Please create an account to participate in the Slashdot moderation system

 


Forgot your password?
Close
typodupeerror
×
The Courts Government News

NuSphere vs. MySQL AB Hearing 184

Posted by michael from the legal-wrangling dept.
An anonymous submitter sent in: "The hearing is over, and Adam Kessel posted a report about what happened in the court room. The judge is probably not going to issue a temp. injunction against NuSphere on the GPL violation, but probably WILL issue an injunction on trademark issues." Politech has another report on the hearing.
This discussion has been archived. No new comments can be posted.

NuSphere vs. MySQL AB Hearing More

NuSphere vs. MySQL AB Hearing

Comments Filter:

  • More on the NuSphere case (Score:5, Informative)

    by Anonymous Coward on Wednesday February 27, 2002 @11:58PM (#3082315)
    Here's some background: Boston judge to hear first test of GNU license used in MySQL [politechbot.com] and NuSphere replies to Politech post about FSF and Boston lawsuit [politechbot.com]. Looks like the GPL isn't an issue here after all -- in other words, it may just be a run-of-the-mill contract dispute between two parties.

  • the short of it.... (Score:3, Informative)

    by univgeek ( 442857 ) on Thursday February 28, 2002 @12:01AM (#3082324)
    The GPL is being treated as any other software licence, and the legality of the GPL itself is not being questioned - yet.


    Also the judge didn't want to get into the technicalities as they were way beyond her at this point.


    MySQL wins on the trademark count (for now at least), but the motion on the GPL violation is being pushed to August.

  • Link to the GPL (Score:2, Informative)

    by Yoda2 ( 522522 )
    Thought a link to the actual GPL might be helpful. Here it is. [gnu.org] My (limited) understanding is that NuSphere released a product statically linked to MySQL, but did not release the software under GPL and therefore violated GPL.

    • Re:Link to the GPL (Score:5, Funny)

      by sinserve ( 455889 ) on Thursday February 28, 2002 @01:16AM (#3082558)
      OMFG!

      Did you just link to the GPL? Dude, I have a cron
      job the searches my harddisk for the GPL and deletes it, incase I have
      installed new software.
      Infact, the perl script invoked by cron not only
      does look for the name "COPYING", it also does an
      MD5 checksum, and knows about the finger prints of
      both GPL versions.

      The next release will be network enabled, and we
      are aiming for a full blown web crawler, of google
      proportions, that does nothing but exploit servers
      and delete GPLs.

      The world needs only ONE GPL, in Stallman's box, and
      the rest of us can symlink to it.

      --
      • So, in a way, I am inventing the world's first
        immunition for a "viral" outbreak ;-D

        --

      • Are YOU going to be the one that pays to buy RMS a box that can take that many hits?


        Oh yeah, a P.S. just for those that moderate this kind of crap up because they think Stallman is an egomaniac, a religious nut that wants everything to run on Gnu/Linux and all that rot.


        Stallman.org [netcraft.com] is running Apache/1.3.6 (Unix) on FreeBSD.


        So there. Get over it.

  • Nusphere fixed the issue (Score:5, Informative)

    by topham ( 32406 ) on Thursday February 28, 2002 @12:16AM (#3082371) Homepage

    Nusphere fixed the basic issue here; they no longer distribute without source code.

    MySQLs actions make little sense, they are acting like spoiled children who didn't get their way. (Incomprehensible to me; I think an ego clash occured)

    MySQL can't claim damages with regard to the GPL because there aren't any. If you look at the information released the other day you'll note near the end (EFF statement) That Nusphere fixed the problem; That EFF encourages compliance, not court action; and the short acknolwedgement that they believe Nusphere violated the license.

    You would note near the beginning they state the purpose of the GPL and their general intent to reach a compliance.

    • Nusphere fixed the basic issue here; they no longer distribute without source code.

      I think the issue is that they STILL have proprietary code (shipping without source code) linking to the mySQL Libraries (thus violating the GPL).

    • Re:Nusphere fixed the issue (Score:5, Informative)

      by yamla ( 136560 ) <chris@@@hypocrite...org> on Thursday February 28, 2002 @12:32AM (#3082426)
      Nusphere fixed the basic issue here; they no longer distribute without source code.

      Assuming this is true (I have read conflicting reports), the GPL does not actually allow you to 'fix the basic issue' and then continue on as you wish.

      Section four of the GPL [gnu.org] states:

      You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance.

      Once Nusphere initially released without source code (they claimed source would be available later), they were, strictly speaking, in violation of the GPL and by section 4, automatically terminated any claims to a license that they had.

      Now, most reasonable companies would allow them to get into compliance with the GPL, then drop the matter. MySQL decided not to. While I do not agree with this action, it is, strictly legally speaking, the right thing for MySQL to do in this situation.

      • Now, most reasonable companies would allow them to get into compliance with the GPL, then drop the matter. MySQL decided not to.

        Normally, yes. However, it's important to note that the offended party has no requirement to forgive such violations, and that is intentional.


        While I do not agree with this action, it is, strictly legally speaking, the right thing for MySQL to do in this situation.

        If this were a simple, onetime mistake I would agree with you. However NuSphere has been making an arse of themselves and being a thorn in MySQLs side for years. There is a lot of history here, and NuSphere's part is VERY blamewourthy. Given the history, I don't blame MySQL in any way for this - if I were them I'd do the same damn thing. Make no mistake, NuSphere picked this fight, not MySQL. MySQL just happens to have section 4 on their side.


      • Read Section 6:

        "Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions." (my emphasis added).

        All NuSphere has to do is prove that MySQL AB distributed the program again and their licence was reinstated. The concept that they've lost their licence to redistribute because of past transgressions is baloney.
        • Except that it's invalidated by section 4 once they were in violation of it. This is why they would need to be forgiven.

          Like it or not, that's one of the reasons of why the license works. If you violate it, you lose all rights to use/redistribute.
          • It isn't invalidated by section 4 for the simple reason that it is a new license. They were in breach of the old contract but section 6 is quite clear in that you get a new grant of the license every time you receive the code, hence you get a fresh chance to do the right thing.

            Like it or not, if you violate the GPL (IMHO) it's quite clear that you regain your rights to distribute once you fall within compliance again.
            • To say that it is circular and obtuse logic is an understatement and clearly not within the spirit of what was placed on paper. Cry all you like, the intent is rather obvious and this generally carries significant weight when being reviewed by a court.

              Using your logic, one is encouraged to break the law as there is no penalty which was CLEARLY not the intension. Rather, it's intent (which is equally clear) is to restrict your freedom once you've decided to spit on the GPL. This only makes sense as it's quid pro quo. That is, since the violator has decided illegally to restrict everyone else's rights to what should be freely available, which flies in the face of the intent of GPL, it in turn, ensures the violator has no rights which forces you back to the table greatly weakened.

              • I'm interested in where this obvious "intent" is. Everything I see in the license reads very much of the intent to allow source to get out there among entities that are in complaince with the license.

                You seem to forget that the discussion isn't about entities that are CURRENTLY in violation of the license. There is a clear penalty for breaking the GPL still - you cannot distribute the code you produce until you are in compliance with the license. The whole need for "forgiveness" from the copyright holder is a rather weak interpretation which comes from people who see a need to perpeptually punish individuals or organisations that step over the line rather than the clear reading of section 6 which gives the intent to allow any company that is in compliance with the GPL's statutes to distribute further.

                You also forget that even under some wild interpretation of the GPL which bars an entity forever from using a certain piece of software that it is a trivial exercise to create a separate entity which gains all the IP of the previous entity and yet is not encumbered by the "need for forgiveness" that you are clutching at.

                In the end the "intent" of the GPL is to make software free (as professed ad infitium by the FSF). Chaining down companies for past mistakes is NOT making software free - it is a pursuit of vendettas. The FSF has also stated many times that their goal is to bring companies into compliance with the GPL rather than force them out of business. I'm sorry, but there is no intent (clear or obscure) to hold back companies which are currently in compliance with the license's terms.

    • Re:Nusphere cannot fix the issue on their own (Score:4, Informative)

      by VP ( 32928 ) on Thursday February 28, 2002 @12:35AM (#3082432)
      The GPL is a distribution license; it gives the licensee rights that they would normally not have, namely to distribute the code. By violating the GPL, NuSphere lost these rights, and until the copyright owner decides to give it to them again, they cannot distribute the code. This is the main argument, and it's validity may be decided by the court case.

      • until the copyright owner decides to give it to them again, they cannot distribute the code.

        ITYM "The copyright owner or anyone else who has licensed the code from the copyright owner under the GPL."
        • No, don't be silly. Only the copyright owner can grant the right to copy again. Otherwise Section 4 would be pointless.

          • RTFGPL! Anyone who has licensed the code under the GPL can, in turn, license the code under the same terms to another party. I could go download a copy of MySQL right now and license it to NuSphere, and they would get from me exactly the same rights they got from MySQL AB under the GPL.

            Now if you want to argue that this is not the _intent_ of the GPL, then we're just violently agreeing. But it's not the intent that matters - it's the text. The FSF will have to fix this loophole in GPL v2.0.
      • They immediately gain permission to redistribute as soon as they download a new copy. Section 6 clearly states:

        "Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions." (my emphasis added).

        In other words, they are automatically relicensed by the terms of the GPL as soon as they got a new version from MySQL-AB.
        • "Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions."

          Subject to these terms and conditions - so the new licence does not shrive the sins comitted.

          • The new license is a new contract and hence is indepenant of the original (broken) contract. It's evident from Section 6 that a new license can be gained to distribute the software so longs as that new distribution is under the terms of the GPL (as you so graciously highlighted).

            A new license does indeed shrive the sins comitted - in fact it's the only way to shrive them.
    • I think the point of not letting Nusphere fix the problems after the fact, come from the following delimma.

      Do we let anyone release binary modifications, and only punish them after they get caught?

      The GPL becomes mostly toothless, if you can abuse as much as you like, but when you get caught, you "quick, release the source..."

      So, the point can be made that Nusphere, it seems needed to be "taught a lesson" and used to make an example.

      I don't know that I agree with the result, I don't really know enough about the case. But I'm sure that this is something to think about.

      It's more about obeying the spirit of the law, rather than the technical license of the law. The GPL can't be too loose on enforcement, otherwise everyone will learn that you can fudge, but fix it if and when you get caught and think it might cost you...

      Just some thoughts.

      Cheers!
    • As you may notice, the action involves a contractual dispute, trademarks, and copyrights, not just a GPL violation. Furthermore, the MySQL suit was filed in response to a lawsuit by NuSphere.

      So, I find your assertion that the MySQL people are acting like "spoiled children" unfounded--there is more going on here. In fact, it isn't even clear to me that NuSphere fixed the problem. Even if they did, it isn't their's to fix: once they fell out of compliance with the GPL, they can't fix it--they have lost all rights to using the code.

      As for the issue of reaching compliance, if anybody can violate the GPL with impunity and then just fix it up without consequences, that will simply encourage a lot of companies to do that. Open source is at a serious disadvantage here because most GPL violations are very hard to detect. It makes sense to pursue some GPL violations seriously even if the violators eventually come into compliance.

      Finally, as for damages, that's bogus. Open source companies live off things like consulting and documentation. If some other company comes out with an enhanced closed source version that ends up getting used widely, the open source companies lose business. The argument isn't as simple (or simplistic) as the MPAA's "1 million people copied Rocky XVII illegally", but there are still damages. And there are damages to the public as well (although they may be harder to enforce).

      • As you may notice, the action involves a contractual dispute, trademarks, and copyrights, not just a GPL violation.

        It's rather hard to violate the GPL without involving copyright infringment and a contractual dispute anyway. Since the GPL is all about contract and copyright law in the first place.
        The tradmark bit is is a separate issue. Maybe related to something like NuSphere passing off their product as MySQL or similar.
    • I do recall, though, RMS causing a bit of a stir when TrollTech finally made Qt free. He pointed out that previously distributing otherwise-GPL code linked against Qt was a GPL violation, and, while the problem no longer existed, prior violators had to be *technically* absolved of their previous sin before they could redistribute once more. IIRC, he also urged the relevent copyright holders to forgive the gaffe.

      At the time, the /. crown seemed pretty upset that RMS was being so pedantic about license complience, though I found that he was just pointing out a technicality.

      So, it would appear that RMS, at least, intends for GPL violators to be forever prohibited from redistrubution of the relevent work unless forgiven. This is not inconsistent with also suggesting that returning to complience result in forgiveness - it just gives the GPL some teeth.

  • Good summary (Score:2, Interesting)

    by Anonymous Coward
    The NewsForge summary seems to be pretty good, especially when the author ponders about just what the infamous termination clause means, and how someone might be able to get re-licensed once s/he's violated the license. (Are you unable to use that code for life? Can you get a new license by re-downloading the code?)

    IMO, this is a very nasty detail that's left hanging in the license, and should be fixed in the next version. If nothing else, the license should explicitly say that you lose all rights to use the software "until you are granted explicit permission, in writing, by the original author", or something similar.

    • The GPL doesn't say more than a few words about "use" of software and they are pretty much "use it all you want". The license is about distribution. I think you meant to say distribution, since that is the context of the discussion, but it seems to me the distinction is important.

      No one is saying NuSphere can't *use* the software, what they're saying is that NuSphere intentionally violated the GPL and that in addition they're being fookin' annoying to MySQL AB so the AB folks are not going to just let them start *distributing* the software again, compliant or no.

      As to some of the issues you mention, I think it's fairly obvious that "you" in the license means the person or entity involved. So conceivably NuSphere could engage in some old shell game and maybe become a different legal entity and then be able to distribute again-- but that's not easy, cheap, or likely to make much sense to their customers.

      But the whole thing about sharing the code doesn't make sense at all. If you share the code, you are following the GPL. If you share binaries, eventually you're going to want to recompile (if you're a distributor) -- not possible without the code. Which throws that whole hypothetical paradox on its ear if you ask me.
    • It's very dangerous to say "original author" however. RMS (and folk) coded the license very carefully not to include such phrases, because the original author may be different than the current project leader, especially in code forks. If the code forks, it may fork to change the current leader (in disgust, or creative differences, or whatever), and you don't want the "original author" to have any more control over it.

      The GNU license is open-ended in such a way that you could set your "command structure" any way you want it. By its very nature, the code/document it protects is public-domain, and no "admin power" is give to anybody.

      Perhaps a time of about a year might work, but then again, anybody trying to screw over the licence like that deserves to never touch the project again. Future generations will learn to look over their licenses very carefully if they don't want to get burned.

      I'm sure this will give people a message to never fuck with the GNU license, the projects it protects, and never EVER fuck with the FSF!
      • The GNU license is open-ended in such a way that you could set your "command structure" any way you want it. By its very nature, the code/document it protects is public-domain, and no "admin power" is give to anybody.
        This isn't true. Even if you fork the project the "original author" still has copyright on their work. He's just not able to revoke the license as long as you comply with the terms of the GPL. So, if you decided to do something like change the license it would still require approval of the "original author". Or if someone way down the line violates the GPL with your fork, the "original author" still has rights to enforce copyright law on the portions he wrote.
  • I like that part about how everyone should sign their code over to the FSF so that it could be defended by an organization with money. Is this their only purpose of existence? Beware of your friends when lots of money is involved.

  • Technology and Judges (Score:3, Insightful)

    by Max the Merciless ( 459901 ) on Thursday February 28, 2002 @12:28AM (#3082412) Homepage
    It seems there is a desperate need for some tech savvy judges. Perhaps a "Court of Technology" should be established.

    The current system seems kinda like slashdotters passing judgement on the latest fashions from Paris.

    • Comment removed (Score:5, Insightful)

      by account_deleted ( 4530225 ) on Thursday February 28, 2002 @12:35AM (#3082433)
      Comment removed based on user account deletion

      • Re:Technology and Judges (Score:3, Interesting)

        by sconeu ( 64226 )
        I have full confidence that the DMCA will be destroyed in court - it just takes time for a good test case to come to light

        What was wrong with MPAA v. 2600? Or the Felten case? Or the Sklyarov/ElcomSoft case (though that has yet to come to trial)?

        In case you don't remember, in the 2600 case, the DMCA came up, and free speech got reamed.

        • Comment removed (Score:5, Interesting)

          by account_deleted ( 4530225 ) on Thursday February 28, 2002 @01:30AM (#3082596)
          Comment removed based on user account deletion
        • It is my opinion the EFF has bungled every attempt it has made to get the DMCA declared unconstitutional. The EFF has poured millions of dollars into various defences and failed to deliver even one decisive action in its favor, even though supporters of the DMCA has failed to prove any loses what so ever to piracy. The problem seems to be is the EFF is playing a defensive game, it is only reacting as issues arise. Most of the time they react badly.

          The RIAA and the MPAA fight dirty, they only go after cases where they know they will win and public opinion will be with them. They have deep pockets and lots of lawyers. But there is a chink in thier armor, when they threatened Felton, the EFF reacted by immeaditly going public, when public opinion turned against the RIAA they backed down, Felton gave his speech and then the EFF lost in court by winning. What should have happened is when Felton recieved the threatening letter he should have forced the RIAA to file the suit by writing the following message back.

          TO: RIAA
          From: Prof. Felton
          Subject: DMCA
          Body: BLOW ME!

          This almost certainly would have produced the desired effect, once the lawsuit was filed, and the rest would have been history. Of course 20-20 hindsight is a wonderful thing.

          Another thing the EFF should consider is going on the offense and filing a class action lawsuit against everyone who has presued prosecution under the DMCA or possibly the US Government. In order to do this real harm must be shown, and this is were the real problem is. I personally consider loss of freedom of speech, freedom of the press and fair use, to be real loss. However in post 9/11 America, I seem to be alone, personal freedom is at an all time low and no one cares. In order to carry out a class action lawsuit, the EFF would need to prove real financial loss on the part of the victims and prove direct profit on the part of the defendent. Not an easy task when the RIAA and the MPAA have already characterized themselves as the victims of Hacker, Pirates, Thieves and dope smoking hippies.
      • Well it seems the Judges and lawyers waste valuable time and a huge amount of money while the judge grapples with understanding the technology involved in each case. Even then there is a risk that judges are making decisions an unsound technological understanding. A speciality court could assume a great deal of knowledge on behalf of judges resulting in quicker, cheaper and more informed decsions.

        How many small time companies can afford to pay lawyers while a judge spends day after day trying to get his or her head around the basics of the technology?

        How open to FUD is the system while judges don't know what they are really dealing with?

        The Supreme Court is a speciality court dealing with constitutional law. The Family Court specialises in family law. This is because they are very complicated areas. In other areas of society we employ specialist professionals (an aeronautical engineer handles different work to a naval engineer or civil engineer), why should the law necessarily be different?

        I'd rather have a judge who knows that Word Perfect and MS Word are different products making decsisions whihc have big ramifications on the Net and technological sector!

        • IIRC, the judge in the Microsoft monopoly case removed the Internet Explorer from Windows (something Microsoft had said was impossible) by himself.
        • Well it seems the Judges and lawyers waste valuable time and a huge amount of money while the judge grapples with understanding the technology involved in each case.

          Simply because a case involves technology does not mean that the details of the technology are relevent. Copyright infringment is not fundermentally different in the case of this kind of GPL infringement from a publisher failing to pay an author or someone hiring a band to play a venue then trying to change the rider after the performance...

          A speciality court could assume a great deal of knowledge on behalf of judges resulting in quicker, cheaper and more informed decsions.

          Or it could be a small clique easily bought off by monied interests...
      • violation of licensing agreement should be enforced the same whether it is MS's EULA.

        The MS EULA isn't a valid licensing agreement. It doesn't even meet the basic requirements of contract law. No court has upheld a EULA as a valid form of contract.

        Max
      • We do not need a special court of technology, for several reasons.

        Nor should we need special laws for specific technologies either. A well written law should cover doing something old with new technology perfectly well. Only if a new technology makes something new possible do you need new laws. N.B. simply making something easier does not make it "new".

        First, the law is The Law. It's interpretation should not change based on the plantiff or its paritcular singular application - a violation of licensing agreement should be enforced the same whether it is MS's EULA or the GPL.

        IMHO It makes more sense to group the GPL with publishing contracts or even contracts for live music performances
      • The judge in this case is ill qualified to sit it, plain and simple. she seemed to think "linking" code was analogous to hyperlinks on the web, and that a single package was like a single icon you would click on on the desktop. This is incompitence. As I wouldn't hire a lawyer who didn't understand the specifics of tech law I shouldn't have to put up with a judge who doesn't understand the specifics of tech law. How can justice be served when we're trying to have sophisticated conversations about what is legal and what is not legal and the judge is sitting there saying "what's a binary?"

  • Irreperable Injury (Score:5, Interesting)

    by bwt ( 68845 ) on Thursday February 28, 2002 @12:34AM (#3082430)
    The judge seemed to weigh that putting NuSphere out of business was a big deal in the "balance of harms" and didn't see clearly what the "irreperable injury" the other way was. As Jack Valenti would say, if Nusphere has based their business model on piracy, then they should be imprisoned, not just slapped with an injunction.

    The GPL is not a gift. It is a barter offer: I allow certain carefully constrained uses of my IP and in return you offer any intellectual property you create some as a result of my good will back to "the pot". In software development, time is a critical element: if one competitor can benefit from the IP of the other and mix in its own IP without disclosure, then the cheater has gained an unfair advantage. If its cheating is part of a business venture, then its entire profit stream is usurped from the good faith party as a direct result of its bad faith dealings.

    For this reason the GPL essentially says "if you ever cheat, you get the death penalty". Without this clause, freeloaders could perpetually use gamesmanship and delaying tactics to obtain competitive advantage in the bartering of "hot" IP. To assure that bad faith cannot usurp the original author's valuable IP contributions, the GPL revokes all rights for the smallest infraction in order to assure that any "negotiation" will occur with the position of strength emphatically given to the original author.

    The irreparable harm inflicted by not issuing an injunction is that NuSphere can continue to profit from a business model that is based on outright theft of intellectual property. The moment they were notified of GPL noncompliance, section 4 of the GPL revoked their licence. They are distributing and indeed modifying a copyrighted work in direct defiance of the authority of the copyright holder with the intent to profit. That is not only irreperable harm, it is stealing, and it is CRIMINAL if it is done willfully (17 USC 504).
    • A small complaint. I don't think that copyright violations should be called stealing or theft. It may be a violation of the license of copyright, but I disagree with calling it stealing.

      Sorry if you disagree. I don't mean to mimimize the impact, but I do disagree with the term.

      By the way, very good comments. I do like the "Death Penalty" analogy. I also the barter analogy. It seems as though many argue that the GPL doesn't put any restrictions on you, but it does. (Perhaps they don't like the word "restrictions" huh? [Grin]) Anyhow, thanks for the comments.

      Cheers!
    • Some people claim that NuSphere is now distributing source with their binaries. If that is true, they may still have lost the right to distribute the software in the long term, but the "irreparable harm" argument for a preliminary injunction seems lessened because it puts them back on equal footing with MySQL right now.
      • If they are shipping MySQL without a licence, even if they are complying with the terms of the licence they would like to have, they are essentially pirating MySQL and basing a business on piracy. The irreparable harm to MySQL is that NuSphere is stealing business based on theft of intelletual property.
    • While you may see this, the judge does not. The problem with the GPL is that it is new for the legal system. And as such they cannot simply say "Oh lets knock off this company because they are illegal".

      Right now legally speaking the company is innocent until proven guilty. (Even if we know otherwise or think otherwise) This is a justice...

      The comparison to piracy is incorrect since that has already been determined to be illegal. Remember how long it took to make hacking and cracking illegal.
      • While you may see this, the judge does not. The problem with the GPL is that it is new for the legal system.

        But copyright isn't new, contracts licencing a third party to make copies of a copyright work (in return for consideration) arn't new either...
    • The GPL is not a gift. It is a barter offer: I allow certain carefully constrained uses of my IP and in return you offer any intellectual property you create some as a result of my good will back to "the pot".

      Also the GPL has a lot more in common with regular contracts than any EULA. Indeed it isn't an EULA in the first place it's a copyright licence...

      To assure that bad faith cannot usurp the original author's valuable IP contributions, the GPL revokes all rights for the smallest infraction in order to assure that any "negotiation" will occur with the position of strength emphatically given to the original author.

      Or if you prefer it ensures that the original author maintains the position of strength which copyright law gave them in the first place.
      Breaking the GPL is conceptually little different from an author licencing a book to a publisher and the publisher failing to pay in accordance with the agreement.

    • While you are essentially correct, copyright infringement is NOT theft. It's copyright infringement. Courts have traditionally interpreted copyright in a different light from fee-simple property rights, so this shouldn't be any big news.

      • Re:Irreperable Injury (Score:3, Insightful)

        by bwt ( 68845 )
        And in every one of the cases, the infringed party has stood up and used words like "theft". It's called advocacy -- use the words that make your case sound strongest.

        By the way, by 17 USC 504(a)(1), if NuSphere's actions were "willful" then they are in fact criminal.
        • Can you explain how their actions are criminal? 17 USC 504(a)(1) says:

          (a) In General. -

          Except as otherwise provided by this title, an infringer of copyright is liable for either -
          • (1)the copyright owner's actual damages and any additional profits of the infringer, as provided by subsection (b); or

          So the way I read that (seems pretty clear), if found guilty, NuSphere is liable for MySQL's actual damages and any money NuSphere made off of MySQL. I don't see how this court-enforceable liability makes NuSphere guilty of criminal acts, whether willfull or not.

          Can you explain?

  • From Kessel's post:

    (1) NuSphere probably figured they could probably violate the GPL and MySQL would not have the resources to sue them. This is probably true of most free software developers who have not assigned their copyrights to the FSF. After watching this hearing, it seems to me imperative that developers assign their copyrights to some party that will have the resources to enforce it (whether the FSF or some other organization). These violations can only become more common as GPL software becomes more powerful and widespread, and it is quite expensive to litigate against them.

    Uh, say what? Does he mean that the FSF has no legal right to assist another party monetarily in courst unless the FSF has the assigned copyright? I somehow don't think legal rules are that restrictive. I read that as, "if you don't give your code to us (the "free" world)", we don't care spit about you if someone hijacks it, GPL or no."

    Please tell me I missed something here...

    • Kessel is exactly right: you cannot sue on someone else's behalf. So FSF cannot, for example, sue NuSphere on MySQL AB's behalf. (They can, however, testify as an expert witness). That is the reason FSF requests the assignment of copyright, contrary to what certain trolls claim.
      • They couldn't assemble a legal team for the original developers ala ACLU or EFF?
      • To expand on another comment at my level...

        The FSF could provide legal counsel at minimal cost to ANYONE they desire. (Remember Paula Jones - she was represented free of charge by a highly partisan group, if I recall correctly)

        You can't sue _George_ for injuring _me_. Only the injured party can *ACTUALLY* sue George. *BUT* you can help me pay for or provide legal council for me to sue George.

        For the FSF to say they can't help is a crock. Perhaps they choose not to, or perhaps they don't have the funds, but they CAN help.

        Cheers!
        • The author(s) could also assign their rights, after the fact to the FSF or anyone else.

          But this brings up another point. If you assign all your rights to the FSF, you wouldn't have the right to then sell your code to some commercial entity (sans any added GPL additions) because you wouldn't have any rights other than those granted under the GPL.

          Would it be possible to grant just the GPL portion of your rights to the FSF, and retain the original rights, so you could take your original code private/commercial/closed. (I don't know much about assigning rights to the FSF, so it might be that this is how it works now, but I suspect not...)

          This might make it unadviseable to assign rights to the FSF if you might want to exercise your right to close or commercialize your software. Again, I do understand that you couldn't take the resulting GPL code with other's changes and close it. But you could with the original.

          Just some thoughts...

          Cheers!
        • You can't sue _George_ for injuring _me_. Only the injured party can *ACTUALLY* sue George.

          The injured party can however appoint someone to act as their behalf. Depending on the context this is known as either "appointing a proxy" or "granting power of attorney".
    • For any given piece of free software, there might be any number of contributors. Getting two dozen authors on four continents to sue together against a GPL violator would be very difficult. If just some of them sue, then we're into the messy area of who owns what part of the project (do the plaintiffs own enough of the project to have standing to sue?).



      If all of these contributors assign to the FSF the copyrights on their respective portions of the project, then it is much easier for the case to proceed, since there is just one plaintiff, and no nitpicking over who owns what part.



      So I don't think that the FSF will refuse to help you if you don't assign them your copyright; it just makes the proceedings much simpler if you do.

  • Clarification on GPL vs LGPL for dynamic vs static (Score:3, Informative)

    by MattW ( 97290 ) <matt@ender.com> on Thursday February 28, 2002 @12:41AM (#3082450) Homepage
    I'm a bit vague now on GPL and LGPL for static vs dynamic linking. Let me note that NuSphere makes themselves sound crooked when they claim that a mail client connecting to a GPL'd mail server would be polluted. That's absurd.

    So, I've always gathered that the LGPL exists to permit a program to link to libraries - but my impression was that the LGPL was phrased such that code could statically link to it -- such that the result would be a single executable, and that even under the GPL, it was legally permissible to write code that would rely upon a dynamically loaded library. Can someone confirm this is the case? IE, you could write a program which required that a person have libmysqlclient.so* installed for dynamic loading and that program could be closed source, but you could NOT produce an executable which statically linked that into the executable unless libmysqlclient.so was distributed under the LGPL (which, of course, it is not).

    • Re:Clarification on GPL vs LGPL for dynamic vs sta (Score:5, Interesting)

      by bwt ( 68845 ) on Thursday February 28, 2002 @12:53AM (#3082481)
      NuSphere is purposefully trying to confuse the issue with their code linking argument. The copyright in source code is completely separate from the copyright in object code. That is old hat.

      Even if the source code is not "integrated", when the result of compiling is an indivisible executable object code, the result is a derivitive work. If I write a library and you use it, the resulting object code is a derivitive of both source files. The GPL specifically mentions and includes this case. The LGPL is offered as an alternative when the harsh consequences of the GPL are not what the author desires. End of story.

      • dynamic vs static linking (Score:3, Interesting)

        by MattW ( 97290 )
        Yes, but dynamic lining does NOT produce indivisible executable object code. I understand that NuSphere was statically linking, so they're in the wrong undoubtedly. But the point I'm raising is that you can dynamically load a shared object (ie, libmysqlclient.so), from a proprietary piece of software, without running afoul of the GPL.

        All your source -> your executable

        All GPL source -> GPL shared library

        When your executable runs, it then loads the GPL shared library in order to have access to the required object code. If it doesn't find it, it has unresolved symbols and craps out. It is, therefore, clearly dependant upon that shared library, but it is, by the same token, clearly NOT a derivative work, since 'your executable' is derived without any GPL source code.

        The static executable under GPL is a clear cut violation. But that leaves static binaries under the LGPL (I'm unsure), the dynamic under the GPL (unsure, think legal), and dynamic under the LGPL (must be legal -- hell, what would the LGPL be good for if it wasn't? :))
        • Yes, but dynamic lining does NOT produce indivisible executable object code.

          Correct, but that's not an issue here, since my understanding is that NuSphere is statically linked.

          If I'm wrong about that, then we need to delve deeper. If executable A can be extended by dynamically linking binary B, then the two executables are A and A+B, both of which are obviously derivitives of A when they are in memory. I also think that the law is pretty well settled that in-memory copies of code are "copies", so that even the A+B case is an integrated copy once its in memory, which would still meet the definition of a derivitive work. Moreover, if B was designed from the outset to only be used with A, then A+B is a derivitive.

          The judge appears to have asked the right question when she asked "can NuSphere's binary run without MySQL installed on the same machine?" Answer No. Conclusion, this is not two separate programs interoperating, but a derivitive form of one program.
        • And here is the kicker, straight from the GPL:

          These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it.

          The only way they can get around not GPLing their software is if it is independant and not distributed as part of a whole.

          In a nutshell this means that the software must not completely depend on the library to function, it must be able to survive having that library removed and still able to operate or else is it not independant but dependant on the library. Also it cannot be distributed with the GPL'd library as part of a whole program which does NOT necessarily mean statically linked but in the same package! Be that tarball or CD-ROM.

          If these two conditions are not met then the software must be bound by the GPL regardless of object or source code, static or dynamic library.

          -- iCEBaLM
        • I was under the impression that Nusphere -Was- dynamically linking. I thought that when Nusphere and MySQL AB were getting along, MySQL AB included the 'hooks' to Nusphere's row level locking functionality. Once these hooks were included in the product, Nusphere could add the transaction support thru dynamic linking, without violating any agreement.

    • I'm a bit vague now on GPL and LGPL for static vs dynamic linking. Let me note that NuSphere makes themselves sound crooked when they claim that a mail client connecting to a GPL'd mail server would be polluted. That's absurd.

      Clearly absurd, and unfortunately typical of NuSphere so far as honesty goes. Which is why I am NOT in any way angry at MySQL AB on this case - if you've followed it at all you know that NuSphere have been arses to them repeatedly, while relying on their code to sustain their business, so it's hard to criticise MySQL for just saying enough's enough, cease and desist you dumb f*ckers.


      So, I've always gathered that the LGPL exists to permit a program to link to libraries - but my impression was that the LGPL was phrased such that code could statically link to it -- such that the result would be a single executable, and that even under the GPL, it was legally permissible to write code that would rely upon a dynamically loaded library. Can someone confirm this is the case? IE, you could write a program which required that a person have libmysqlclient.so* installed for dynamic loading and that program could be closed source, but you could NOT produce an executable which statically linked that into the executable unless libmysqlclient.so was distributed under the LGPL (which, of course, it is not).

      The Lesser GPL is indeed so phrased, and that is why Stallman recommends its use only when a pre-existing proprietary library beat us to the punch so far as functionality goes. If there is a reasonably priced proprietary library that gives you function X, then a GPL library that gives you function X isn't particularly attractive. So in that case it should be LGPL instead.


      As to the rest... I think it's pretty clear to anyone with a small fraction of a clue that a statically linked program is a derivative work. The question remaining is whether a dynamically linked work is a derivative work as well. That's really the only area where FSF interpretations are likely to face any significant court challenge, and the FSF is and has been aware that their view may not prevail from the beginning - their publications make this clear.


      The "official" FSF interpretation, if memory serves, is that dynamically linking a GPL library creates a derivative work. Yeah, that's a period.


      Before the peanut gallery starts raving about the FSF being hypocritical here, let's just remember that this is a standard tactic in any legal or political arena, and the FSF would be selling us all short if they didn't use it.


      Based on legal history, if it came down to the wire a court would be unlikely to uphold that claim in the event of a program which was dynamically linked against a GPL library but would also function if dynamically linked against another library instead. However, if there was no non-gpl library it could be dynamically linked against instead, the courts would likely side with us. Courts may not understand code, but they do understand pretty easily the difference between a work which has no value without X and a work which can work identically with X or Y.


      Even if there is a non-GPL library now, if the work in question was distributed before the non-GPL library was available, the courts would likely call the work a GPL-derivative. Just one more reason to license your library under GPL NOT LGPL unless you are certain your library simply duplicates functions already available.


  • Not About Linking (Score:2, Informative)

    by SpringRevolt ( 1046 )
    Despite Nusphere's comment, the point is that it is *not* about linking (perhaps Nusphere where trying to cloud the issue).

    Nusphere lost the right to redistibute MySQL because they made a modified version without the source code for that version being available: "Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License."

    The MySQL authors therefore have the right to stop Nusphere redistributing their software.

    MySQL AB undoubtedly have their own reasons for being so stiff necked.

  • Amend the GPL (Score:4, Insightful)

    by bwt ( 68845 ) on Thursday February 28, 2002 @01:23AM (#3082574)
    It might be a good idea to add a little bit to the GPL regarding the "death penalty" provision.

    Specifically, I would propose that use of GPL software be conditioned on legal stipulation of the proposition that continued violation of the terms of the GPL after notification is "irreparable injury" because it is theft of intellectual property.

    Similarly, you should have to stipulate that the "balance of harms" of a violation is completely in favor of the author, and that this applies regardless of any business interest that exists as a result of the exploitation of the licence. The author offered a zero-cost good faith licence to use and extend their IP subject only to the condition that IP extentions be returned to the commons on demand. In many cases the GPL'ing author makes this deal believing that it will produce more value than retaining tight proprietary interest and selling the software for its market value. If someone else builds a business by skirting the extremely minimal requirements of the IP barter arrangement, then any revenue stream that results should be viewed by stipulation of the licence as being misappropriated and unjustly earned. Indeed pirated or stolen.

    Thirdly, I would propose that the section 4 "death penalty" provision be made substantially more blunt. In particular, a "take down" process should be invoked (somewhat like the DMCA provision) that says if you recieve a letter of noncompliance from an author that your rights cannot be restored other than by a written order of the author, an arbitrator (see below), or a judge. Some people have noted that if you receive another copy that the licence can be read to re-grant a new and separate licence. All such weasle room should be removed by stating that this doesn't apply if you're licence was revoked for noncompliance.

    Fourthly, I would propose that the licence should designate that any disputes be settled by arbitration by the FSF or by the EFF if the FSF is a party with the loser to bear all reasonable costs incurred by both sides.
    • The GPL was never intended to stiffle development, which is what a "death penalty" would accomplish. Lest we forget, Nusphere is actually ADDING features to MySQL, particularly in the area of complex "transactions".

      As far as I can tell, this entire case is MySQL trying to protect its market share. Nusphere's strategy seems to have been to drag its feet releasing source code to make it a more attractive partner than MySQL. It could offer to support MySQL code, but not vice versa. If you're a firm, your choice of partner is a no brainer.

      The problem is that the GPL was not designed as a commercial license. Firms are not EXPECTED to make money selling code.
    • Specifically, I would propose that use of GPL software be conditioned on legal stipulation of the proposition that continued violation of the terms of the GPL after notification is "irreparable injury" because it is theft of intellectual property.

      Why do you need to ammend the CPL to cover this. Copyright infringement is already against the law. If someone distributes the software in violation of the GPL that is exactly what they are doing.

      Similarly, you should have to stipulate that the "balance of harms" of a violation is completely in favor of the author, and that this applies regardless of any business interest that exists as a result of the exploitation of the licence.

      Again this would appear to be redundant since there is plenty of statute and case law related to copyright infringment.

      The author offered a zero-cost good faith licence to use and extend their IP subject only to the condition that IP extentions be returned to the commons on demand.

      It's not "zero-cost", simply that any cost is not in terms of money. To be valid contracts require transfer of "consideration", this includes money, but also includes barter of goods or services, etc.

      n particular, a "take down" process should be invoked (somewhat like the DMCA provision)

      Why not simply use the provisions of the DMCA?

      All such weasle room should be removed by stating that this doesn't apply if you're licence was revoked for noncompliance.

      THe only thing which might be a useful ammendment.
    • Fourthly, I would propose that the licence should designate that any disputes be settled by arbitration by the FSF or by the EFF if the FSF is a party with the loser to bear all reasonable costs incurred by both sides.

      This would make it unnecessarily difficult to enforce the GPL outside the US (ie, with both parties outside the US). Also, some countries do not have arbitration at all, and it's hard to predict what a judge in such a country would say to such a designation.

      • Good point. That leads to the other obvious missing element: a "forum selection" clause. All disputes should be settled under US law, which does afford for very strong arbitration support.

        Also, there is no reason why such disputes have to be arbitrated with both (or even either) parties physically present.
        • That leads to the other obvious missing element: a "forum selection" clause. All disputes should be settled under US law.

          I see one risk with such a clause. People in other countries could be tempted to release their software under a license in which they replace "US law" by "law of country X" (with X being their own country). This would make it even harder to enforce the GPL.

          On a more personal note, since I come from a country which doesn't (yet) have software patents and a DMCA-like law, I'm quite glad that the GPL doesn't refer to US law specifically. :-)

    • I tend to support this, after all the GPL asks so little. But, I would also add a "reprieve" clause, that is, if complience is restored within some reasonable length of time, say 30 days after notification of non-complience, the license is automaticaly reinstated. In unusual circumstances, a court might order an extention of the reprieve period.
      • Why put anything in the licence that benefits the infringer? Let the primary author have the heavy hand. Most of the time such people have an incentive to respond positively to a good faith effort to stop the infraction (after all they licenced it under the GPL in the first place), but these provisions are needed so that the author can negotiate from a position of dominance with the rare asshole who wants to get away with things.
        • Why put anything in the licence that benefits the infringer.

          Because infringement can be inadvertent and accidental. I'd rather have a "grace" period spelled out, than have to rely on the forgiveness of the license holder. 30 days strikes me as reasonable. Of course, any time you put a specific limit on something, it will be too long for some, and too short for others.

          The recent affidavit before the court is a testament to the fact that the FSF prefers complience to retribution (as, it appears, do most slashdotters). If that is the case, formally giving up the right to vengeful persecution for the slightest inadvertent infraction, is certainly in the correct spirit.

          Of course, this defense should not be available for repeated, intentional, violations opening a 30 day window of opportunity each time (and I suspect that is what you fear). I think that acceptable wording could be drafted to balance heavyhandedness for blatent intentional infringement against grace for temporary lapses, even though reasonable dilligence was exercized.

          The bottom line is that while I like the idea of a GPL "death penalty", I also think that some safeguards be in place so it is not abused.

          Let me give you an example of an inadvertent infringement: I once sent a RedHat-derived CD to someone (with our own GPL code and source) and forgot to include the Red Hat sources. Technically, I was in violation of the GPL even though I already knew they had the relevant sources. They didn't even want me to send them something they already had. However, the GPL did not cover the relationship betweem me and them, but betweem me and the copyright holders so what they wanted was irrelevant: I either had to provide the sources, or a transferrable offer to provide them. I was not prepared to do the latter, because we were not prepared to provide source to any and all who asked for it and didn't obtain binaries directly from us. So, I made sure I sent a source CD with the next update.

          RMS was aware of this slip (we were discussing subtle nuances of GPL complience at the time and the pragmatic problems some requirements caused (hint: making a supportable source distribution is not always easy)) and, IIRC, considered their statement of already having the sources "good enough" to qualify as our meeting the requirement that they have them. However, the FSF was not the only copyright-holder affected, so I made sure to dot the i's and cross the t's as it were and sent sources anyway.

          Should I suffer a "GPL death penalty" because of this gaffe? I don't think so, because (a) I was not grossly negligent, (b) I compensated for the harm caused (sent source), (c) acted in good faith. If the case went to a civil court, I'm sure I'd prevail: civil courts are about restitution, and not revenge.

          Omiting reasonable "forgiveness" clauses in a license leaves the matter entirely up to the courts, and they might be more forgiving than the copyright holder might wish. Putting them in strengthens the copyright holder's position, IMHO, and does not weaken it.

  • I see strong supporters of GPL/L-GPL liscensing jumping on the idea that programmers are protected. What about entire groups of programmers, as the case is here? It's not just John & Jimmy Doe versus Joe Schmoe from XYZ company.

    "...thus likely that she will not actually address the terms of the GPL itself in her decision, which is a good thing, since there was so little expert testimony about it..."

    It's as if open-source software is destined to transcend limits of a hobbyists project and mutate into a corporate-sponsored system. Programmers fill a need with MySQL. Pretty soon, it is MySQL that fills needs within other corporations. Broadcast 2000 video editing software from Heroinewarrior.com was the best bet going. It worked, cost no bones, and featured GPL liscensing. Soon enough, dot bombs pop up and Bcast2k is pulled because of liability concerns. Corporation Soon2be Inc., need video editing and fills that need with Broadcast 2000. Who's liable if a setuid/gid root /usr/local/bcast2000/bcast nukes their production box and months work is lost?

    it's looking scary for hackers who share their toys and get organized / sponsored. pitch your product to company Soon2be Inc., and when they merge with Around4Ever Corp., watch yer back.

  • Fat Lady Not Sung (Score:5, Informative)

    by Euphonious Coward ( 189818 ) on Thursday February 28, 2002 @01:56AM (#3082672)
    This is just a hearing about a preliminary injunction, nothing final. The questions considered are not who's right or wrong, but how much harm is being done leaving by leaving matters alone until the case is over.

    It's clear that allowing NuSphere to continue shipping code they have been shipping for the last seven months won't, by itself, do much more harm, if any, so the GPL issue was left alone.

    Trademarks are another matter -- the more and longer they are abused, the greater the harm to the owner.

    When the case itself goes to court, the text of the GPL will leave the judge little choice: NuSphere's product really is a derived work, and there is lots of case law about derived works, even with software.

    For some background, see my letter to LWN [lwn.net] last year. (Scroll down to the end. Incidentally, it appears I was the first person to tell Monty about this feature of the GPL.) Evidently it took this long to establish that NuSphere just wouldn't figure out where they stand without help from a judge.

    • It's clear that allowing NuSphere to continue shipping code they have been shipping for the last seven months won't, by itself, do much more harm, if any, so the GPL issue was left alone.
      Trademarks are another matter -- the more and longer they are abused, the greater the harm to the owner.

      Usually the same reasoning is applied to copyright infringement too, though.
  • "I'm a free-software publisher who has dealt with similar license issues, even so far as bringing up concerns about the GPLv2 with Bradley Kunz and in turn, Eben Moglen."

    My appologies to Mr. Kuhn. I had met him at the open source convention in San Diego, and spoke with him about the FSF, which he is Vice President of -- the man on the street, if you will. He was extremely helpful and quite expedient in helping me with my GPL'd project and understanding license issues. (Thanks Brad - My bad!)

    Anyway - I wish I had read this before I came down so soundly on MysqlAB's side: http://www.nusphere.com/misc_stuff/declarationofbr ucefwebster.pdf [nusphere.com]

    Why does the name Bruce Webster sound so familiar? He must have argued this GPL issue before?

    Abe

  • Section 4 isn't the end of it (Score:4, Insightful)

    by throx ( 42621 ) on Thursday February 28, 2002 @02:29AM (#3082725) Homepage
    While Section 4 states quite clearly that if you break the GPL (which, let's assume they did for now) then all rights to distribute under the contract you entered into with the license are revoked.

    That's not the end of the story as most people seem to be professing. The real issue comes when someone from NuSphere downloads ANOTHER copy of MySQL including the source, especially if this is a newer or later version or if they obtained it from a different party. Does this new download constitute a new contract between the parties, and if so then all a GPL violater has to do is download a new version and/or download from a different source and wham - new license, new contract, all sins are forgiven.

    Failing that argument, the (rather simple) alternative is to create a new and independant legal entity, assign all your IP rights to the new entity (including the trading name) and the new entity downloads a new copy of the source and is immediately in compliance with the GPL.

    The real danger here is that the GPL may and probably does have a "built in automatic forgiveness" clause. If you break the GPL then it seems likely that bringing yourself into compliance and relicensing is a trivial exercise.

    Note: IANAL... I did see one on TV once though... In Australia... It was a British show... Ahh whatever.
    • Well what they can do is something like this.

      Someone download MySQL X.XX, that is not employed by NuSphere (could be someones brother for that matter). They go though the code and rewrite portions, add stuff, and rename is to XxSQL and distribute to NuSphere.

      I would say that it is legal as long as it is all GPL the entire way. There is NOTHING in the GPL to keep me from renaming a project (AKA CODE FORK).

      Sure it would not be MySQL anymore, but there is no reason why they can't track the protocol and ABI. It is GPL remember?

      How does that one sound?
      BWP
      • My point exactly. The FSF's idea that you need to be "reblessed" once you've sinned is (IMHO) very shaky. I don't see how they can possibly argue anything different when Section 6 clearly states:

        "Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions." (my emphasis added)
  • After reading some of the comments on the audience and the judge's reaction to the open source movement, I can only say:

    If you want to support someone at any future GPL related hearing, please stay home (unless you're going to act like a grownup).

    All we need to do to convince a judge that we're a bunch of kooks is dress like we're silicon valley engineers (or worse), play Quake on our laptops, and snicker when the judge makes technical mistakes. I learned not to be a snot to the authority figure (think: professor) back when I was in college, but apparently, some people never learn.

  • This apears to be a classic case of confusion over concepts of IP.
    THe tradmark issue would apply only to NuSphere attempting to pass their product off as being MySQL. With the GPL issue the relvent matter is copyright violation.
    The first linked article completly muddles the two up. The second article does at least mention that there are at least two separate issues to the lawsuit.
  • I know next to nothing about this case but it seems strange to me that a Swedish company chose to pursue the case in the US courts.

    Does the GPL not hold outside the US? Do judgements in foreign courts not hold any weight in the US? Or is it simply because they had a better chance of winning a trademark case in the US? (or is their Trademark registered in the US?)

    They should have known that a US judge is never going to make a decision that harms a US company to the advantage of a foreign one.

    • One poster makes the point that MySQL AB is actually the defendant, thus they had no choice of venue. However, they could have sued NuSphere independently in Swedish courts, but it would be next to impossible to get enforcement in Boston of a decision by a Swedish court, if they could even get jurisdiction over NuSphere for a trial.

  • Thoughts (Score:2, Interesting)

    by Spazmania ( 174582 )

    From Adam Kessel's Comments:

    [The Judge] seemed to think "linking" code was analogous to hyperlinks on the web

    In the case of dynamic linking, isn't it? I can link your graphic from your web server inside my web page. In both cases I'm telling the software (either a browser or the OS) where to go to get the rest of the relevant information. In the abstract sense, how is dynamic code linking unlike web linking?

    It seemed like she [...] had trouble seeing how [immediate and irreparable injury, loss or damage] was the case

    This is the clincher for pretty much any temporary injunction filed based on the GPL. The GPL requires that no money can be due to authors of the work for its use. No money, no damages. No damages, no injunction without a full blown trial.

    The only way I can see to get around this would be to offer the software under a dual license where the second license required significant remuneration. That way you could argue that if they failed to comply with the GPL then the second license applied, which would mean a breach involving a lot of unpaid cash.

    <Rant On>

    Why are we fighting this fight anyway? If we as software developers choose to give our stuff away for the common good, what does it matter if someone who doesn't share our values manages to make a buck or two off of it? If the Open Source / Free Software doctrines are sound, then such folks are going to lose in the long run anyway. Its not as if there is a shortage of folks out there who are willing to be a part of this without being forced. Not for the better part of a decade.

    Think about it: If Microsoft could swipe the Linux kernel, wrap Windows around it, and sell it without the source, what would the fallout be? Millions of users would see a more stable and secure operating system, hardware manufacturers would have an easier time of providing drivers for Linux (since they would be essentially the same), and Redmond would be financially motivated to keep coming back to the core linux source tree rather than a single-shot fork. Socially, Linux in the public domain would be far more powerful than Linux under the GPL.

    Besides, saying that I "freely" offer you my stuff but you must subscribe to my social values to get it seems a little like a 21st century version of Marxism. Who do we think we are that we should have such power? Wouldn't we be better off to just release in the public domain?

    <Rant Off>

    • Think about it: If Microsoft could swipe the Linux kernel, wrap Windows around it, and sell it without the source, what would the fallout be? Millions of users would see a more stable and secure operating system, hardware manufacturers would have an easier time of providing drivers for Linux (since they would be essentially the same), and Redmond would be financially motivated to keep coming back to the core linux source tree rather than a single-shot fork. Socially, Linux in the public domain would be far more powerful than Linux under the GPL.

      Give me all your money. Socially, I can be more powerful with your money than you so give it all to me.

      So many people miss the idea of the GPL. It's not just about sharing code, but also protecting the individuals who share their code. It makes sure that noone screws me if I'm doing something nice for them.

      It would be horrible if MS stole the Linux source code and never rereleased it. I would never again be able to hack the kernel when I needed or add an unsupported device driver.

      It is not marxism, it's capitalism. The GPL is oriented towards preserving individual rights of a piece of code. Public domain abandons individual rights. I think you are a tad confused here.
    • My short thought on '21st century Marxism' is what's happening with the copyright cartels (MPAA, RIAA). 'You may enjoy our media if you subscribe to our values system - which are Sell Out, Pay per Play, and No Fair Use. Be a Good Little Sheep.'

      But that's totally irrelevant here. :) So please feel free to disregard. Maybe we just can't get away from ideology when we're trying to publish or sell bright ideas.

    • [The Judge] seemed to think "linking" code was analogous to hyperlinks on the web.

      In the case of dynamic linking, isn't it?

      Yes, except the Judge clearly wasn't thinking of things like image tags; she was analogizing the linking of code to the links you click on when you are viewing a web page that take you to other sites.

      By the end of the hearing, I think she understood that linking was more complex than underlined words. She also knew there wasn't time to deal with the issue in this hearing.

    • This is the clincher for pretty much any temporary injunction filed based on the GPL. The GPL requires that no money can be due to authors of the work for its use. No money, no damages. No damages, no injunction without a full blown trial.

      I'm not sure whether that point is valid. Just because I choose not to sell something doesn't mean stealing it, is ok. If I lend a book to a friend, I don't loose ownership of it - he still has to give it back. The same applies if it's not a book, but say a bicycle which I made myself, and which I lend to all my friends freely.

      Courts are quite used to assign values to goods, even if they don't have a price sticker. I doubt they'd be stumped in this case.

    • Free Software can't win so long as we give a 100% subsidy to proprietary competitors. If millions of people are using the GNU System, I can make a good living making work-for-hire improvements that groups of people decide they need. If millions of people are using a proprietary system, they're all at the mercy of its vendor for any customization.
  • The FSF has posted an Affidavit of Eben Moglen on Progress Software vs. MySQL AB Preliminary Injunction Hearing [fsf.org]. It is pretty good reading.

  • Judge Saris rocked the house (Score:3, Informative)

    by jonathanjo ( 415010 ) <jono.fsf@org> on Thursday February 28, 2002 @12:41PM (#3084729) Homepage
    I was in the courtroom. I was very impressed with Judge Patti Saris and the way she handled the cas e. She was the kind of tough-as-nails, heart-of-gold Boston lady I've learned to admire in my years here; and she brought a plain common-sense attitude to the case that no geek could have brought.

    Clearly she didn't know or care that this is considered the big "test case" for this culturally iconic GPL. To her, this was just another license and copyright case. The question of whether the GPL is "valid" never came up; it was assumed to be as valid as any license. She knew, probably from the 200 pages of briefs she complained loudly about having had to read, a bit of the history and significance of the FS/OSS movements (when the MySQL lawyer began a diatribe about the importance of keeping software free, she interjected, "Yeah, yeah, I know, it's like a religious movement, it's Open Source!")

    It was just as clear that she knew nothing about software and made no bones about it. She started out pronouncing the product in question "Mice Quill". (And then she has to deal with the plaintiff calling it "My Ess Cue Ell" and the defendant calling it "My Sequel"!) And, as has been remarked, when the MySQL lawyer used the phrase "single executable file," she interrupted and demanded that he clarify, "Is that like, when I click the icon for Microsoft Word Perfect, just one window comes up?"

    However, like any good judge or lawyer, she was focused on avoiding time-consuming litigation with expert testimony and the like. She saw past all the technical arguments and FS/OSS crusading and recognized this case for what it was: a business relationship gone bad. She stopped in the middle of NuSphere's testimony and asked if the two businesses wanted to work together again. Both sides were taken way aback, there formed these two little clusters of suits as the lawyers and the company reps conferred; each lawyer in turn blustered for a minute about his client's demands and then said yes, they'd love to. (Significantly, this is where the judge got NuSphere to promise to take the EULA off their product.) Judge Saris then got her bailiff to get on the phone and find them a qualified arbitrator *right away* and offered them her courtroom for the rest of the afternoon. In essence, "Will you kids please just get a room and figure this out!"

    It was a bit disturbing that she was so reluctant to see irreparable harm in letting a company abuse the GPL like NuSphere did and get away with it. I think that was mainly her wanting to avoid a long trial unless it proved essential, and making NuSphere quit using MySql would basically shut down NuSphere, and would thus require a long fierce court battle.

    It is important that even cases like this, dealing with highly technical matters, be judged in a regular court by regular legal types with no prior knowledge of the field. Just like a case between a bricklayer and her construction company, or a musician and his record label. Because they need to be judged based on regular law, in proceedings comprehensible to an intelligent but uninformed outsider. If the judge had known the difference between static and dynamic linking, she would have been an industry insider and would have already had a bias one way or the other.

Slashdot Top Deals

Neutrinos have bad breadth.

Close