners and licensed to recipients under terms stated. In the United States there is a foundational legal principle of “freedom to contract [or license] under any terms you want,” and GPLv3 takes full advantage of that principle to build a collection of free software available to all under copyleft terms.

But that GPLv3 software isn’t the only such collection of free software in the world. Many distributors, including most prominently the Apache, Eclipse, Mozilla and Perl foundations, have their own licenses and their own collections of software. I support the work of those other foundations also. The very first sentence of GPLv3 summarizes it nicely: “The GNU General Public License is a free, copyleft license for software and other kinds of works.” Notice that the license doesn’t claim to be “the” license, or “the best license for all purposes.” The GPLv3 license is merely (but emphatically!) the next generation of a leading member of a community of licenses for free and open source software.

GPLv3 is a statement of the rules by which certain owners of software will distribute their works. From my perspective, it is a far better license for those purposes than many other licenses, although you probably will not be surprised to learn that I still prefer my own Open Software License (OSL 3.0). No single license can satisfy everyone. Richard Stallman and the FSF are entitled to the terms and conditions they believe in—to satisfy their own goals and purposes—for their own intellectual property. They are entitled to this, just as my other friends and colleagues in the community are entitled to their open source licenses, indeed just as the owners of intellectual property are entitled to license it under proprietary terms if they wish.

My major objective when I agreed (many long months ago) to help Eben Moglen and Richard Fontana on Committee A was to ensure that GPLv3 played well in the entire ecosystem of free and open source software. I define those terms more broadly than some in FSF, but I specifically include software available for free under very generous terms from the Apache Software Foundation. By way of disclosure, I am an attorney representing the Apache Software Foundation. License compatibility between GPLv3 and the Apache License was as important to my clients as it was to Richard Stallman.

Apache software can now be included in GPLv3 projects. I’m particularly delighted that these two large communities are now able to work together compatibly in this way!

More broadly to my satisfaction, though, the GPLv3 license contains clearer language that will encourage its software to be used in “aggregations” (what in United States law we call “compilations” or “collective works”) that include many independent works. Computer professionals will better understand these independent works as modules of code, so I’ll use that word instead. As long as those modules are independent works under copyright law, and as long as their own independent licenses allow this to be done, verbatim copies of those modules can be combined to build larger systems of increasing sophistication and breadth. GPLv3, as I now read Draft 4 the license, allows that.

I have long counseled that GPLv2 allowed that result also, but some argued otherwise. All the FUD that I and others have spread over the years about linking—static, dynamic, otherwise—didn’t help matters much. This argument can be replaced now by intelligent conversation about the actual words of GPLv3, which I understand to allow aggregations of verbatim copies of independently-written modules taken from the worldwide portfolio of free and open source software. I have always read the Open Source Definition as mandating that open source licenses permit the copying and distribution of verbatim copies including for collective works, and this added clarity in GPLv3 supports that interpretation. Perhaps now our software commons is vastly larger for collecting independent open source modules into interoperating solutions for our customers.

Of course this all remains, despite the increased clarity of GPLv3, a complex legal topic requiring careful counsel: From a copyright law perspective, the original licenses continue to apply to those verbatim copies, but not necessarily to the collective work as a whole. Collective works are independently copyrightable to the extent they required creativity to collect. Also under copyright law, don’t confuse collective works with derivative works. The improved clarity of the definitions in GPLv3 Section 0 may help explain these concepts to developers and distributors of software, for which I thank the authors of the license. But you should still consult with your own attorneys to ensure that you clearly understand the legal distinctions I’m drawing here.

I predict that one of the biggest success stories of GPLv3 will be the realization that the entire universe of free and open source software can thus be combined into comprehensive open source solutions for customers worldwide.

My satisfaction with this aspect of GPLv3 is partially offset by my concerns that the patent provisions in Section 11 will impede the overall success of the GPLv3 license itself. I have fought over the years about the wording of many a patent provision in many an open source license, and I have discovered that some provisions make licenses unacceptable to big patent holders—and thus impede adoption of open source software. Simply put, companies can conclude that free software is not worth the loss of their patent portfolios; if those companies are your potential customers or business partners, they may not accept your software under your GPLv3 license.

The Free Software Foundation is right to be concerned about software patents and right to want to defend our community from them. But I do not believe that the best way to do so is to threaten to emasculate the patent portfolios of big companies so that they refuse to play with us at all.

These debates about patent provisions in licenses usually take place in private because big companies that have commercially-significant patent portfolios aren’t anxious to disclose their intellectual property strategies in public. (Recent press about the potential use of one company’s patent portfolio against open source illustrates the public relations risk of saying aggressive things about patents in public!) For GPLv3, negotiations over its patent-related license provisions were undertaken primarily in Committees B and C on which I didn’t serve, and so I don’t have much insight into the justifications for the compromises that apparently were made. But I personally fear that the resulting compromises, and the words of GPLv3 Section 11, will impede the development and distribution of free software by frightening patent owners away. That is a shame.

Bottom line: I believe GPLv3 is a good license that many of us can live with, but it will never be the only license that serves the free and open source community. Owners of software have other good licenses to choose from that will meet their own philosophical and business goals. The Apache, Eclipse, Mozilla and Perl foundations and many other projects will continue to build upon their own visions of free and open source software using their own licenses. To the extent that GPLv3 embraces participation in that larger community, I’m quite satisfied with it, even though it won’t necessarily be the license I’ll recommend.

[1] Copyright © 2007 Lawrence Rosen. (See

Licensed under the Academic Free License version 3.0. (See

None of the opinions in this article necessarily represents or reflects the views of my clients, including the Apache Software Foundation or any of the other projects or organizations mentioned herein and whose work I support. My views about GPLv3 are my own. I thank Eben Moglen for inviting me to participate on GPLv3 Committee A, and I thank Richard Fontana for ably representing FSF’s interests on that committee. Lawyers and engineers alike should recognize that giving birth to a new license is always a difficult task, particularly in a very public process where there are almost irreconcilable views about important legal matters. I congratulate FSF as they approach the end of that process. Richard Stallman and the Free Software Foundation now have the license they want, and they are entitled under the law—and under free and open source software principles—to apply it to their valuable software. I wish them success.