If I understood it, his opinion was that the license terms of the GPL would not survive resale, due to the well established “first sale doctrine” and its EU equivalent “exhaustion”.  It basically means that the copyright holder cannot stop you reselling your software, and that the license terms will not apply to the guy receiving it.

I tried to understand this further, but Alexander was not always easy for me to comprehend and had then a habit of linking to his own posts elsewhere to bolster his position, leading to a kind of echo chamber of Terekhovs all nodding vigorously at each other.  He also back then and evidently more recently too explained legal decisions that did not fit his understanding by calling the Judges in question “morons”, etc.  Well the forum I met him at had a very high trolling quotient so it just joined the rest of the anti-GPL sentiment there for me in the end and I ignored it.

GPL is a license too

But I was reminded of this last night when I read about a recent decision against Autodesk which is being widely seen as a victory for Joe Softwarebuyer.  From the Patry blog post link above:

…many software companies have taken the position that they can convey the copy to the customer in an over-the-counter transaction for a one-time payment, but describe that transaction as a license; as a license, the first sale doctrine doesn’t apply, meaning copyright owners can prevent further distribution of the copy…

Doesn’t this vindicate Alexander’s position?  How can GPL terms stick past resale if Autodesk EULA ones don’t?  Nothing stops “built-in” or “automated” resale to clense software of any licensing restriction.

A lot of people seem to be happy about the paid-for world being freed from license conditions, are they going to be happy if it turns out that everyone is also freed from GPL conditions?

Civil infringement and Punishment

What effect would this have on contribution I wonder.  It seems to me the real-world advantages from being active in a project by contributing will still apply.  But it will enable private proprietary forking for products, the kind of thing that Harald Welte’s gp-violations.org has had success attacking and punishing to date.  Contributors will see their work used in commercial products without the changes being open.

But the BSD folks seem to survive this outrage without it removing their motivation.  And from time spent looking at music licensing over the years, I kind of recognize an element of proprietary vindictiveness in gpl-violations… of course the member companies hiding behind the RIAA attacks are also “perfectly within their rights” to embark on much worse vindictive destruction, but they are not entirely dissimilar and that always bothered me.

Playing ball or going home?

Well, this decision is subject to appeal, will only apply to the jurisdiction of that court, etc, so the sky didn’t fall in already.  But there is quite a bit of harmonization of copyright law thanks to the insistence of rich rightholder companies mainly from the US side.  But if this is upheld, it may come to contaminate most Western countries and turn GPL terms in unenforcable noise — the choices would be in effect public domain or closed.

I guess some people will go closed rather than have their work exploited, but I expect most people will just continue on, and contributions will continue to come perfectly fine.  The advantages from being a visible contributor and taking upstream directly are still going to apply, so will the bitrot that happens to any additional code put on top and maintained privately.

Too mature to care?

Maybe now we reached a point that the social, financial, engineering and public advantages from cooperation are ingrained enough that we don’t need a license to protect them anyway?  But I read this and I feel a sinking feeling about the naivity of such a proposal.

RMS (Richard Stallman) just had an interview with the fairly major investment magazine Red Herring. A lot of the content isn’t new from him, but as usual some of it is spot on.

Q: Do you think you will ever achieve that goal?
A: I don’t know because it depends on you. That’s why I resist these self-fulfilling prophecies. If enough of us demand freedom we’re sure to win and if few of us demand freedom we will almost surely lose. It’s entirely up to the readers of this article. As so many issues are to the extent that we still have democracy; so if people were told by businesses they want this and you know you can’t [oppose] businesses so just get used to it, go along, suffer. If people lie down and take it then they will lose. So what do [businesses] do? They are smart: they encourage people to lie down and take it.

[This happens on] many issues and not just this one. Pick any political issue in which things get worse and you’ll find people telling the public: ‘It’s inevitable. Don’t try and fight it; it’s useless. of course, if we did bother trying to fight it, we might win.

A couple of items down on LWN where I found the link, there is another story, about ESR (Eric Raymond) joining the “Leadership Team” of Freespire, and it is the juxtaposition of the two stories that is the real story.

Although the Press Release says people were surprised “in recent weeks” about ESR’s speaking out about the need to work with proprietary data formats, his interest in the issue goes back as far as March 2006. He was on fedora-devel then making the same suggestions with the same urgency, basically that although Linux was doing quite well, it was really let down by the lack of support for proprietary codecs. There followed a discussion about why there was no MP3 support out of the box in Fedora; the patent situation appeared to be news to him. He was basically campaigning for Redhat to start defying patentholders and ship mplayer and other such contraband to allow the most complete possible proprietary data format out of the box. It was explained to him that if we as generally pretty penniless individuals decide to download and use mplayer, that is one thing; if a cash-rich American corporation like Redhat decide to start distributing some of the mplayer stuff, which is fairly clearly containing copyright and patent infringements in the US at least, it would strongly motivate the rightsholders to mount an attack to separate the proposed Redhat-idiot from his pile of cash. Redhat are lawyered up enough to be completely alive to the danger, even with regards to MP3. After several days of back and forth, with Alan Cox weighing in quite negatively towards ESR, he moved on to greener, well, less Red, pastures.

So as it happened it was my good self that perhaps brought linspire’s more commercial attitude towards Linux media player apps to the attention of ESR. To underline a point I will be developing in a moment, I sent him a link to linspire’s legally patent-licensed DVD player app several months ago. (Michael Robertson, Linspire honcho, earned some serious gratitude from my family and I for funding, via a prize after the fact, the original Xbox Linux hacking work that I got a fair chunk of, in fact that kept us afloat for about a year).

Now one of the things I realized during that thread, which is the point of this post, is that the conspiracy between copyright law enabling the licensing of works how the rightsholder sees fit (consistant with compulsories that exist in the case of music) and patent law enabling the holder of the patent rights to control the ability of people to play back content that can only be decoded according to their patent, gives proprietary software like Windows a niche that it can’t be winkled out of by FOSS equivalents.

If content rightsholders insist on patented codecs for their content, well, that defines their content as needing licensed playback devices, and that in turn (exceptions like the recent weirdo-licensed free MP3 license for FOSS aside) insists that there is paid-for player doing duty, which violates the share and share-alike basis of pure FOSS. If content rightsholders insist on end-to-end TPM-backed crypto lockups in addition, well that requires a proprietary hardware system with a proprietary OS.

Therefore proprietary software is validated and given meaning by the rights conferred by Copyright and Patent laws. It’s given a future by the deeply embedded and accepted laws that underpin expressions of creativity in developed countries. FOSS can’t equally compete without violating laws that have been proven many times to have vicious teeth: this is an area where FOSS can’t do what it is doing in the areas that are not so wrapped up with globally enforcable rights.

And now we come back around to the elements at the start of this post. RMS knew this long before I worked it out in the middle of an argument with ESR. RMS does not have a lot of time for the traditional media channels

Q: What is the solution to making the free software movement successful?
A: People should boycott all digitally restricted media and if you can’t get your computer to copy it then you shouldn’t buy it. If you don’t have free software to read a DVD you shouldn’t get a DVD. We are calling a boycott on things like HD-DVD and Blu-ray. The solution is to eliminate DRM. There is no situation in which DRM is excusable. Maybe you will be able to access peer-to-peer networks to these songs and movies; I hope so. At least that won’t put chains on you, so it’s ethically legitimate.

Q: So, you don’t watch any movies on DVDs?
A: I have a few DVDs that are not encrypted and I don’t have anything that would play an encrypted DVD. Hollywood sets out to make crap and most people who see it already know that it’s crap before they go to watch it. It’s not quite the same as boycotting all movies. Boycott all movies that you don’t have a reason to feel that they’re good. And it’s obviously different from the simple boycott but the practical result is the same.

Q: Which movies, according to you, are “good” movies that you have watched?
A: My memory isn’t very good but I have seen movies that I feel like are not crap. I like Spike Lee movies and I also liked Galaxy Quest—it’s a comedy which makes fun of Star Wars and its fans, and turns into science fiction. It’s rather fun. I also like Spartacus.

Q: Which was the last movie you saw?
A: The last movie I saw was Spike Lee’s Inside Man and I saw it on an airplane to India, which is where I end up watching most movies.

Well he goes too far with “There is no situation in which DRM is excusable.” in fact I demand some kind of DRM on my bank account, you can call it privacy or encryption but it is in fact Digital Rights/Restriction Management. As he correctly points out in his nomenclature, success against DRM depends on the generic consumer rejecting it. In my philosophical terms, the evil is coming out of the DRM consumer. Without the gormless willingness of the consumer to accept the restrictions inherent in what they chose to give money for, an evil action of the rightsholder cannot bear fruit. But where my own data is being kept from me for a larger reason — so I cannot manipulate my own bank balance by hacking it — this is in a larger interest shared by everyone.

But still: RMS understands what ESR rejects, it is not enough that the code is Free, the data, the content has to be Free too. mplayer, bittorrented mainstream content, they are free but they are not truly Free, and they cannot be with the laws as they are. It is a shift as big as FOSS vs proprietary to move to a world where the data that you consume has the same Free rights as the code used to render it. How can the philosophical shift from proprietary to FOSS be played out on the content? jamendo.com shows the way but how can the content take advantage of the same aggregation advantages that code can?

As I described in my previous post, until now nobody seems to have thought through the ramifications of allowing a distributor to specify the license version he chooses to distribute under, or how he should specify what he has done in a non-conflicting way.

There must be a surefire way to use and distribute a GPL 2 “or later” package under GPL 2 rules. For example, so that a recipient cannot try to apply the GPL 3 “give me your crypto keys” rules when the distributor honestly and correctly wants to play only under GPL2 rules. But as I described in the previous post, merely noting that you distribute the sources under specifically GPL2 rules leaves the package you pass on in conflict with itself. The source files in the package are still annotated to allow “modification” under the terms of the GPL 2 “or later”.

What I proposed on the busybox list here and explained more clearly here was that a very strong way to make it clear how you are distributing would be to modify the GPL terms shown in the source files, ie, to show that you are distributing under GPL2 only as allowed to by the license, removing the “or later” part from the originally GPL2 “or later” sources. This has the advantage that the distribution method is consistently shown throughout the sources, and the action is ’sticky’, ie, if you were given those sources under GPL2 only then they remain GPL2 only through further distribution.

While the maintainer seemed to like the idea, there was a cognet objection from Glenn L McGrath, who says

GPLv2 clause 9 states "If the Program specifies a version number of
this License which applies to it and "any later version", you have the
option of following the terms and conditions either of that version or
of any later version published by the Free Software Foundation."

I dont see where it implies the right of licensees to remove the "or
later" statement. For all i know removing the "or later" clause may be
adding a further restriction to the license which isnt allowed according
to clause 6.

In addition to that concern, Clause 1 of GPL2 also seems to not like the idea as I noted in a reply

''1.  You may copy and distribute verbatim copies of the Program's
source code as you receive it, in any medium, provided that you
conspicuously and appropriately publish on each copy an appropriate
copyright notice and disclaimer of warranty;  keep intact all the
notices that refer to this License  and to the absence of any
warranty; and give any other recipients of the Program a copy of this
License along with the Program.''

Well something will have to give. What is needed is some definitive recipe to tell people who wish to distribute GPL2 “or later” projects under specifically GPL2 terms, as the license grants them the right to do so, such that unfair – unfair because they in good faith chose to play under GPL2 rules – GPL3 demands cannot be applied to such distributors.

Bruce Perens and the current busybox maintainer ended their argument for now by agreeing to both consult with the Software Freedom Law Center, a place that has found its moment to leap forward with advice I should think. So hopefully instead of IANAL there will shortly be an opinion from someone who IAL and who has a grasp of the confusing niceities of the text for a change.