Kuhn: Clarification on Android, its (Lack of) Copyleft-ness, and GPL Enforcement

Posted May 21, 2011 15:20 UTC (Sat) by FlorianMueller (guest, #32048)
In reply to: Kuhn: Clarification on Android, its (Lack of) Copyleft-ness, and GPL Enforcement by JoeF
Parent article: Kuhn: Clarification on Android, its (Lack of) Copyleft-ness, and GPL Enforcement

I do read LWN, but not each article, and not on a daily basis. I don't see how there's any "FUD" of mine "exposed" here. I saw on Naughton article in the HuffPo and felt that it raised a serious issue, so I blogged about it. If he now writes other articles or posts comments to sites like ITWorld, I may or may not agree in each case. If he again does something I consider very important, I may blog about it again, just like if anybody else writes something I believe my readers should know about.

Apart from that, the only way anyone could "expose FUD" is to tell people they have legal certainty. Brad Kuhn and even Linus Torvalds stopped short of that. Brad Kuhn still stops short of that and keeps a door open for the possibility that Naughton was right.

I recently saw a video (can't find the URL right away) of an assistant law professor who's totally against all software patents (not just against some of them, which is Google's position) and all for open source, and she said in her presentation in front of an audience in Red Hat's backyard that "The GPL is not what Richard Stallman tells you it is. It's not what Eben Moglen tells you it is. It's what judges will say it is." (You could substitute Brad Kuhn or Linus Torvalds for the two names in the previous quote.) I still don't see that there's certainty that judges will interpret the GPL the way Brad Kuhn, Linus Torvalds and other people with all their agendas and their interest in playing favorites with some companies would like them to.

Kuhn: Clarification on Android, its (Lack of) Copyleft-ness, and GPL Enforcement

Posted May 21, 2011 15:46 UTC (Sat) by JoeF (guest, #4486) [Link] (2 responses)

Yup, FUD again, this time the old, tired, and wrong "the GPL has not been tested in court" FUD...
According to your "logic", click-through licenses are invalid because they, unlike the GPL, haven't been tested in court.

Kuhn: Clarification on Android, its (Lack of) Copyleft-ness, and GPL Enforcement

Posted May 21, 2011 16:04 UTC (Sat) by FlorianMueller (guest, #32048) [Link] (1 responses)

@Joef, I don't know why you don't understand me. Maybe you can't; maybe you don't want to. At any rate, something untested is not automatically invalid. I never said that or implied that; I never would say or imply that. Such hyperbole is not a contribution to a discussion that you could be proud of.

Compared to its ubiquity, the GPL is -- so far -- rather untested. I believe that to a certain extent that's due to the fact that the FSF and others know that especially GPLv2 is irretrievably ambiguous in some key areas, so they try whatever they can to prevent further clarification.

Every somewhat independent legal expert who has so far analyzed the GPL has concurred that it's irretrievably ambiguous in certain respects, especially in connection with the copyleft borderline dispute. But some "luminaries" try to use strong words to distract from that irretrievable ambiguity. However, no matter how aggressively they speak out, they don't change the law.

Also, the current situation is quite a nice one for the likes of Eben Moglen and Brad Kuhn to be in. The lack of case law enables a few people close to RMS to run a GPL interpretation cartel. Some of them do that for financial gain; others are idealists; some represent a combination of both. When anyone from outside of that cartel, such as Naughton in his article of two months ago and apparently a more recent one, dares to discuss possible interpretations that may not be entirely consistent with the party line, they get very nervous. Brad Kuhn's post displays a lot of fear of "competition" that the cartel to which he belongs could increasingly face.

Finally I'd like to add something to why I didn't pay attention to this article here before. These days my focus has been, besides Oracle vs. Google (I commented on Oracle's and Google's joint memorandum on Wednesday within an hour and a half of its discoverability on PACER, while Groklaw commented two days later; in terms of content, Groklaw's explanations and mine are similar), on the terrible problem posed by trolls attacking defenseless little iOS app developers (see this EFF statement, MocoNews/paidContent.org, or my FAQ). Obviously, free software folks don't care about iOS app developers, although RMS repeatedly said that closed source developers should support the fight against software patents). So I'm more focused on helping programmers understand and evaluate this situation than on the sensitivities of the GPL interpretation cartel.

Kuhn: Clarification on Android, its (Lack of) Copyleft-ness, and GPL Enforcement

Posted May 21, 2011 20:12 UTC (Sat) by Zack (guest, #37335) [Link]

>>The lack of case law enables a few people close to RMS to run a GPL interpretation cartel. Some of them do that for financial gain;

Now this is getting rich. There's one voice I consistently hear when it comes to someone trying to profile himself as an open-source-savy independent "intellectual property" consultant for hire.

GPL not legally vetted?

Posted May 21, 2011 15:57 UTC (Sat) by vonbrand (subscriber, #4458) [Link] (7 responses)

While there are details of GPL that aren't tested in court of law, its meaning is clear enough that controversies around it haven't reached the courts. Most are silently settled. And what RMS and Linus think the license means has some legal relevance, even while not completely binding. And it can't be said that theirs is just a random layman's interpretation, RMS and the FSF had first rate legal advice on crafting the license, Linus did likewise consult competent lawyers on the issue. And the folks at Red Hat (whose livelyhood depends on assorted licenses) and the people from IBM (who invested heavily in Linux) did too.

GPL not legally vetted?

Posted May 21, 2011 16:10 UTC (Sat) by FlorianMueller (guest, #32048) [Link] (6 responses)

While there are details of GPL that aren't tested in court of law, its meaning is clear enough

You might want to read Heather Meeker's book "The Open Source Alternative", particularly the chapter on the GPL Borderline Dispute (concerning copyleft). The GPL is irretrievably ambiguous in that area (and in others, too).

And what RMS and Linus think the license means has some legal relevance, even while not completely binding.

What they think has relevance to a court's decision only as what may be regarded as persuasive authority, but persuasive authority ranks relatively low among all of what goes into a proper construction of contract terms.

And the folks at Red Hat (whose livelyhood depends on assorted licenses)

That's a non sequitur, at least without a convincing explanation. Red Hat itself supports weak copyleft interpretations (closed-source nVIDIA drivers etc.), but apart from that, I can't see how Red Hat's business model depends much on the copyleft borderline issue. They could always adjust their business model to court decisions (should they ever have to).

the people from IBM (who invested heavily in Linux) did too

IBM also does a lot of business based on permissive licenses (particularly Apache). So again, I don't see the business logic of what you're trying to say.

GPL not legally vetted?

Posted May 21, 2011 18:19 UTC (Sat) by clump (subscriber, #27801) [Link] (5 responses)

You've made a very disappointing selective quote. I'll provide it in it's entirety:

While there are details of GPL that aren't tested in court of law, its meaning is clear enough that controversies around it haven't reached the courts. Most are silently settled.

Now here's something you've said with regard to Nvidia:

Red Hat itself supports weak copyleft interpretations (closed-source nVIDIA drivers etc.)...

Perhaps you missed that Red Hat sponsors and distributes only the open source nouveau driver? Are you intentionally being dishonest?

GPL not legally vetted?

Posted May 21, 2011 18:28 UTC (Sat) by FlorianMueller (guest, #32048) [Link] (4 responses)

If anything here is "very disappointing", it's the quality of your comment I'm replying to.

Your entire quote concerning "tested in court of law" doesn't change anything about the accuracy or relevance of what I said. I stopped at "clear enough" because that's the point I commented on. It was a legitimate way to stay focused. It was not like taking anything out of context.

Concerning nVIDIA, what difference does Red Hat's sponsorship and distribution of "only the open source nouveau driver" make? The question is whether Red Hat believes that closed-source Linux drivers are acceptable from a GPL point of view. Red Hat's position (and that of its EU evangelist taken in a Twittersation with me) is that it's OK to do such drivers as long as they're distributed separately. Of course, since their position is that the GPL can be circumvented by distributing such code separately (I don't think so, but they apparently do), they can't distribute it themselves along with GPL'd code.

But the thing they would do, if they really wanted strong copyleft, is to sue nVIDIA for release of its source code, claiming it's a derivative work.

GPL not legally vetted?

Posted May 21, 2011 18:57 UTC (Sat) by clump (subscriber, #27801) [Link]

Your entire quote concerning "tested in court of law" doesn't change anything about the accuracy or relevance of what I said. I stopped at "clear enough" because that's the point I commented on. It was a legitimate way to stay focused. It was not like taking anything out of context.

Except that you selectively quoted, and answered out of context. This is unfortunate behaviour of yours.

Regarding Nvidia, you made a dishonest and uninformed argument. Worse, you're attempting to change the topic again. Please, take a moment to understand the nouveau project.

I suppose I'm flattered by your disappointment.

GPL not legally vetted?

Posted May 22, 2011 5:10 UTC (Sun) by cmccabe (guest, #60281) [Link] (2 responses)

> But the thing they would do, if they really wanted strong copyleft, is to
> sue nVIDIA for release of its source code, claiming it's a derivative work

So not suing over something automatically means that you "believe that behavior is acceptable"? Since you're the "expert," can I quote you on that?

GPL not legally vetted?

Posted May 22, 2011 6:05 UTC (Sun) by FlorianMueller (guest, #32048) [Link] (1 responses)

Note that I said all of this against the background of vonbrand claiming that Red Hat's livelihood depends on this. If it livelihood depended on strong copyleft for Linux, they would definitely fight to have closed source drivers by such a major player declared illegal. They might not care about little or subtle violations, but again, if (as vonbrand claims) this was about their livelihood, they would take action.

GPL not legally vetted?

Posted May 22, 2011 15:01 UTC (Sun) by clump (subscriber, #27801) [Link]

There's a certain irony in your words here. Vonbrand didn't claim Red Hat's livelihood depends on Red Hat suing Nvidia. Your entire position is that litigation is evidence that you support, or do not support, positions. Surely your "Twittersation" -- your words -- clears all of this up.

I keep mentioning the Nouveau project because it's a concrete position taken by Red Hat with regard to Nvidia drivers. Yet you mentioned Nvidia in an attempt to spread misinformation. You've been caught.

Kuhn: Clarification on Android, its (Lack of) Copyleft-ness, and GPL Enforcement

Posted May 22, 2011 13:29 UTC (Sun) by AndreE (guest, #60148) [Link] (13 responses)

It's a disingenuous to say that "the GPL hasn't been tested in court" :

1) Some aspects HAVE been tested in court. For example, both the netfilter and JMRI projects had certain clauses of the GPL upheld by courts in their respective jurisdictions. You have to be specific in naming which specific parts of the GPL are problematic and in which jurisdictions.

2) Most licenses haven't been tested in court. The formulation of a license is certainly no test of it's validity, but that is true for ALL licenses, not just the GPL. I can make all sorts of claims that the Adobe license for Photoshop 5 or that the Windows 7 license haven't been tested in court.

AndreE, your reading comprehension is your problem, not mine

Posted May 22, 2011 13:33 UTC (Sun) by FlorianMueller (guest, #32048) [Link] (12 responses)

Before you talk about disingenuity you might want to check on the facts. You were replying to a post of mine. I never said that the GPL hasn't been tested in court. I'm well aware of the fact that some enforcement has happened (particularly by Harald Welte). I had said "rather untested" and that certain aspects including the copyleft borderline dispute haven't been tested.

So @AndreE, don't claim I said something I never said. Maybe LWN should require people to take a basic reading comprehension test before they're allowed to comment here.

AndreE, your reading comprehension is your problem, not mine

Posted May 22, 2011 14:05 UTC (Sun) by AndreE (guest, #60148) [Link] (6 responses)

Firstly, please point out where in your post you talked about the copyleft borderline dispute. I only see mentioned a video of a law professor and the uncertainty surrounding GPL interpretation.

Regardless, my second point still stands. Uncertainty is a part of ALL licenses. Proprietary, copyleft, whatever. All licenses have certain aspects untested, so it is not particularly interesting in the case of the GPL that some elements are untested.

Harping on about the uncertainty of GPL interpretation is a tactic generally adopted by competitors of open source products seeking to gain some form advantage.

AndreE, you have to prove your (fake) quote

Posted May 22, 2011 14:08 UTC (Sun) by FlorianMueller (guest, #32048) [Link] (5 responses)

You're making things worse this way. You claimed that I claimed "the GPL hasn't been tested in court". But I never said or implied that. It's not me who has to prove anything here. You came up with a TOTALLY FAKE QUOTE and called it "disingenuous". Don't manufacture fake quotes. I don't have to prove it's a fake quote. You have to prove that I said that.

AndreE, you have to prove your (fake) quote

Posted May 22, 2011 14:14 UTC (Sun) by AndreE (guest, #60148) [Link] (4 responses)

Yes, now I see the reply below. You didn't say that the GPL was untested. I apologize for that.

Are you suggesting that being "relatively untested" is problematic for the GPL?

AndreE, you have to prove your (fake) quote

Posted May 22, 2011 14:15 UTC (Sun) by AndreE (guest, #60148) [Link]

'rather untested'

Thanks for clarifying this

Posted May 22, 2011 14:19 UTC (Sun) by FlorianMueller (guest, #32048) [Link] (2 responses)

I appreciate your clarification. I was really upset because it's hard to have any reasonable discussion on these issues if there's a need to deal with misattributions.

To answer your question about whether "rather untested" is problematic for the GPL, I don't think it's a reason not to use the GPL or not to use GPL'd program code. It is, however, a reason not to place blind faith in persuasive authority (i.e., luminaries like Torvalds) in connection with untested aspects of the GPL, especially the copyleft borderline dispute.

Thanks for clarifying this

Posted May 22, 2011 14:52 UTC (Sun) by AndreE (guest, #60148) [Link] (1 responses)

Right, I read the whole thread (it's late here). I agree about your point about persuasive authority, but I don't really find it a compelling revelation about the uncertain aspects of the GPL, for the reason I stated above.

However, I realise you were responding to someone's assertions, and not actually highlighting it as your own particular point of issue with the GPL.

Perhaps I should indeed take a comprehension test...

Thanks for clarifying this

Posted May 22, 2011 14:56 UTC (Sun) by FlorianMueller (guest, #32048) [Link]

By now it's clear what I said and what I didn't say. I for my part don't have any issue with the GPL. In fact, MySQL's GPL-based business model worked out very well and I (re-)stated my pro-GPL views last August.

AndreE, your reading comprehension is your problem, not mine

Posted May 23, 2011 17:00 UTC (Mon) by pboddie (guest, #50784) [Link] (4 responses)

I understood the "fake" quote to be paraphrasing what you wrote, which was the following:

I still don't see that there's certainty that judges will interpret the GPL the way Brad Kuhn, Linus Torvalds and other people with all their agendas and their interest in playing favorites with some companies would like them to.

That is, the usual finger-wagging "wait till a judge takes a look at it" style of response. Note that I'm not directly quoting you here but merely characterising a sentiment that is frequently expressed.

As always, I find it helpful to think of the GPL as a big "view source" button: if a program is licensed under the GPL, you should be able to invoke that option even if it isn't actually achieved by pressing an actual on-screen button. How deep the viewable source goes depends on limitations of what people (yes, judges and legislators included) regard as reasonable, but you can't blame the people who write these licences from telling people what they had in mind.

People keep telling us that a common understanding of the intent of a licence does matter when a judge has to decide on a case. You're suggesting that some of the people who contribute to this common understanding are actually holding back in favour of various corporate interests, are you not?

AndreE, your reading comprehension is your problem, not mine

Posted May 23, 2011 17:05 UTC (Mon) by FlorianMueller (guest, #32048) [Link] (3 responses)

Common business practice is also a relatively weak factor (as is persuasive authority). A judge may come down on that side -- or may not.

I'm not saying that holding back is a matter of what some large corporations might want them to do. At least I don't see any particular indication for that theory. However, the current state of affairs in the interest of entities like the FSF and the SFLC.

AndreE, your reading comprehension is your problem, not mine

Posted May 23, 2011 22:04 UTC (Mon) by pboddie (guest, #50784) [Link] (2 responses)

So what do you mean by "their interest in playing favorites with some companies"?

AndreE, your reading comprehension is your problem, not mine

Posted May 23, 2011 22:06 UTC (Mon) by FlorianMueller (guest, #32048) [Link] (1 responses)

That didn't relate to their hesistance to have the GPL clarified by courts but to their inclination to issue statements like "I'm not 100% sure, and I don't know the details, but everything's fine" in connection with what companies like Google and Red Hat do.

AndreE, your reading comprehension is your problem, not mine

Posted May 24, 2011 11:22 UTC (Tue) by pboddie (guest, #50784) [Link]

And have there been cases where other parties have done practically the same thing and those people have condemned it?

Kuhn: Clarification on Android, its (Lack of) Copyleft-ness, and GPL Enforcement

Posted May 24, 2011 6:49 UTC (Tue) by cmccabe (guest, #60281) [Link]

You could say equivalent things about any legal document. "The American constitution is not what Thomas Jefferson said it was. It's what judges will say it is." Statements like this are empty of content to those of us who have even even a high-school level grasp of the US legal system.

However, judges will consider the intentions of the people who drafted the legal document when deciding how to interpret it. For example, if you write a will stating "I leave this brick house to my daughter," and the house is later extended so that part of it is made of wood rather than bricks, the judge will most likely give that part to your daughter as well. That was obviously your intention.

If you want to discuss ambiguities in the GPL, any version, let's discuss it. There is a lot to discuss and I've had many interesting conversations about the subject. If you want to make vague, content-free negative statements about Bradley Kuhn, Google, Red Hat, or various other individuals and organizations, maybe you should post somewhere else.