Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
GNU is Not Unix

MySQL AB and Nusphere Go to Court Over GPL 238

Posted by michael from the fate-of-the-Free-world-at-stake dept.
A little fairy whispered in our ear: "MySQL AB is seeking a temporary injunction against NuSphere, even though they've finally released the source code for Gemini and MySQL Advantage. According to the GPL, NuSphere lost the right to redistribute when they violated #3 by not providing the source code originally. The FSF will testify tomorrow in court, according to this Newsforge article." Newsforge and Slashdot are both part of OSDN. We've done a couple of previous stories about the MySQL AB vs. Nusphere conflict: the original story, a follow-up, and a note about a countersuit. Update: 02/26 21:15 GMT by T : bkuhn (Bradley Kuhn of the Free Software Foundation) writes: "The FSF has a press release on the matter and affidavit that we filed is also available."
This discussion has been archived. No new comments can be posted.

MySQL AB and Nusphere Go to Court Over GPL More

MySQL AB and Nusphere Go to Court Over GPL

Comments Filter:

  • Ramifications? (Score:5, Interesting)

    by nakhla ( 68363 ) on Tuesday February 26, 2002 @04:00PM (#3072485) Homepage
    What will be the ramifications if the GPL doesn't hold up in court? If the GPL hasn't been challenged in court thus far, this could have far reaching implications within the Linux community. If the GPL doesn't hold up, does that mean Microsoft is free to take large chunks of GPL'd software and make it proprietary?

    • Re:Ramifications? (Score:4, Informative)

      by Anonymous Coward on Tuesday February 26, 2002 @04:06PM (#3072527)
      Trust me..if it was ever possible for anyone to make GPL'd code proprietary, without a reasonable amount of legal danger to the party stealing the code, universities would already have done it a LOOOOONG time ago. The GPL was written expressly so that institutions could not lay claim to code written to be distributed and used by the public. The BSD license, on the other hand, more or less allows anyone to take anything written for it, so long as they provide credit to the original authors.

      • Re:Ramifications? (Score:2, Informative)

        by Mr. Slippery ( 47854 )
        The BSD license, on the other hand, more or less allows anyone to take anything written for it, so long as they provide credit to the original authors.

        That's the "old" BSD licence. Problem is that you can quickly end up with pages and pages of author credits. The "new" BSD licence - which is essentially the same as the X11 licence - no longer requires this.

        • All versions and variants of the BSD license require that credit is given to the copyright holder. The only difference with the "new" version is that credit no longer has to be given in advertising material that mentions use or features of the code.

          Credit still has to be given in all documentation provided with the software, as well as within the source code. "Give credit" in this case really means that the whole software license, including a disclaimer, has to be reproduced with every copy.

    • Re:Ramifications? (Score:2, Interesting)

      by rutledjw ( 447990 )
      While this may have ramifications, I don't think they will be great. Nu DID end up releasing the source code, so it may be a moot point. I think what's most significant is that the FSF has proven they will fight to protect and enfore the GPL.

      If MS came out after this "viral" license I think that IBM and others who have strongly supported Linux (and hence, the GPL) would make that a costly proposition. Regardless of what IBM may think about the GPL, having the legal foundation of your marketing brain-child ripped out from under you is not a pleasant proposition. Plus, would people continue to develop code for Linux w/o the GPL? Another reason for IBM (and others) to step in.

      Additionally, potential gov't actions - not just the US gov't, but also European governments who take a dim view of MS - may also be a deterrent to MS trying that little strategy. Even so, I wouldn't put it past MS to try...

    • Re:Ramifications? (Score:3, Insightful)

      by JordoCrouse ( 178999 )
      If the GPL doesn't hold up, does that mean Microsoft is free to take large chunks of GPL'd software and make it proprietary?

      Good question! What if portions of the GPL are declared to be bad, or if the whole GPL is declared invalid, does that mean that a new licence can be drawn up and all the existing projects can be allowed to relicence themselves under the new licence, or are they stuck as GPLed forever?

      • Re:Ramifications? (Score:3, Interesting)

        by Drone-X ( 148724 )
        Good question! What if portions of the GPL are declared to be bad, or if the whole GPL is declared invalid, does that mean that a new licence can be drawn up and all the existing projects can be allowed to relicence themselves under the new licence, or are they stuck as GPLed forever?
        The GPL says that programs licensed under it can be regarded as licensed under that version or higher versions. What this means is that if the GPL was invalid, current GPL code would fall under standard copyright law (currently you can choose not to accept the GPL and use it under standard copyright either, but that grants you zero rights). The FSF could then release the successor to the GPL (which they are working on) and everything would be OK again.

        But there's more, the GPL includes a statement that if part of the GPL was regarded as invalid in court then that shouldn't invalidate the rest of the license.

        One more thing though, Linus has in the past changed Linux's license to GPL v2 and "no other version". This he was not allowed to do in the first place because he didn't ask permission to the many contributors, but if he wanted to bump the version number or say that it again falls under GPL vX or higher version he'd have to ask permission to everyone that has contributed since the license change. But of course everyone will again just let Linus get away with that.

        • Re:Ramifications? (Score:3, Informative)

          by Dwonis ( 52652 )
          The GPL says that programs licensed under it can be regarded as licensed under that version or higher versions. What this means is that if the GPL was invalid, current GPL code would fall under standard copyright law (currently you can choose not to accept the GPL and use it under standard copyright either, but that grants you zero rights).

          Nitpick: The GPL doesn't actually say this, but the text declaring that the work is licensed under the GPL usually (though not always) does.

        • Linus has in the past changed Linux's license to GPL v2 and "no other version". This he was not allowed to do in the first place because he didn't ask permission to the many contributors [...]

          This is incorrect. Before the change in terms, all GPL code in the kernel could be licensed under the GPL v2 or any later version, at the licensee's option. That means that I can take that code and and use it under the terms of the GPL v2 specifically. These terms allow me, among other things, to incorporate the code into any product distributed under the GPL v2.

          This is essentially what Linus did: All the GPL code contributed to him for inclusion in the kernel was under the GPL v2, though the authors may have allowed other licenses also. The authors of such code, by allowing Linus to license it under GPL v2, gave him explicit permission to redistribute it according to the terms of that license. The "or higher version" clause does not require that he follow the terms of all available licenses; it allows him to choose among the allowed licenses and use one of them.

      • well since the GPL is designed to give MORE rights to an end user. If it was declared invalid, i would Assume that standard copyrights would be in effect. ie, unless you have permission to use but not copy without my written permission etc. And thankfully for the clause in the GPL that collapses the license back to standard legal IP rights, anything found to be a problem would yank everything in the GPL until it can be fixed. SO A, B and C are rights in the GPL. If A is found to be invalid, B & C are revoked also.

        Any clarification on this if im wrong?

    • Re:Ramifications? (Score:4, Insightful)

      by lupercalia ( 310569 ) on Tuesday February 26, 2002 @04:15PM (#3072622)
      We have absolutely no reason to think it won't hold up in court. Have you ever heard of that happening to any other software license, free or not? I haven't. This speculation is all based on the fact that it hasn't been taken to court, not on any question as to its legality.

      In fact, the most likely reason it has never been tried in court is because nobody seriously thought they could overturn it, so they complied rather than face what they knew to be a losing court battle.

    • Without GPL you simply have no license to (Score:5, Informative)

      by Vicegrip ( 82853 ) on Tuesday February 26, 2002 @04:18PM (#3072651) Journal
      distribute your derivative work.

      The GPL is clear on this point. If, for whatever reason a court of law decides the GPL is invalid in a particular case, then you lose any right to distribute derivative works.

      Thus, there is no incentive for a company to try to invalidate the GPL, because without it they have no rights to distribute derived works. The only option a company has is to prove it is in compliance with the GPL.

      The one interesting thing here I observe is that once you have been found in violation of the GPL only the holder of the copyright can reinstate your ability to distribute derivative works-- and there is no obligation to do so.

      Thus, companies using GPLed code in works they distribute need to pay special attention to compliance as none-compliance carries with it the possibility being permanently blocked from using the GPLed code again.

      Admitedly though, it's not complicated. Simply ensure you distribute your source along with the product, and all is well.

      Otherwise put, if you want to use GPLed code, you have to share the code of your derivative work with everyone.
      • If you want to distribute a product based on GPLed code, then you have to share the code of your derivative work.

        You are allowed under the GPL to create a derivative work, say a modification particular to your business, and as long as you don't distribute the application outside of your business, then you are not obligated to share the code of the derivitve work (although you're encouraged to if the code might be useful to others).
      • The one interesting thing here I observe is that once you have been found in violation of the GPL only the holder of the copyright can reinstate your ability to distribute derivative works-- and there is no obligation to do so.

        This doesn't seem right. Why can't I, a random person with a fresh copy of MySQL plus source, modify MySQL very slightly then pass this derived work onto the defendant under the GPL, thus reinstating their right to redistribute.

        The article refers to article 4 of the GPL, but there's nothing there that makes special reference to the copyright holder. In fact, clause 6 says that if I redistribute the program then the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions.

        Surely that means that even if NuSphere were once in violation of the GPL, the copyright holders cannot take away their right to distribute derivative works so long as someone (e.g. Red Hat, Debian, the cat's mother...) is willing to distribute to them.
        • If the GPL is an invalid license, nobody but the copyright holder has the right to distribute copies, because the copyright holder only granted that right to others under the provisions of the GPL (which suddenly is no longer valid). The random person with a fresh copy would lose their license just as much as anyone else if the GPL were found invalid, which is the scenario being discussed here.
      • >The GPL is clear on this point.


        yes, but . . .


        >If, for whatever reason a court of law decides
        >the GPL is invalid in a particular case, then you
        >lose any right to distribute derivative
        >works.


        I am an attorney, but this is not legal advice. If you need legal advice, contact an attorney licesed in your jurisdiction.


        This is *far* from clear, though possible. I assumes that part of the GPL is stricken, the rest upheld, and no other defenses prevail.


        Other possible outcomes:
        *finding other terms to the license to replace the stricken terms (reformation)
        *finding that the failed licensing placed it in the public domain (unlikely without odd facts)
        *finding that the copyright holder is estopped


        While I believe, in the general case, that the most likely result flowing from a purported violation is that the "copier" has no rights, the other outcomes are possible, at least with the right facts. Of them, I would hazard a guess that reformation would be the most likely to arise.


        hawk, esq

      • The one interesting thing here I observe is that once you have been found in violation of the GPL only the holder of the copyright can reinstate your ability to distribute derivative works-- and there is no obligation to do so.
        Exactly. This is cause of the whole "RMS forgives KDE" incident of a few years ago. Most people felt he was just being an asshole when infact he was doing what he felt needed to from a legal point of view.

        • The RMS Problem (Score:3, Insightful)

          by alexhmit01 ( 104757 )
          The problem with what RMS did wasn't what he did legally, we all recognized that. It was his being a jerk about the entire situation.

          The KDE Team felt they were within the bounds of the law, FSF felt otherwise. Either there was no FSF code involved or they felt that their case was week, so they focused on complaining and launching a competing project.

          When the FSF and Trolltech worked out their differences regarding Qt licensing, RMS issued a statement applauding the change, forgiving KDE and it's users, and cheering on GNOME. Once Qt went GPL, there is no reason for the FSF to support GNOME (which sits on top of libraries with the "bad don't except under special circumstances" LGPL license ) over KDE except for NIH.

          RMS handled it with less tact than he normally uses, that is what pissed everybody off. The "forgiveness" could have been done in nice legalese on their website without trying to get it coverage.

          Alex

          • Re:The RMS Problem (Score:3, Informative)

            by Trepidity ( 597 )
            RMS handled it with less tact than he normally uses

            Don't you mean "the same amount of tact that he normally uses"? The man isn't exactly fames for his tact...
        • No code belonging to the FSF or RMS was in KDE. And they knew it. If RMS was right about KDE needing forgiveness, then the forgiveness should have come from only two parties: Linus Torvalds (for kfloppy) and Aladdin (for kghostscript). No other KDE code was a derivative work of any prior GPL code.

          To date, neither Mr. Torvalds nor Aladdin have forgiven KDE. I can only assume then KDE is still illegal and I am still a criminal for distributing it. Come arrest me.
      • Let's assume GPL is deemed invalid:

        The work is still copyrighted by the owner. If GPL was invalidated, you would need to negotiate an agreement with the relevant owner(s) if you wanted to use it. With any amount of collaborative input ("we used the libraries from project X and the drivers from project Y which was derived from project Z") this would be a real minefield that could see you sued further down the track (ala GIF). In the meantime, a new GPL would be circulated and pretty much overnight, most people would be distributing under "son of GPL".

        Me thinking out loud:

        I think it would take a ruling that took copyright away from owners to compromise things. Imagine a "Fair and Reasonable Corporate Access to Copyrighted Materials" Act that watered down copright for individuals. Say under the excuse that copyright shouldn't be used to protect DCMA circumvention and such copyrights are therefore forfeit.

        Xix.

    • Re:Ramifications? (Score:2, Interesting)

      by JonWan ( 456212 )
      I don't see it causing much of a problem in that someone (microsoft) could come in and "steal" code. Because it's still copyrighted by the author(s) that wrote it. It might be fun trying to figure out who wrote what, but for the most part things would continue with a new version of GPL.

  • Hmm.. are they the same?? (Score:4, Insightful)

    by lemonhed ( 412041 ) on Tuesday February 26, 2002 @04:03PM (#3072504) Journal
    NuSphere may not be 'obscure', but the mysql.org website seems a little deceptive. If I didn't investigate it more, I wouldn't have known the difference between mysql.org or mysql.com just by looking at the front page of the website.

    MySQL.org presents their website in a manner that does not give credit where credit is due.
    For one, if you go to mysql.org you will find that it doesn't say whether it developed the software, it just says it's got mysql software available for 'free' download. But the mysql.com guy is correct, in that you cannot download any software without registering with mysql.org first. That indicates to me that it's not 'free' because now mysql.org has customer data to use to market their product to. Plus, it doesn't say: "Hey, we didn't write the software, mysql.com is where you can find that info. We just improved upon it." Besides, I don't think .org websites should ever be for-profit businesses as that is not how that domain was intended to be used.

    • Re:Hmm.. are they the same?? (Score:4, Interesting)

      by jsprat ( 442568 ) on Tuesday February 26, 2002 @04:12PM (#3072597)
      Minor nit: mysql.org and mysql.com now point to the same IP. The canonical address is/was mysql.com, but the real MySQL site can be found using either address. This was settled (or maybe not, read the link ;) soon after the original story broke.

      NuSphere clearly didn't do "the right thing" and I hope they get their butts kicked.

    • Yeah, I mean, I'd never even think of using a commercial .org site. Especially to read news.

    • NuSphere may not be 'obscure', but the mysql.org website seems a little deceptive. If I didn't investigate it more, I wouldn't have known the difference between mysql.org or mysql.com just by looking at the front page of the website.

      MySQL.org presents their website in a manner that does not give credit where credit is due.
      For one, if you go to mysql.org you will find that it doesn't say whether it developed the software, it just says it's got mysql software available for 'free' download. But the mysql.com guy is correct, in that you cannot download any software without registering with mysql.org first. That indicates to me that it's not 'free' because now mysql.org has customer data to use to market their product to. Plus, it doesn't say: "Hey, we didn't write the software, mysql.com is where you can find that info. We just improved upon it." Besides, I don't think .org websites should ever be for-profit businesses as that is not how that domain was intended to be used.

      I completely agree the site is deceptive, although I'm not sure that is a violation of any law. Same thing with not giving credit for the code on the page. It may not be good practice, but as long as they follow the terms of the GPL (making source available, etc) then they are in the clear (legally at least). And I fully agree with the .org not being for profit (although it's a completely lost cause now). I have to take exception with the part about it not being free because you have to register. To be free (as in beer) means you don't pay anything. Saying it's not free because you have to register is like saying something isn't free because you have to go pick it up. Of course I still can't figure out what it being free or not has to do with anything, but free has nothing to do with registration.
    • That indicates to me that it's not 'free' because now mysql.org has customer data to use to market their product to.


      You can't change the licence, but you can charge *anything* (no, not equal to cost) for getting it from you. Translation: I can charge you $1000 to download MySQL from me, but you're free to distribute it at no cost. Read the licence, it's all in there.

      Kjella

  • An important case (Score:3, Insightful)

    by baka_boy ( 171146 ) <lennonNO@SPAMday-reynolds.com> on Tuesday February 26, 2002 @04:10PM (#3072566) Homepage
    Personally, I don't have much of a vested interest in who distributes MySQL, since I don't use it, and probably won't any time soon. However, as one of the first court cases to test the enforceability of the GPL, it think that it's critical that MySQL AB win. If they can't take on another small company over a GPL violation, how in the hell can we expect anyone to be able to stand up to a BigCo that decides to rip off their GPL'd code?
    • NuSphere is not at all a small company. It's GIANTLY HUGE ;-))

      NuSphere is owned and financied by PROGRESS
      http://www.PROGRESS.com/ which is really a
      giant company like Oracle...
      • Yeah. It's also not a company that "gets" Open Source, in some ways. And don't even mention the "F" word! They have enough trouble with "Open", never mind "Free".

        The involvement with MySQL is just about soaking up some of the low-end database market.

        Pass them by, and if you need a better database than vanilla MySQL, go for PostgreSQL. It has many features of the expensive Progress RDBMS, and can do some things that are far from easy in Progress.

        I might start recommending Progress again when they post Python bindings or protocol specs on their website.

  • I'm confused.. (Score:2, Interesting)

    by XaXXon ( 202882 )
    So the article said that NuSphere registered MySQL.org, and AFAIK the real MySQL web page is at MySQL.com.. but --

    [xaxxon@chopper log]$ nslookup www.mysql.com
    Server: localhost
    Address: 127.0.0.1

    Non-authoritative answer:
    Name: www.mysql.com
    Address: 64.28.67.70

    [xaxxon@chopper log]$ nslookup www.mysql.org
    Server: localhost
    Address: 127.0.0.1

    Non-authoritative answer:
    Name: www.mysql.org
    Address: 64.28.67.70

    Did I mis-read?

  • Section 4 of the GPL (Score:5, Informative)

    by lupercalia ( 310569 ) on Tuesday February 26, 2002 @04:21PM (#3072680)
    The most interesting comment raised in the article is the invocation of section 4 of the GPL:

    Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License.


    Basically, since the GPL is the only document granting you permission to use the software, violating the GPL revokes your rights under it. That means that if it is found that they violated the GPL (which seems a foregone conclusion if the reporting is accurate), they will no longer be able to distributed MySQL code at all. In other words, put completely out of that business.

    Now that is something other companies will take seriously in the future. (IANAL and all that of course.)

    • (disclaimer: IANAL, but I have some real-world experience with IP law)

      Enforceability of the GPL is problematic at best. It's no more legal than any other software license, and consider how many objections the average /.er raises to (say) Microsoft's clickwrap licenses. Those same objections apply to the GPL, but even more strongly.

      After all, if you can make the legal case that "clicking OK doesn't really count as accepting a license", you can easily make the case that "doing absolutely nothing (which is all that is required to accept the GPL) sure as hell doesn't count as accepting a license". If Microsoft can't implicitly force you to accept a contract, neither can the FSF. The law doesn't care if you're nice or not.

      However, even if (as I believe) the GPL is legally unenforceable, that probably doesn't hurt anything. GPLed code is still protected by copyright, so even if the GPL itself is powerless, the copyright holder (if such can be established) can still sue to prevent redistribution. The GPL, in this context, basically constitutes a non-enforceable statement of "we won't sue if you release source code".

      Even this might not hold up in court. Trademarks are lost if you don't protect them. I don't believe the same applies to copyright, but only an IP lawyer would know for sure the legal ramifications of selectively suing people who don't follow your (arbitrary, non-legal) license.

      Hopefully the courts will shed some more light on this soon.
      • Enforceability of the GPL is problematic at best. It's no more legal than any other software license, and consider how many objections the average /.er raises to (say) Microsoft's clickwrap licenses. Those same objections apply to the GPL, but even more strongly.

        Ok, if the GPL is found unenforcable, where do these people think that they get any rights to distribute the program at all?

      • accepting the GPL (Score:4, Informative)

        by brlewis ( 214632 ) on Tuesday February 26, 2002 @05:17PM (#3073208) Homepage
        More concisely,
        5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it.

      • Re:Section 4 of the GPL (Score:5, Informative)

        by bwt ( 68845 ) on Tuesday February 26, 2002 @05:36PM (#3073368)
        Enforceability of the GPL is problematic at best. It's no more legal than any other software license, and consider how many objections the average /.er raises to (say) Microsoft's clickwrap licenses. Those same objections apply to the GPL, but even more strongly.

        Totally wrong.

        My objection to Microsoft's clickwrap licence is that it only purports to grant me a licence to "use" the software on a single machine, which I already have by 17 USC 117, since I am the "owner of a copy". Since their contract does not give me anything I don't already have, it is unenforcable because there is no "consideration".

        The GPL is a unilateral grant to do something that you cannot otherwise do without violating 17 USC 106. It is not a contract at all, but a unilateral grant. If the GPL is unenforcable then NuSphere is commiting copyright infringement by distributing a derivitive work.
        • Since their contract does not give me anything I don't already have, it is unenforcable because there is no "consideration".

          Bravo! Your payment for the software doesn't count as consideration, because you paid for it *before* you clicked the button, and you paid that fee to someone *other* than Microsoft (like to the clerk at CompUSA). The MS EULA is not a legally binding contract.

          Trying to unilaterally impose a contract after the fact is ludicrous.
          • Exactly.

            Even though Softman v Adobe didn't reach the matter of shrinkwrap licences, it all but decides it.

            When you walk into the store or whatever and hand over cash for the cardboard box, you get a little piece of paper called a receipt, which lawyers will call a "contract of sale" because it specifies item, quantity, and price. At that point you are the owner.

            Since the software vendor probably wasn't even a party to the contract of sale, they cannot even claim that it is the "completion" of the sale by "acceptance of goods". That undercuts the only theory that any court has employed to hold any type of shrinkwrap licence enforcable.

            In fact, I don't even think the shrinkwrap should count as an "offer" even if it did purport to offer you something in consideration, because its method of acceptance produces no manifestation of assent outside of your own property. I call this attack by offer. It goes something like this: "Microsoft, I make you the following offer: if you would like to grant me a worldwide licence to all of your intellectual property in exchange for an unqualified option to have my old socks, then simply have one of your executives turn his or her monitor off and on in the next day".

            It doesn't work because the fact that you are claiming that an otherwise legitimate use of their property is acceptance does not allow you to attach a meaning to that use. The fact that they did it may simply indicate that they believe you are irrelevent. You have the right to install and use the software by 17 USC 117. The fact that it purports to offer you an "accept" option does not mean that you cannot click your mouse on the YES button as you please. Similarly, the MS executives can turn their monitors off and on as they please without assenting to your offer.
            • See comments at the end of FEDERAL APPEALS COURT HOLDS SHRINK WRAP LICENSE UNENFORCEABLE [lgu.com] on this topic, written by someone who doesn't have to add "IANAL". :-)
              • That article was written in 1992, so it's very out of date. In the case of a warrentee disclaimer, a different set of notification laws apply. "Service sold separately" is a completely reasonable point of view, given proper notification. Most EULAs attempt to use "notification" as an excuse to take away the user's rights. Only a contract can do that.

                It's important to know that all of the caselaw on this subject makes it very important to look at the exact circumstances. For example, if you are installing from one disk onto more than one computer, then you DO need a licence and in that case it is sort of like the GPL: if you don't accept the EULA then nothing else grants you multiple machine use. The "nothing else" part fails in the case of single machine use because there is something else: 17 USC 117, which is not subject to approval by the vendor.
            • I don't know if you'll see this...but *THANKS*

              I didn't follow this discussion originally, but I really like your description, and the clear concise way this was described.

              I would assume that you are a lawyer? (I'm sorry for all those lawyer jokes... [grin])

              Is this why the software makers want UCITA so bad? I would assume that the legality of EULA's are on VERY shaky grounds, having never been tested, and they know it. Knowing this, they want some REAL legal cover. Reply if you can.

              Thanks again!

              Cheers!

      • Re:Section 4 of the GPL (Score:5, Insightful)

        by egomaniac ( 105476 ) on Tuesday February 26, 2002 @06:27PM (#3073863) Homepage
        I don't normally reply to myself, but since everybody has completely and totally missed my point...

        Yes, you are all absolutely correct that without the GPL, it is illegal to distribute the software at all.

        However, I evidently didn't explain myself well enough. My point was not that it would be legal to distribute software without the GPL, far from it.

        The GPL grants you additional rights if you follow certain provisions. If you don't follow those provisions, you're subject to ordinary copyright law (which prevents you from redistributing the work). If you do follow those provisions, you have the legal right to redistribute the software.

        Now, work with me here. You cannot sue somebody for a GPL violation. Period, end of story. All you can sue them for is copyright violation, since without the GPL's provision you can't copy the software. *All* GPL violations will be tried in court as copyright violations, because that is the only law you could have broken. The only penalty for breaking the GPL is revocation of your license, which leaves you subject to copyright law.

        You're all looking at the enforceability issue backwards. The enforceability of the GPL does not *ever* protect the people who offer to license the software. You don't need the GPL for that, because you have copyright law -- copyright law is completely sufficient to shut people down from using your software. You don't even really need a license for that, because you're free to sue company A because you don't like the way they are using your software, but leave company B alone because you're happy with what they are doing. This is completely legal, and you don't need the GPL for that. (I realize that the GPL fulfills a very important role as far as formalizing the agreement and making it easier to get people to comply, but legally it isn't necessary. You could just sue anybody whose use of your code you disagreed with.)

        The GPL is just a formalized statement of "I won't sue you if you distribute source". It protects the people *using* the software, because while copyright law would ordinarily say "you can't do this", the GPL says "you can if you distribute source". The GPL does not grant one iota of extra power to the people licensing the software, it grants it all to the people using it -- a statement of protection from lawsuit.

        So firstly, it's pretty much irrelevant. The legalese of the GPL could be replaced with "I promise I won't sue you if you ..." and it would have exactly the same effect.

        As far as the legal enforceability of the GPL, it may or may not hold up in court. Since everybody is going to get this backwards, I'll spell it out. "Hold up in court" means use it to *protect* yourself, not to sue somebody else. Again, we've already established that the only law being broken is copyright -- you cannot use the GPL to attack somebody for violating it. The GPL is only good for *defending* yourself, to say "no, look, I'm not violating copyright because the GPL allows me to do this". *That* is what needs to be tested in court, and that is what I'm not sure will necessarily work. Theoretically, you might be able to successfully sue somebody for using GPLed code completely in accordance with the license, and *that* would be the "not holding up in court" that I'm talking about.

        It would obviously be horribly unethical to do so, and any reasonable judge might well throw the case out, but a good lawyer could probably make a case against the ability to use GPLed code in the absence of a formal agreement.

        Hope that clears up what I meant by my first post.

        • Re:Section 4 of the GPL (Score:3, Insightful)

          by ahde ( 95143 )
          and the argument will be that "because this product was licensed under the GPL, it was represented as being freely available to copy, modify, and distribute. If it had not been released under the GPL, copyright would still be in effect. But, a derivative work was made (at great expense my clients part) in good faith, with the understanding that the product could be distributed."

          Basically, they'll argue that you can't give away 99% of something and then come and ask for the whole thing back like an Injun Giver. The motion will be for the GPL to be ruled "equivalent to public domain."

          A contract does not mean only what the contract writer wants it to mean. That is why you have recourse *heh* if you sign a misleading contract. The analogy will be drawn that you could buy a house for $1 on the condition that you must paint it pink at midnight on July 4th, 2015. When 2015 comes around you have a vested interest in the house beyond your initial contract $1 -- therefore the pink-at-midnight clause could be challenged.

          • 'The motion will be for the GPL to be ruled "equivalent to public domain."'

            I don't think that is remotely possible. If there's a weak point in law it's that it tends to side with licensors and those who make the rules. GPL is a way of making the rules... and unlike, say, the BSD license, it makes very STRICT rules of what you must do to comply.

            In essence, to a lawyer, the GPL is totally unlike giving away. It is laying strict rules for a collaboration that HAPPENS to be with people you don't even know. It has REQUIREMENTS that must be obeyed, or the deal is off. The fact that it produces a pool of 'cooperating' software is of no interest to a lawyer... that's a social effect, we're talking about what the contract is...

            If it ever did become invalid, first of all I think it would become invalid only under certain conditions (don't ask me what, because I _don't_ think it will ever be 'ruled' something other than what it is). And second of all, it would absolutely not fall through to being public domain. It would fall through to being normal copyright law, and people affected would have to renegotiate their licenses, perhaps they would have to get direct consent from the authors of code they used or something. There is just no way it would be deemed public domain- it so plainly is NOT trying to be. In some senses (ask a GPL hata ;) ) it is not 'giving away' anything...

      • After all, if you can make the legal case that "clicking OK doesn't really count as accepting a license", you can easily make the case that "doing absolutely nothing (which is all that is required to accept the GPL) sure as hell doesn't count as accepting a license". If Microsoft can't implicitly force you to accept a contract, neither can the FSF. The law doesn't care if you're nice or not.

        Unlike shrinkwrap licenses, the GPL does not resrtict rights that the user would otherwise have. Rather it gives the user a right to redistribute which copyright law by default prohibits. If you want this _extra_ right then you also have some responsibilities. This could not be further from a shrinkwrap license. On the other hand, I'm not a fan of either.

    • In a normal contract... there is a point where the contract an agreement is reached, the contract is signed. If a clause in the contract causes the contract to be terminated, it's null & void.

      But you see, with the GPL, there is no negotiation. What prevetns someone from re-licencing the software under the GPL again? Nothing. As long as they are currently complying with the terms, they can keep using it.

      Why is the GPL different? Because nobody CHOOSES to let someone license it.
    • if it is found that [NuSphere] violated the GPL (which seems a foregone conclusion if the reporting is accurate), they will no longer be able to distributed MySQL code at all. In other words, put completely out of that business.

      I think this is an overbroad interpretation. NuSphere violated the license, so MySQL AB can sue for copyright infringement. However, it seems to me that all it takes for NuSphere to get a new GPLed copy of MySQL is to download it again. Presto, blammo, reinstated--no "forgiveness" necessary.

      It's the same as if Microsoft caught me pirating Office. If I have any money left after the BSA and MSFT lawyers are done with me, I can still wander down to Best Buy and get a licensed copy of Office.

      In the case of GPLed software, it's just a lot cheaper to get a new copy.

    • Basically, since the GPL is the only document granting you permission to use the software, violating the GPL revokes your rights under it. That means that if it is found that they violated the GPL (which seems a foregone conclusion if the reporting is accurate), they will no longer be able to distributed MySQL code at all. In other words, put completely out of that business.

      I agree that GPL compliance is important, but I am not clear on the legality of this license termination business. What is to keep NuSphere from just downloading a fresh copy of the mySQL sources -- from MySQL AB or another GPL licensee -- and working from there? That copy would, after all, come with a fresh new license.

      The doctrine that MySQL AB seems to be pushing is that once an entity violates GPL on a product, not only has that entity's existing license to the product been terminated, but that entity is tainted, forbidden from accepting any future license to that product. This does not appear to be what the GPL itself says. The GPL speaks of voiding the instant license, but not of tainting the offending licensee from accepting a future one.

      Does anyone have an explanation for the disparity?

    • Basically, since the GPL is the only document granting you permission to use the software

      Wrong. The *law* gives you the right to use any software you have legally obtained. Period. You won't have the right to create derivative works, or to redistribute it, but you do have the right to USE it.

  • GPL enforceability (Score:2, Insightful)

    by Anonymous Coward
    As usual, it is worth pointing out that if the GPL were ever found unenforceable, then you have NO right under copyright law and international treaty to use the code - it doesn't suddenly become public domain.
    This is one of the reasons the GPL tends not to get challenged - it's a lose-lose situation for a GPL infringer who challenges it in court:

    Scenario a: GPL challenge fails. Infringer has to GPL all derived work of original GPL code that he wants to distribute. Presumably he didn't want to do this, otherwise he wouldn't have taken it to court...

    Scenario b: GPL challenge succeeds. Infringer has to stop distributing all derived work of original GPL code, as he now has no rights granted to him by the original copyright holder to use the code.
    • Bingo, I've got an (albeit small) amount of code floating around in a few of the various Linux distros. Under copyright law in the U.S., and most other industrialized countries, if the GPL were to fail the rights on my code would revert back to me. At that point I would be legally allowed to charge a license fee for its use, completely ban its distribution or use, or relicense it under a more bulletproof version of the GPL (which is what I would do).

      It would NOT be in the best interest of any company wanting to use GPL'd code to try and overturn the GPL.
    • Here's a question to that, which I don't know the answer to, but seems rather interesting.

      Here's what I'm mentally debating; let's say the GPL is struck down and invalidated, wouldn't that then mean that anyone distributing a derivative work would be in violation of the original creator's copyright? As you said the GPL no longer gives them right to create derivative work?

      Now I'd imagine that most/all of the original coders probably wouldn't sue, but could they if they wanted to? Could someone sue Redhat for distributing their formerly GPL'd code out of spite (they could just say the license they were using is not valid anymore, so what defence would there be to being able to use the code).

      Just doing a little musing... I'll let someone with more license knowledge debate the question, just throwing it out there.

    • One word: Microsoft (Score:3, Insightful)

      by QuantumG ( 50515 )
      Get sued random small ass person who has code under the GPL. Get the GPL invalidated. Shut down all the millions of free software projects that make use of the GPL. Hell, forget Microsoft, how do we know that Theo da Raadt or some other cat stroking evil genius isn't gunna do it (no offense Theo). If the GPL is declared invalid then surely you have to be wrong about this whole "no rights to distribute at all" stuff. Surely 99% of people using the GPL would immediately issue statements saying in essence that they are switching to a BSD license. Ahhh, fun with copyright law.

  • More information on FSF Website (Score:5, Informative)

    by bkuhn ( 41121 ) on Tuesday February 26, 2002 @04:31PM (#3072755) Homepage
    FSF has a press release [fsf.org] and the affidavit we filed [fsf.org] available on our website.

  • Timeline on Source Release? (Score:3, Interesting)

    by bhsx ( 458600 ) on Tuesday February 26, 2002 @04:40PM (#3072823)
    It is also acceptable for the license to require that, if you have distributed a modified version and a previous developer asks for a copy of it, you must send one.
    and also...
    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,

    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 on a medium customarily used for software interchange; or,

    c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.)

    I don't see, anywhere in the liscense, a timeline specified as to when any changes to source code must be supplied. Anyone know of a quote from the GPL that specifies that the source must be made available at the time of binary release? AFAIK, they don't have to release the code to anyone who doesn't specifically request it, and who has a copy of the binaries; and I don't think time-frame is brought into it.

    • I don't see, anywhere in the liscense, a timeline specified as to when any changes to source code must be supplied.

      The timeline is very clear and obvious, at least to anyone who can read english, namely:

      You have to make the source code available the moment you begin distributing the GPLed software.

      Even if you choose option (b) or (c), the source code has to be available, and provided on request, the moment you begin distributing your derivative work.

      Where is the rocket science in this?
  • Alot has happened in the MySQL AB vs. Nusphere fight. First Nusphere took the mysql.org domain and acted as if they were the not-for-profit corp behind the software. MySQL AB was right to be frustrated about the situation. But for what we know, this question got resolved, as mysql.org now points to the original mysql.com site and WHOIS reports MySQL AB as the owner.

    So where do we go from here? These guys believe that they can enforce article 4 of the GPL against Nusphere for an infraction to the GPL that occured in the past and which has been corrected since then as Nusphere published its source code. I personnaly believe this is not The Right Way. It totally goes against the spirit of the GPL: software must be free, and although we want to protect the original copyright holder, we must not discriminate against the users.

    Many people could try to profit from the GPL in an illegal way, does it mean that we can put them on a blacklist? If they decided to change their attitude and in turn abide to the license, I don't see why we would still punish them. They can be a valuable ressource in improving the software, even though it clearly goes against MySQL AB's agenda.

    Hey, if you're going to act that way, why did you release it under the GPL in the first place? By retaliating the way they do MySQL AB makes me wonder who between them and NuSphere has the weakest principles.

    Using the GPL as a tool for revenge is definitely not what Stallman & followers originally intended.

    Of course, MySQL AB could and should sue NuSphere for misrepresentation, moral copyright infrigment and dubious corporate conduct. They should seek damages, which they are fully entitled to under their country's Copyright Act. I for one find it unacceptable to simply lock Nusphere out.

    It is sad to see that the first judicial test of the GPL is witness to such a hijacking. And the worst thing: the FSF has it's arms in the mud upto the shoulders.

    This is one of these situations where Stallman actually should open his big mouth like he usually does.

    I don't think that the judge will let that injonction go through. Their case is too weak to achieve that. Anyways in the worst case, what will prevent Nusphere from starting another company?

    pertools [pertools.com] will bring you the thruth.

    • Article 4 of the GPL is critical (Score:4, Insightful)

      by JoeBuck ( 7947 ) on Tuesday February 26, 2002 @06:50PM (#3074115) Homepage

      The reason Eben Moglen has gotten dozens of companies to give up and to submit is because of section 4. Without it, we'd have a lot less free software than we do now. In the past, the threats of nuclear war have been private, but very serious (if you're in the Linux business and lose your right to distribute, say, glibc, you're dead meat).

      It's important for everyone to understand that if you violate the GPL, it's not sufficient to just stop violating, you need to get the copyright holder's explicit permission before you can ever start copying, modifying, or distributing the program whose copyright you violated ever again. People got pissed off when RMS talked about "forgiving" the KDE project, but too many people don't realize that from a legal standpoint this forgiveness was required (though evidently only a couple of less-important KDE applications ever had any FSF-owned GPL code in them). Certainly RMS could have been more diplomatic (though maybe not, it isn't one of his talents).

      • >People got pissed off when RMS talked about "forgiving" the KDE project, but too many people
        >don't realize that from a legal standpoint this forgiveness was required (though evidently only a couple of less-important KDE applications ever
        >had any FSF-owned GPL code in them). Certainly RMS could have been more diplomatic (though maybe not, it isn't one of his talents).

        Actually the "forgiveness" combined with "Go GNOME" was a bad insult. It has not been forgotten by anyone. It is good to see though that RMS finally wants cooperation between KDE and GNOME.

        Add to this insult (for creating the first GPLed Desktop Environment!) the fact that NO KDE application had ANY FSF or RMS-copyrighted code in it. There was a little code from Alladdin (kghostscript) and the linux sources (kfloppy). That was it

        Finally until this day it is not certain if the RMS strict interpretation of the GPL is valid or not, because QT might well be considered as part of the Linux distribution.
    • Progress did not just download the MySQL source and develop their NuSphere database. They entered into a contract with MySQL AG for approximately $3,000,000 for software and services. Some (maybe all) of the work done by MySQL AG has since been released subsequently under the GPL, but this is more than a case of GPL violation. Progress Software used proprietary software created by and in collaboration with MySQL AG. They then broke the terms of their contract, specifically relating sections of the GPL.

  • Basically, MySQL AB is arguing that because NuSphere violated the GPL *in the past*, they have forfeited their right to distribute a GPL app, even though the source code in question has since been released in compliance with the terms of the GPL. What are the ramifications if they succeed? Does this mean NuSphere will be unable to distribute any GPL applications? Will the NuSphere contributions be deemed invalid? Will MySQL AB then distribute MySQL with Gemini tables while NuSphere is not allowed to? (Gemini tables are a NuSphere contribution.)

    Isn't it sadly ironic? The first time GPL is tested in court, and I hope they loose, for all our sakes. No matter which way the decision goes, the dispute hurts the entire free software community, and will make corporate contributions to code base that much harder to come by. And think of what Craig Mundie from Microsoft will have to say about this! All over what was originally a domain name dispute. (If your new to this conflict, NuSphere paid MySQL a huge chunk of change for the right to distribute MySQL, (that's right, paid for the right to distribute a GPL app.). In the process, they created a mysql.org web page that did not give any credit to MySQL AB, and indeed, only made SQL Source Code available to registered users. MySQL AB charged NuSphere with GPL violation. In response, NuSphere made the source code more available and released the source code for their own proprietary modules. It is still not clear to me, however, what exactly MySQL AB thinks the millions they were paid by NuSphere was supposed to be for. Somehow, the right to distribute a GPL app for a limited time just doesn't make sense.)

    MySQL AB, I thank you very much for your hard work and development of MySQL. But please, stop pissing on the GPL to advance your own agenda. You can debate whether the money you were paid gave NuSphere the right to hijack MySQL trademark until the cows come home; but all GPL issues have been put to bed months ago!

    • Basically, MySQL AB is arguing that because NuSphere violated the GPL *in the past*
      [snip]

      Has NuSphere released the source code for all previously distributed versions of their bastardized version of MySQL?


    • We have said earlier that we welcomed the opensourcing of Gemini when it finally happened. But NuSphere still denies having violated the GPL in the first place, which, if left at that, may set a damaging precedent and is a potential threat to ANYONE developing GPL'd software. I believe it is in everyone's interest that the GPL is a defendable licensing model.

      And, for those who may not know, we at MySQL AB have made numerous attempts to settle the case out of court.

      Marten Mickos, MySQL AB
  • The early version of NuSphere was clearly in violation of the GPL and thus _de facto_ renounced their rights to redistribute that version of MySQL. However, it seems that they've cleaned up their act for the next version of NuSphere.

    What I'm wondering is if that revocation is permanent, absent forgiveness by the licensor of the MySQL code? It will obviously be so for that particular version of MySQL. But what about later versions of MySQL? New code, new license. Does their initial unrepentent infringment of the GPL mean that they are forever barred from redistributing newer versions of MySQL whose license they haven't violated?
  • Seems to me that Nusphere is trying to greedily free-ride off the hard work of Open Source developers. Simply put: Nusphere doesn't get it. They are one of many who still mistakenly believe that proprietary software is the only way to make money. As such, I personally hope they get ripped to shreds in court and all their proprietary modifications get forcefully released to be freely assimilated into the GPL codebase as MySQL developers see fit.
  • Nusphere should continue to be able to release their software in compliance with the GPL, however for each copy of their software they distributed without complying with the GPL, they should be forced to negotiate with MYSQL AB etc. and agree on a fee to be paid for each copy of the software distributed illegally.

    If precise figures cannot be had, then an estimate should be created based on the length of time NuSphere was distributing product that did not comply with the GPL.

    A price per copy should be constructed based on similarly featured products in the marketplace. It could get expensive for NuSphere, but commercial software is an expensive commodity.

    As MySQL is open-source, it may make more sense to have NuSphere donate the money to an organisation like the EFF or the FSF, to support Open Source software as a whole.

Slashdot Top Deals

Repel them. Repel them. Induce them to relinquish the spheroid. - Indiana University fans' chant for their perennially bad football team

Close