I attended and spoke at the FOSDEM 2012 Legal Issues DevRoom (Update 20120217: slides, blog posts) organized by Tom Marble, Bradley Kuhn, Karen Sandler, and Richard Fontana. I understand the general idea was to gather people for advanced discussions of free/libre/open source software legal and policy issues, bypassing the “what is copyright?” panel that apparently afflicts such conferences (I haven’t noticed, but don’t go to many FLOSS conferences; I bet presenters usually get the answer only superficially correct). I thought the track mostly succeeded (consider this high praise) — presentations did cover contemporary issues that mostly only people following FLOSS policy would have heard of, but I wished for just a bit more that would be news or really provocative to such people. In part I think 30 minute time slots were to blame — long enough for presenters to belabor background points, short enough for no substantive discussion. Given only 30 minutes, I personally probably would have benefited from a 15 minute speaking limit, thus being forced to state only important points, and leaving a little time for participants to tear those apart. Yes, I should have imposed that discipline on myself, but did not think of it until now.
Philippe Laurent gave an overview of cases involving “Open Licences before European Courts”. He did not list one recent “open content” case, Gerlach vs. DVU.
Ambjörn Elder on “The Methods of FOSS Activism” spoke about political activism; a worthy topic, but I hope for more discussion of activism for software freedom, rather than against ever worse policy.
In place of Armijn Hemel’s “Goes into an Executable? Identifying a Binary’s Sources by Tracing Build Processes” (missed flight) Kuhn and Sandler excerpted from a presentation on and took questions regarding nonprofit homes for free software projects. Writing this reminded me to make a donation to Software Freedom Conservancy, of which Kuhn and Sandler are respectively ED and Secretary of. Somewhat tangentially, I don’t find the topic boring, but I do find the lack of information, informed-ness (including mine), and tools regarding it boring. I don’t know of any libre documentation on running a nonprofit — I’d love to see a series of FLOSS Manuals on this. OneClickOrgs is a fairly new free software project to handle some aspects of governing a small organization, but I don’t know how useful it is at this point. Related to lack of documentation, some of the Q&A emphasized how little people know of these topics across jurisdictions — nevermind rule minutiae, even the existence of relevant “home” organizations.
Dave Neary on “Grey Areas of Software Licensing” questioned whether one could legally do various things, using examples largely drawn from GIMP development. The answer is always maybe. Fortunately developers sometimes take that as yes.
Allison Randal gave an overview of FLOSS history with a focus on legal arrangements in “FLOSSing for Good Legal Hygiene: Stories from the Trenches”.
Michael Meeks on “Risks vs. Benefits on Copyright Assignment” made the case that assignment (and some non-assingment contributor agreements) is harmful to participation, and proprietary re-licensing has not proven a good business, so a corporate sponsored software project ought to either be free (sans assignment and potential for propreitary relicensing) or proprietary, and fully enjoy the benefits of one or the other, rather than neither. He also indicated that permissive licensing can be better than copyleft for a free software project with copyrights held by a corporation, as the former gives all effectively equal rights, while the latter abets proprietary relicensing and ridiculous claims that the corporate sponsor will protect the community. Meeks repeatedly called on the FSF to abandon assingment, as for-profits disingenuously cite FSF’s practice in support of their own (FSF ED John Sullivan responded that they are getting corrections made where FSF practice is inappropriately cited and will work on explaining their practice better). Finally, Meeks requested an “ALGPL” which would require sharing of modified sources used to provide a network service, like the AGPL, but allow modifications that only link to or the equivalent ALGPL codebase to not be shared. I don’t know whether he wants GPL or LGPL behavior if such modificaitons are distributed. I was somewhat chagrined (but understanding; just not enough time, and maybe nobody submitted a decent proposal) that this was the only1 discussion of network services!
Loïc Dachary on “Can for-profit companies enforce copyleft without becoming corrupt like MySQL AB?” said yes, if they aren’t the sole copyright holders; on projects he is hired to work on, he seeks out additional contributors who will hold copyright independently.
John Sullivan in “Is copyleft being framed?” presented some new data, apparently replicable (based on Debian package metadata), showing that GPL-family licenses are used in the vast majority (did I hear 87%?) of Debian packages. Update 20120217: I did hear 87%, in 2009, and 93% in 2011. Note some software available under multiple licenses. Slides.
Richard Fontana on “The (possible) decline of the GPL, and what to do about it” suggested the need to start thinking about GPLv4, but I’m not sure for what issues2 — doesn’t matter; if the particulars of licenses can make a big difference, requirements for the next version of important ones should always be a relevant topic, even if there is no expectation of creating another version for many years. Fontana also indicated that perhaps the next (massively adopted, presumably) copyleft might not be created by an existing steward3 (meaning the FSF, or obviously CC in many non-software fields), which I take as an indication that license innovation is possibly more important than compatibility and non-proliferation.
I don’t remember much of panels with Hugo Roy, Giovanni Battista Gallus, Bradley Kuhn, Richard Fontana on application stores and Ciarán O’Riordan, Benjamin Henrion, Deb Nicholson, Karen Sandler on software patents, as I was probably preparing for my talk, but I trust that free software is still important if mode of delivery changes slightly and that software patents ought be abolished.
Most of the slides from the day should be available soon on the DevRoom’s page. Some audio might be available as well eventually.
Kuhn demonstrated his qualifications for another fallback career: crowd crontol. Fontana blogged a summary of the devroom. Sandler gave the most important talk on FLOSS policy (but not at FOSDEM). Marble apparently did almost all the organizing. Thanks to all! There will be another legal/policy devroom next year.
Addendum 20120210: Richard Fontana offered these corrections:
1“re network services, I mentioned rise as factor in possible GPL decline, coupled with AGPL pwned by dual-license hucksters”
2“main reason for GPLv4 right now is GPLv3 is needlessly complex, limiting popularity of strong copyleft.”
3“growing concern that anti-license-proliferationism concentrates power in privileged Establishment organizations”