Forgot your password?
typodupeerror
Android Open Source News

Does Android Violate the GPL? Not So Fast 186

Posted by Soulskill
from the sanity-check dept.
jfruhlinger writes "Patent gadfly Florian Mueller's latest post has made a fairly bold claim: that virtually all Android licensees are violating the GPL because of their failure to redistribute the code, and have thus lost their rights to redistribute Android. Mueller here is mostly promoting ideas put across by patent lawyer Edward J. Naughton. But others in the community are skeptical of the claims. Software Freedom Conservancy head Bradley Kuhn says he's never heard from Naughton. 'Don't you think if he was really worried about getting a GPL or LGPL violation resolved, he'd contact the guy in the world most known for doing GPL enforcement and see if I could help?'"
This discussion has been archived. No new comments can be posted.

Does Android Violate the GPL? Not So Fast

Comments Filter:
  • Troll is troll (Score:3, Informative)

    by Anonymous Coward on Monday August 15, 2011 @04:46PM (#37099160)

    Florian is a net-kook, not of course on the level of some others like JVM and such. Of course the decade is still young and he has plenty of time to improve his kook ranking

    • Re:Troll is troll (Score:5, Informative)

      by poetmatt (793785) on Monday August 15, 2011 @05:05PM (#37099444) Journal

      Why didn't anyone link to the http://www.groklaw.net/article.php?story=20110815131443415 (Groklaw Debunk) [groklaw.net].

      The writers are soapbox trolls, who have ties to microsoft but have hid them. And they write articles about how GPL violations are in force (which it is the owner's decision to enforce or not)? Say it ain't so!

      It is literally Florian quotes Naughton as his source, yet Naughton is basically a FUD machine.

      • by cHiphead (17854)

        As soon as I saw Florian Mueller I knew it was all utter bullshit. I consider him in a lower, more scumbag league than even half-assed MG Siegler on Techcrunch.

    • He? Seriously?

    • by exomondo (1725132)
      FM;DR
    • by e70838 (976799)
      The claims are stupid. Linus has clarified how and why the way Google has used kernel headers is fully authorised. But it is not a question of Linus opinion, the reasoning is easy to undestand:

      - Linux kernel can be used by non GPL programs. It is an indisputable goal of Linux.
      - To satisfy that goal, it was necessary to allow the usage of kernel interface description by non GPL programs. This authorization has been clearly given and do not restrict the manner this interface will be used.
      - The interface d
  • More Florian? (Score:5, Insightful)

    by painandgreed (692585) on Monday August 15, 2011 @04:46PM (#37099164)
    I'm beginning to remember the days of Dvorak articles as the happy times.
    • by MrHanky (141717)

      Just be happy that he's no Jon Katz.

    • by MrEricSir (398214)

      I liked Florian better when he was still in Kraftwerk. And Dvorak should stick to inventing new keyboard layouts.

  • WTF (Score:5, Insightful)

    by Anonymous Coward on Monday August 15, 2011 @04:46PM (#37099166)

    Why do you continually link to this sensational asshole astroturder?

    Does Slashdot get kick backs from his ad revenue?

    • by oakgrove (845019)
      I actually like the fact that florian gets linked on Slashdot. If we're lucky, when people google him, the Slashdot articles will be toward the top. When they click, maybe there is the slim chance they will be greeted with highly rated comments exposing Mueller for the fraud he is. Keep +1ing the truth, people!
    • Why do you continually link to this sensational asshole astroturder?

      Does Slashdot get kick backs from his ad revenue?

      You are a visionary, using a word that's not even in urban dictionary yet! http://www.urbandictionary.com/define.php?term=astroturder [urbandictionary.com]

      It will be later today, though if I can figure out what it means!

  • Even if he's right (Score:5, Insightful)

    by afidel (530433) on Monday August 15, 2011 @04:48PM (#37099188)
    Even if he's right, do we really want the GPL to be a revokable license where an tiny mistake that might throw you out of compliance requires a Herculean effort to re-establish rights? That would make all GPL code nuclear hot for any and all commercial interests which would probably see 80-90% of all code development on GPL projects dry up.
    • by s73v3r (963317)

      Even if he's right, do we really want the GPL to be a revokable license where an tiny mistake that might throw you out of compliance requires a Herculean effort to re-establish rights?

      I wouldn't call failing to distribute the source a "tiny mistake".

      • by poetmatt (793785)

        I wouldn't call completely invalidating compliance anything other than a "nuclear response", either.

        • Wait, we live in a world where a copied cd is a serious offence right? GPL violators must be protected because they are an economic force and screw music buyers which are a comparable one? no wai. Screw those that pollute GPL code with their crap and hide it, no matter what you think of the GPL.

          • by poetmatt (793785)

            What kind of idiot are you? This is the weirdest nonsequitur ever. I think you're up to one per sentence. Who said GPL violators must be protected? I sure didn't. Those who pollute GPL code? You mean like the mono project?

            There is ground between invalidating compliance and not taking action. The world is not "they're free to kill off GPL/must be sued into oblivion". If it was, people wouldn't even code.

      • by afidel (530433)
        What if just one branch for a new update doesn't get pushed out on the day of release, technically that's a violation which under a strict reading of the license could terminate your rights. I'm not sure where the line lies as far as compliance and regaining rights goes, but I think that actors who deal in good faith should not be punished disproportionaly to their transgression. In my mind Google's stance on Honeycomb was wrong, their partners don't have the right to withhold the source that they are distr
        • What if just one branch for a new update doesn't get pushed out on the day of release, technically that's a violation

          I don't think so. You can fulfil the source code requirements of the GPL by giving anyone who asks for it the source code. If nobody asks, you are fine. If someone asks, I don't think there is a definitive requirement how much time you would have.

          • It goes against general operating practice and tradition, but strictly speaking, the only parties with the RIGHT to demand the source are parties who are licensed users of the GPL-licensed application itself. In other words, if HTC sells you an Android phone, you have the inalienable right to demand the GPL'ed source code. If I buy a Samsung Android phone, I have no right to demand that *HTC* furnish ME with a copy of the source, because I'm not their customer, and they never licensed Android to me. HTC can

            • Actually, you can't demand anything. You can ask. If they ignore you, then the COPYRIGHT owners can go after them. But not the end users. The GPL provides no legal recourse for users.

            • Oh, and actually, the GPL v2 reads that you must fulfill the request from *any* third party, not just the people whom you distributed the binary to. Specifically, in these cases, the written offer words:

              " 2. b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above

              • by jeremyp (130771)

                Actually, that's not 2 b, that's 3 b and you don't need to do that, you can do 3 a instead which is

                Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange

                • My bad, I copied it word for word, and inserted the '2' instead of '3'.

                  It is rare, however, for anyone to use clause 3. Most use the written notice buried in the credits pages.

        • by JAlexoi (1085785)
          FYI: All the GPL code in Honeycomb is available. The ASL licensed code is not, however....
      • by WarlockD (623872)
        A tiny mistake is "Hey sorry, we thought we posted all the changes. We will post them now" A serious one is, "Sorry, we put some of our proprietary code in there so we are not going to show it in like ever" I believe it needs to be revokable, but GPL has always been kind of murky on what enforcement will be about it. As a newer poster said, if you have a problem with GPL use MIT or something else.
      • Re: (Score:2, Informative)

        by multi io (640409)

        I wouldn't call failing to distribute the source a "tiny mistake".

        http://android.git.kernel.org/ [kernel.org]

        Looks distributed enough to me.

        And I know that's only the kernel -- but that's all that matters. The userspace stuff isn't "derived" and thus doesn't have to be licensed under the GPL.

        And I also know that the above is a Google-provided site, not one provided by the handset manufacturer. But that doesn't matter either as long as the handset manufacturer doesn't himself modify the GPLed parts of Android (i.e. the kernel) and then fails to distribute the source for that. As

        • by Wovel (964431)

          I believe you are correct. However, It is improper to refer to Android (At least Honeycomb) as an Open Source Operating System. It uses an Open Source Kernel, I am not aware of any other components of the OS that have been released. This is a change from all previous versions of the Android OS.

          • by poetmatt (793785)

            Was it ever said that they *won't* release it? Or just that they haven't *yet*? I don't mean this wholly as a counterargument, more of I'd like your input on that so I can understand better.

        • by jeremyp (130771)

          That's not good enough. The terms of the GPL v 2 section 3 make it explicit that the distributor (mobile phone manufacturer) has to make the source code available. They can't just say "get it from that other web site over there". So to be compliant, HTC (for example) needs to make available the source for the exact version of the kernel that is installed on your phone.

          • That seems like a strange argument. Of course you're right that the exact sources required for building the HTC code need to be available; the GP claims that the Android resellers don't modify kernel source, which may or may not be true. If that's true, however, I don't see why it would matter who hosts the code.

            For all we know Google has an agreement with the resellers that they will host the sources for the vanilla kernel. I'm sure HTC could make an agreement with another party -- say, github, or sourcefo

        • Nearly every handset manufacturer has to modify the kernel just to get Android ported to their phone.

          That said, I have never in my life seen a single phone model that doesn't have it's kernel released. I don't know what Florian is going on about, but he has obviously never browsed xda-developers.com. Quite often the source for the kernel is released, re-built, and ported to custom roms *before the manufacturer has even released a ROM to the carriers.

      • That's for the copyright owners to decide. That's the point.

      • by rtb61 (674572)

        It is, if your contributions to the code are nothing more than re-branding ie making it seem like it is your application in the retail space for marketing purposes, basically supplied for free with the hardware. In turn providing some direct sponsored contributions to the core code, some improvements you wanted in your retail branded version but you still didn't want to maintain and distribute the source, security etc.

        It seems pretty un-necessary to distribute 'your' source if it is pretty much identical

      • by firewood (41230)

        I wouldn't call failing to distribute the source a "tiny mistake".

        If your server goes down or gets unplugged or rebooted for any reason, you are temporarily failing to distribute source. How long before that's not "tiny"?

    • Well, we do want it to be a revocable license. We don't want it to be revoked for piddly shit like a line here and a couple of code comments there accidentally making it into an Apache-licensed module rather than a GPL-licensed module. It needs to be revocable for intentional violations that are not being corrected when brought to light. Otherwise, if it could not be revoked, then people would have the right to continue to violate it and continue shipping code in violation of it. What should be necessary t

    • by JAlexoi (1085785)
      GPL is revokeable. However, is it is also automatically issued. By returning into compliance, you regain the license automatically...
    • by Wovel (964431)

      Seriously? So you are advocating not enforcing the GPL? You want companies to profit without contribution off the hard work of those who contribute to GPL projects. Perhaps you are saying we should reinstate their license if they provide a simple apology.

      I am not a big fan of the GPL for a lot of reasons, but not enforcing it or weakening enforcement seems rather pointless.

    • A separate license can be granted by the original author on the code base they maintain. This is fairly commonly done actually, and many projects use specifically this tactic to sell what is more or less the right to distribute something closed source. Off the top of my head ARTag is an excellent example.

    • by Jonner (189691)

      Even if he's right, do we really want the GPL to be a revokable license where an tiny mistake that might throw you out of compliance requires a Herculean effort to re-establish rights? That would make all GPL code nuclear hot for any and all commercial interests which would probably see 80-90% of all code development on GPL projects dry up.

      That's just one of the ways GPL v3 [gnu.org] improves on earlier versions. Of course many Linus and other Linux contributors remain irrationally opposed to it. From the GPL v3

      However, if you cease all violation of this License, then your
      license from a particular copyright holder is reinstated (a)
      provisionally, unless and until the copyright holder explicitly and
      finally terminates your license, and (b) permanently, if the copyright
      holder fails to notify you of the violation by some reasonable means
      prior to 60 days after the cessation.

  • by jeffmeden (135043) on Monday August 15, 2011 @04:55PM (#37099280) Homepage Journal

    Patent gadfly Florian Mueller's latest post has made a fairly bold claim: that virtually all Android licensees are violating the GPL because of their failure to redistribute the code,

    Hmm, no, actually most (but certainly not 100% of) manufacturers that embed Android comply with the redistribution clause in the GPL. Samsung (one of the biggest vendors) has a web site set up specifically to redistribute code, and others make it similarly easy. Do you really think that the multi billion dollar likes of Samsung, Motorola, HTC, etc didn't bother having a copyright lawyer look over the situation to make sure things are kosher? Surely there are some vendors out there that are abusing the system and not putting up code as required, but if any one of the major vendors did that why not come out and say who it was?

    • by Andy Dodd (701) <atd7NO@SPAMcornell.edu> on Monday August 15, 2011 @04:58PM (#37099324) Homepage

      In addition, Samsung is extra careful to put up code BEFORE a device is released in most if not all situations.

      The sources for the Rogers variant of the Infuse 4G (SGH-I997R) dropped the Monday before anyone had the device. (Devices showed up two days later on Wednesday.) As I understand it, it was the same with the initial Infuse release on AT&T.

      • So where's the Xoom's source?

        • by exomondo (1725132)

          So where's the Xoom's source?

          Probably on Motorola's sourceforge page.

        • by JAlexoi (1085785)
          nVidia Kernel is right there on android.git.kernel.org
          • Yeah but where's the WebKit based browser's source?

            • by gl4ss (559668)

              the arguable point is indeed that not all of android is gpl and how the non-gpl parts tie up to the gpl'd parts. And well, sitting on honeycomb while devices were shipping kind of proved that, not a big issue for any 3rd party devs surely, but indeed a big issue for any random chinese tablet manufacturer.

              in practice, it's not just having source that matters - it's having the source that's actually in the shipped devices and/or possibility to change the code on that running device. moto's had a few linux dev

              • That's the TiVoization aspect to GPLv3.

                I think it's despicable that Google ran around, and especially Andy Rubin, ran around touting "It's free!" when you now can't build from source AND Google's looking to sue MS for showing an outside consultant 3.x source.

  • by Baloroth (2370816) on Monday August 15, 2011 @05:01PM (#37099378)

    The GPL parts of the code for Android are freely available. Google provides them, and I'd assume any licensee would just point to that if asked for the code (if they don't already make it freely available themselves. Can a redistributor just point to the original source under the GPL if they don't modify it? I assume they can). AFAIK most licensees don't modify the GPL portions of the code, only the front-end etc. which are licensed under Apache. I'm no expert on the GPL, but really this is just FUD created by (as others have commented) Florian Mueller, who just seems to glance at issues and post whatever he "thinks", without actually doing, well, any real thinking, much less actual research.

    Now, if RMS said this, I might stand up and take notice. Doubt he would though, he knows well enough that poisoning GPL code like that would mean the death of OSS.

    • by Microlith (54737)

      . Google provides them, and I'd assume any licensee would just point to that if asked for the code (if they don't already make it freely available themselves. Can a redistributor just point to the original source under the GPL if they don't modify it? I assume they can). AFAIK most licensees don't modify the GPL portions of the code, only the front-end etc. which are licensed under Apache.

      Virtually every Android device provider touches the kernel, as they have to integrate their board support files and hard

      • by jonwil (467024)

        The continued GPL violations by HTC are the #1 reason why I didnt buy a Desire Z and bought a Nokia N900 instead (the locked firmware/bootloader was #2)

        • by dudpixel (1429789)

          The continued GPL violations by HTC are the #1 reason why I didnt buy a Desire Z and bought a Nokia N900 instead (the locked firmware/bootloader was #2)

          cant find what you want at http://www.htcdev.com/ [htcdev.com] ?

          Or maybe you hadn't looked?

          • by jonwil (467024)

            Yes HTC released GPL source code for the Desire Z eventually.

            But some searching with Google and some reading of the archives for the gpl-violations mailing list would show you that they routinely take weeks or even months after the release of the product before they release the GPL source and that in some cases the released source code does not match the binaries.

    • by Chirs (87576)

      Can a redistributor just point to the original source under the GPL if they don't modify it? I assume they can

      You assume wrong, if the redistributor is doing it commercially. In that case they need to either include the source with the binary or else provide a written offer to give the source to anyone that asks.

      For noncommercial distribution, providing a link to the original source is sufficient.

      • by lkcl (517947)

        Can a redistributor just point to the original source under the GPL if they don't modify it? I assume they can

        You assume wrong, if the redistributor is doing it commercially. In that case they need to either include the source with the binary or else provide a written offer to give the source to anyone that asks.

        For noncommercial distribution, providing a link to the original source is sufficient.

        that's incorrect. commercial or non-commercial has nothing to do with it. re-read the GPL license. the only commercialisation allowed by the GPL is that it's ok to offer a warranty (and charge for it) or to charge for the distribution (sending a CD). that's all. there's absolutely no mention of differences between commercial or non-commercial redistributors.

        correct answer: yes, a redistributor can just point to the original source if they distribute binaries which were compiled without modifications to

        • In the GPL v2, section 3 reads as follows...note in particular clause C.

          3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:

          a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,

        • by jeremyp (130771)

          the only commercialisation allowed by the GPL is that it's ok to offer a warranty (and charge for it) or to charge for the distribution (sending a CD). that's all.

          That's for the source code, you can charge as much as you like for the compiled binary.

          there's absolutely no mention of differences between commercial or non-commercial redistributors.

          correct answer: yes, a redistributor can just point to the original source if they distribute binaries which were compiled without modifications to that original GPL'd source code. it's actually a very good way for the redistributor to keep costs down.

          As others have said, you need to read section 3 which deals with distribution of binaries. Suffice it to say, you are wrong.

    • unfortunately your assumptions are incorrect. many of the chinese tablet manufacturers are ignorant of software: they have to get a specialist in, to help them, as they are hardware-manufacturers only, in most cases. so they don't even HAVE the source code! i've repeatedly found this to be the case when dealing with chinese factories. there are many other things that are incorrect about what you've said, and i've cover them here, as well as providing some insight into what exactly is going on: http://ne [slashdot.org]

    • by jeremyp (130771)

      Can a redistributor just point to the original source under the GPL if they don't modify it? I assume they can)

      No. They can't.

      Section 3 of the GPL v2 says they must either provide the source when they provide the binary or provide a written offer to give people the source if asked at the cost of distribution or, for non commercial distribution only point to the written offer they received from upstream.

  • He should have waited for Ice Cream Sandwich before going public with his claims. Now he risks Android being killed before its most crucial update.

  • by lkcl (517947) <lkcl@lkcl.net> on Monday August 15, 2011 @05:50PM (#37100038) Homepage

    there is a lot of misinformation about this topic, because many people do not realise that chinese manufacturers simply have absolutely no software skills - at all - and get supplied with GPL-violating binary firmware from a very limited pool of software specialist companies in china, completely by mistake. they then turn to those software specialists when asked for GPL compliance, and you typically get a very distorted and standard answer which indicates a complete lack of knowledge of the GPL.

    but on top of that, the android source code that comes out of google simply is not enough, on its own. firstly: android is NOT just apache2-licensed applications: it sits on top of a GPL'd Linux Kernel which has had some very specific modifications made to it (mostly in the form of the android "security model"). secondly: many CPU manufacturers have to add hardware-accelerated video and 3D graphics engines in order to meet "cunsumaah dimarnd". these are backed up by proprietary software libraries that qualify - usually - as "System Libraries" under GPL exemption clauses. thirdly: android simply doesn't have a built-in video player nor a video player "API" so it is up to the vendors to put in applications which *do not* come from the android "stock" that comes out of google, and they usually do this by grabbing the nearest GPL source code they can find. fourthly: many software-builders for the OEMs / ODMs simply throw away portions of android source code and utilise the GPL equivalents (such as the GPL version of busybox, not the version that google implemented from scratch just to be able to "cleanse" the core android OS From All Gee Pee Ell code).

    so this is the situation. and, as a result, yes, the vast majority of android devices in the world are GPL violating. let's go through a few coments that have made it past slashdot moderation.

    baloroth states that google provides the GPL parts of android and that they are "freely available". well, yeah, but re-read the above and you'll see that that's completely irrelevant. he then states that he assumes that any licensee would just point to that if asked for the code. well, yeah, but re-read the above and you'll see that that's again completely irrelevant - not least because the licensee is required to provide the code that THEY have distributed, and it's usually been heavily modified by somebody else that they got in to do the software. he then states that "most licensees don't modify the GPL portions" which is wrong: it is absolutely essential to create an android-specific linux kernel which will support the android OS, on a per-CPU and even a per-device basis.

    jeffmeden then goes on to try to state that we should all think that multi-billion-dollar companies like motorola, samsung and HTC don't bother to check if things are kosher? jeff - even a few seconds of checking on the gpl-violations mailing list or even just searching "HTC GPL Violation" would show you at least three GPL violations by HTC within the past year! samsung you're actually right about, and motorola i haven't kept up with but they are just in the process of being acquired by google, which is something that needs to be kept an eye on. it could be good, or it could be bad.

    jeff also states "hmm no actually most manufacturers comply with the redistribution clause in the GPL". this is completely wrong. actually, according to an off-the-cuff survey of android tablets done six months ago by a redhat employee, he found that 95% of the 80+ tablets were GPL violating. he's maintaining the list here:
    http://www.codon.org.uk/~mjg59/android_tablets/ [codon.org.uk]

    this list is so long it would overwhelm the Software Freedom Law Centre's resources to tackle them all at once.

    so... yeah. it would appear that there is a hell of a lot of ignorance surrounding android. the mistake that google made was to try to combine apache-licensed code with GPL code. apart from anything, this gives people the impression that all

    • The Software Freedom Law Center cannot do *anything* without an owner of some of the code requesting they do so. Which, no one has, as of yet. The article is rather correct in this manner. There are moocho, many companies which have not complied. But it doesn't make a lick of difference until the people with the rights, THE OWNERS, decide to do something about it.

    • by poetmatt (793785)

      wtf is this garbage? Could you have posted more fud and crossposted it elsewhere?

      Where have there been actual GPL violations enforced? This matters, because it means it's not a big enough deal for a developer to raise the "violation flag". Show me one. So yeah, it would sum up that your article isn't really accurate in that sense.

    • by dudpixel (1429789)

      Why is Google's source code release irrelevant?

      Android is Google's product, and no one else's, and Google does comply with the GPL. Therefore the headline is incorrect. Android doesn't violate the GPL. Not at all.

      Maybe some obscure chinese manufacturers that the rest of the world have never heard of are violating the GPL, but certainly all the big-name brands (HTC, samsung) are not violating the GPL.

      None of this has anything to do with Android. It is select Android Licensees, and not Android itself, tha

  • For ages the GPL community has been rightfully baffled by the problem of defining what constitutes a derivative work that triggers the GPL. Copyright law itself isn't entirely clear on the matter, nor is making a binary distinction (derivative/not) particularly easily when there is a continuum of derivative-ness. FM seems to be absolutely certain that Android is a derivative work of the Linux Kernel. I'm not so sure, given that it runs entirely in userspace and access the kernel through well-defined interfa

    • by lkcl (517947)

      again, you're completely ignorant of the situation surrounding the actual makeup of the android source code. the core android OS which was released BY GOOGLE is completely apache2 source code. it runs on top of a modified Linux Kernel, which has had patches made to it to add in, apart from anything else, the android security model. for more info see the post i made here: http://news.slashdot.org/comments.pl?sid=2380756&op=Reply&threshold=1&commentsort=0&mode=thread&pid=37100038 [slashdot.org]

      bob, yo

      • again, you're completely ignorant of the situation surrounding the actual makeup of the android source code. the core android OS which was released BY GOOGLE is completely apache2 source code. it runs on top of a modified Linux Kernel, which has had patches made to it to add in, apart from anything else, the android security model. for more info see the post i made here: http://news.slashdot.org/comments.pl?sid=2380756&op=Reply&threshold=1&commentsort=0&mode=thread&pid=37100038 [slashdot.org] [slashdot.org]

        bob, you really really need to get better informed about the GPL before making random comments like the ones you've just made, ok?

        You will not that they haven't released the source for Android 3.X under any license, Apache or otherwise. It does run on top of a modified Linux kernel, the patches to which have been published. The question is whether they can distribute Android 3.X that runs on top of Linux without it being a derivative work.

        And I think I understand the GPL just about as well as it can be understood given the ambiguity and lack of caselaw regarding derivative works.

        See, e.g. http://www.linuxjournal.com/article/6366 [linuxjournal.com]

        But what happens if you merely copy an original program as a component in your own, perhaps larger, work? Does it make a difference that you didn't actually modify the source code to combine the original program into your larger work?

        Does merely linking to a program without any change to the original source code create a derivative work of that program? Almost every program links to library routines. Surely, one doesn't create a derivative work of a library simply by calling a sqrt function in the library. Why should it be any different when you link to something as complex as an enterprise server or database engine? What about linking from a software program, such as when linking your device driver into a GPL- or OSL-licensed program like Linux?

        ~Larr

        • Actually it is rather clear in US courts, there isn't a single court case I'm aware of that has ruled a program as a derivative work of a library it uses, there are plenty of them that have struck that down, however.

          Does it make a difference that you didn't actually modify the source code to combine the original program into your larger work?

          In copyright law, no, because the copied code might not be licensed for such redistribution, so long as that work can actually be copyrighted and is not strictly necessary as an integral part of the work (i.e. code for the purpose of stopping inter-operability instead of as a creative literary

          • If you are allowed to distribute the linked library with the program is another matter (the GPL prohibits it, the LGPL allows it).

            So distributing the Linux Kernel alongside any non-GPL-compatibly-licensed, such as Ubuntu One, which is closed source, is prohibited? Obviously Ubuntu One links to the kernel and uses the various userland kernel APIs to do its work, it could not operate without the kernel at all.

            Even establishing all that, I don't think that makes sense to call it a derivative work. As I see it, a CD with Ubuntu and Ubuntu One on it is not a derivative work of the kernel or Ubuntu One but they remain as two separate works

  • Idiots quote Florian and idiots post links to his nonsense. He is as close to an expert on patent law as any character on Sesame Street. Any journalist that refers to him seriously is even more of an idiot.
  • The title of this article of absolutely Android FUD. GPL conformance by vendors has long been a thorn to end users, and Android is no exception. But this gives the inference that Android itself somehow violates the GPL, which is utter baloney.

How many QA engineers does it take to screw in a lightbulb? 3: 1 to screw it in and 2 to say "I told you so" when it doesn't work.

Working...