Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
Open Source The Courts

The GPLv2 Goes To Court 173

Jason Baker writes Despite its importance, the GPLv2 has been the subject of very few court decisions, and virtually all of the most important terms of the GPLv2 have not been interpreted by courts. This lack of court decisions is about to change due to the five interrelated cases arising from a dispute between Versata Software, Inc. and Ameriprise Financial, Inc.. These cases are dealing with four important terms in the GPLv2: 1) What are the remedies for breach of the terms of the GPLv2? 2) What is a "distribution" under the GPLv2 that triggers the obligations under the GPLv2? 3) Does the GPLv2 include a patent license? 4) What type of integration between proprietary code and GPLv2 licensed code will result in creating a "derivative work" and subject such proprietary code to the terms of the GPLv2?
This discussion has been archived. No new comments can be posted.

The GPLv2 Goes To Court

Comments Filter:
  • by Chirs ( 87576 ) on Monday December 15, 2014 @01:00PM (#48602127)

    This would have been the perfect story to get covered there....

    • After she went on a rant about how Psystar was secretly a "M$" plot to destroy the GPL? Honestly I don't see how anybody could take anything she said seriously after that. I guess too many years of SCO turned her from reporter to tinfoil hatter but after that it didn't matter what it was she was gonna play 6 degrees of M$, kind of like how old Twitter could link everything from global warming to asteroid impacts to Bill Gates.
  • ...he may be the originator of the license, but he'll probably not come across well when talking. Think of Cumberbatch's portrayal of Sherlock Holmes in The Reichnbach Fall as an example. Does more harm than good.
    • by jedidiah ( 1196 ) on Monday December 15, 2014 @01:09PM (#48602211) Homepage

      RMS only commissioned the license. He did not create it. The lawyer that actually drafted the license would likely be a much better "witness" assuming that such things would even be considered in this case.

      Amicus briefs are likely the only means of being seen or heard in a case like this.

      • by TWX ( 665546 )
        I hope you're right. I also hope that the significant amount of published literature discussing the nature of the license can count for enough public comment to deflate arguments that the violator didn't know of any other interpretations of the license beyond their own.
      • by gnasher719 ( 869701 ) on Monday December 15, 2014 @01:47PM (#48602583)

        RMS only commissioned the license. He did not create it. The lawyer that actually drafted the license would likely be a much better "witness" assuming that such things would even be considered in this case.

        Even so, what the lawyers who created the license intended doesn't matter. Or what he was told that the license should achieve, doesn't matter either. The text of the license matters.

        • by TWX ( 665546 )
          Are you certain of that? Bear in mind, when interpreting the Constitution of the United States, judges do look at other influencing documents from the time, like The Federalist Papers, which are not themselves legal documents.
          • by Kjella ( 173770 )

            Are you certain of that? Bear in mind, when interpreting the Constitution of the United States, judges do look at other influencing documents from the time, like The Federalist Papers, which are not themselves legal documents.

            True, but ignorance of the law is no defense. Which basically means that not only must you know the text of the law, but the entire applicable body of law, relevant precedents and current interpretation of the law. Heck, you can still end up losing a trial because the Supreme Court will disagree with your reading of an ambiguous and previously unsettled area of law so being a psychic or clairvoyant could be quite useful. They'll try interpreting the law as intended and you bear the burden if they decide you

            • by Teancum ( 67324 )

              In contract law you're not assumed to know anything about the background or history of the license except as written.

              That isn't true either, particularly with standard contracts that are in widespread usage. A good example of such a contract is the standard Screen Actors' Guild contract, which is something frequently that goes to trial and has had nearly every provision tested in one way or another. Legal precedence in particular (previous court cases that have occurred with regards to that contract) are definitely something you would be wise to pay attention to, in particular any cases argued in Los Angeles County.

              Gene

          • by bws111 ( 1216812 )

            The Constitution is basically a contract between the States. It is not unusual in contract law to look outside the contract to resolve ambiguities. They are looking for the so-called 'meeting of the minds' between the parties to determine the real meaning.

            A license is not a contract. It is completely one-sided. and the licensee has no input and there is no meeting of the minds. Therefore, the licensor has all the responsibility for making the license clear. He does not get to go back later and say 'tha

          • by Rich0 ( 548339 )

            Are you certain of that? Bear in mind, when interpreting the Constitution of the United States, judges do look at other influencing documents from the time, like The Federalist Papers, which are not themselves legal documents.

            Sure, but a law is issued by Congress, and by nobody else.

            If I license my software under the GPL and give it to you, then if you want to understand the intent of the license you need to talk to ME and not RMS. After all, _I_ was the one who licensed the software to you. I just used the GPL instead of retyping my own version of it.

            Sure, you can ask what the intent was, but I think the text matters far more in a situation like this.

      • by Teancum ( 67324 )

        RMS actually wrote the version 1 of the license. He did get some legal assistance in terms of crafting the language of the license, and Eben Moglen was frequently involved in those legal discussion including enforcement of the GPL with regards to software owned (aka the primary copyright holder through various agreements) by the Free Software Foundation (primarily GNU related software). Another significant contributor was Mike Godwin, of Godwin's Law fame, although you could also point to several other pe

  • by bogaboga ( 793279 ) on Monday December 15, 2014 @01:10PM (#48602231)

    ...4) What type of integration between proprietary code and GPLv2 licensed code will result in creating a "derivative work" and subject such proprietary code to the terms of the GPLv2?

    It upsets me that this question will be answered by those who [probably] know nothing about software and code. Why won't they ask those who created this GPL? Wouldn't they surely know better?

    • No, the creators will tell you what it was intended to do, the court will tell you what it actually does. Potentially huge difference there.

      • the creators will tell you what it was intended to do, the court will tell you what it actually does.

        No.

        The creators will tell you what it was intended to do and what it actually means within that narrow GPL context.

        That is what is important and that's why we have the so called fine print in all agreements I have come across. No?

        • Re: (Score:3, Informative)

          by BasilBrush ( 643681 )

          If there's fine print, then there's no need for the opinion of the creators. Indeed there's no need for their opinion either way, because if it's not in the license already, in written words, no one is subject to it.

          • by jbolden ( 176878 )

            That's not true. In all contract law there is a concept called "the form of the contract" which means contracts are interpreted inside of a context and that context is included in the contract unless explicitly excluded. For example if X signs a service contract with Y which lists the services by X and payments by Y but never says that X will perform the services or that Y will deliver those payments it is assumed from the form. On the other hand an insurance list is just a list.

        • by itzly ( 3699663 )
          What the creators intended isn't as important as what they wrote down.
          • The court will interpret intent when the language is unclear, but the "plain language" rule from the supreme court will take precedent.

          • by tiberus ( 258517 )

            Sadly too true.

            There is often a vast gap between saying what you mean and meaning what you say.

            It's a real PITA to have someone interpret your words and force you to accept that they don't mean what you intended them to.

        • Problem is, what it was intended to do means sod all when it comes to determining what it actually does - the court won't hold you to the spirit, it will hold you to the letter.

          Which is why the fine print exists in those other agreements. But it doesn't exist here beyond the wording of the GPL itself. So asking the creators of the GPL in this instance will get you nowhere because their opinion on the matter lacks any weight, its what the actual wording says which determines what you are beholden to. They d

          • Which is why the fine print exists in those other agreements.

            Agreed.

            But it doesn't exist here beyond the wording of the GPL itself. So asking the creators of the GPL in this instance will get you nowhere because their opinion on the matter lacks any weight, its what the actual wording says which determines what you are beholden to.

            So let's agitate for fine print in the GPL to aviod any ambiguity.

          • by bill_mcgonigle ( 4333 ) * on Monday December 15, 2014 @02:11PM (#48602829) Homepage Journal

            So asking the creators of the GPL in this instance will get you nowhere because their opinion on the matter lacks any weight, its what the actual wording says which determines what you are beholden to.

            Most prose can be interpreted in multiple ways and not every interpretation occurs to every human at every time. Courts are well aware of this, which is why they will only ever offer an Opinion about what things mean - never claiming to offer the Truth. Even SCOTUS only offers opinions.

            Now, those courts will also issue orders to men with a violent streak to enforce their opinions, so effectively they are Law. But never Truth, which is why subsequent cases can overturn previous ones. This also means that Law is never Truth, only the prevailing view of the status quo of a given time.

          • by gnasher719 ( 869701 ) on Monday December 15, 2014 @02:19PM (#48602923)
            Now all that said, from all I've heard the authors of the GPL were quite competent in what they were doing, so it is quite likely that the GPL says what they meant.

            I think the most important thing is about remedies, and there is quite strong precedence. It seems that if you use open-sourced software and don't do what you are supposed to do according to the license, you are a copyright infringer, and not someone in breach of contract. So the damages are those of a copyright infringer (up to $150,000 per work, or the proven damage), and not those of a person in breach of contract.
          • It doesn't matter what the aurthor of the GPL meant, it's what the licensor actually said that counts

            • "It doesn't matter what the aurthor of the GPL meant, it's what the licensor actually said that counts"

              No, it isn't. What the licensor actually "said" is the text of the GPL itself, since that's how the work is licensed.

              But then, since it is a license, a kind of contract that you don't get the chance to negotiate, it is not the interpretation of the licensor the one that counts, but the interpretation of the licensee, within reasonable bounds.

              • Not necessarily, anyone can contact the software authors and negotiate license terms; it can become more difficult as the number of authors increase and probably exponentially more difficult. MySQL for an example had multiple licensening options, one was the GPL.
                The reason what licensor actually said that counts is because the licensor is the entity that initiates legal proceedings to remedy perceived violations of the licence. Any way the terms of the GPL are essentially a non-concern until you distribute.

                • "Not necessarily, anyone can contact the software authors"

                  Key word being "can" which is neither "should" nor "must". The user either contacts for renegotiation/clarification of the license, in which case there will be a paper track and it will be the paper what counts, or the user won't contact, in which case, it will be the paper what counts.

                  In neither case the licensor will have the chance to further explain his intent to the court.

                  "The reason what licensor actually said that counts is because the licens

          • It will hold you to the letter until there's something vague and in need of interpretation. Then the court will consult the original party or parties as to the intended spirit of that portion and take it into consideration.

        • The creators will tell you what it was intended to do and what it actually means within that narrow GPL context.

          No, they won't.

          What the creators will tell you amounts to "if we were in charge, these would be the rules of the GPL and what it means". They can tell you their intent, but intent may not equate to any legal weight.

          The problem is that ignores the rest of the legal system and copyright law which gives the GPL its teeth. For that you need an actual court to rule.

          Nothing at all which is told to you

          • by Teancum ( 67324 )

            Except there are judicial opinions and legal precedence with regards to the GPL. This isn't 1990 when there are six websites on the internet and the GPL was brand new with everybody speculating about what a judge might say about the GPL. SCO v. IBM is one example of a set of legal opinions written (in this case from federal courts) by actual judges where the GPL was a central point of contention and real legal precedence can be found.

            Fortunately the GPL was written to be understood by those who were using

        • "The creators will tell you what it was intended to do and what it actually means within that narrow GPL context."

          Too late. The contract is already signed and the license agreed upon.

          "That is what is important and that's why we have the so called fine print in all agreements I have come across. No?"

          No. It is the perfect opposite: the "fine print" is there exactly because you are *not* going to ask the text author about what was the intended meaning. The meaning is what the text says and no more.

          Since it

    • Slashdot (Score:2, Funny)

      by Anonymous Coward

      No, no, no. What they should have done is submitted it to Slashdot and read the posts to find out what the GPL means and what they should do.

    • by itzly ( 3699663 )
      The concept of derivative work itself is not too complicated, and they'll invite experts to testify on the details.
    • by gstoddart ( 321705 ) on Monday December 15, 2014 @01:23PM (#48602357) Homepage

      It upsets me that this question will be answered by those who [probably] know nothing about software and code. Why won't they ask those who created this GPL? Wouldn't they surely know better?

      Because, the GPL is essentially a contract written in the framework of contract law, which says "you have a copyright exemption under the following circumstances".

      There's been tons of confusion about what it actually means, and if it legally means what people think it means.

      Until a court actually rules on this, everything else is an opinion based on someone's interpretation.

      But, from there, if it stands and the court says "this is the impact", then we'll know and there will be legal precedent.

      It's within the realm of the possible the court could invalidate the whole damned thing. And the court could also provide an interpretation which narrows the scope of it. The court could also expand it.

      So, it may upset you, but the foundation of the legal system is more or less until a judge rules on it, and until there is a legal precedent ... you don't really know if it holds water or not.

      Those who created the GPL may not know as much as they think about writing a software license. Or, they knew an awful lot and the court will agree.

      At the very least, this should remove some of the ambiguity and confusion.

      • by linuxrocks123 ( 905424 ) on Monday December 15, 2014 @04:16PM (#48604359) Homepage Journal

        So, it may upset you, but the foundation of the legal system is more or less until a judge rules on it, and until there is a legal precedent ... you don't really know if it holds water or not.

        Bullshit. The entire point of having a legal system based on written law is so that people know what the law is without having to just try things and then see if the executive arrests them. There are places in the law that are rough and where you really don't know what a judge will do -- "new areas of the law" -- but, in most cases, you do know what a judge will do, because of statute and precedent in similar cases. This certainty is what gives the law its value.

        The GPL is a fairly simple document. It's pretty clear what it means, so we really don't need a judge to tell us. This court case might clear up a few corner cases about the consequences of infringement (forced relicensing or simple injunction + damages), but it is effectively impossible the entire document will be held null and void. There's enough precedent that it is possible to conditionally license a copyrighted work that the GPL's general validity is not in doubt.

        • I have long dreamed of inventing a new crime.

          God damn computer fraud and abuse act makes anything a federal judge doesn't like retroactively illegal.

          • There's a law review article I read somewhere arguing that unless federal courts narrow the scope of that law, it will eventually be found void for vagueness. It is very poorly written.

    • There is the original intent from the creators, then the interpretation of the users.
      Being that it is a written document, the intent is not really binding, but the letter of the document needs to show that interpretation is wrong.

    • by Kjella ( 173770 ) on Monday December 15, 2014 @01:26PM (#48602405) Homepage

      Well, that part is actually not up to the GPL to define it's a key part of copyright law, if it's not derivative it's not covered by copyright so the GPL wouldn't apply.

      • Well, that part is actually not up to the GPL to define it's a key part of copyright law, if it's not derivative it's not covered by copyright so the GPL wouldn't apply.

        Well... Copyright law is about different things, mostly about copying and creating derivative works. A judge would decide whether someone made a copy, or whether someone created a derivative work, according to the law, and the GPL cannot override this. With that decided, the judge will then conclude that either you did need permission by the GPL, or you didn't need permission by the GPL to do what you did.

        On the other hand, the GPL can say in its own terms what it allows you to do. They could say "you ar

    • Answer to #3 is clearly NO - that's why GPLv3 was created, which pissed off a large number of companies, and drove them to BSDL solutions, like LLVM/Clang instead of GCC. For #1, the remedy is that they either stop distributing it, or re-license the combination under GPLv2 and distribute it under THOSE terms.
      • by caseih ( 160668 ) on Monday December 15, 2014 @01:59PM (#48602713)

        No there's a third remedy. Negotiate with the copyright holder for more suitable terms of license. So to summarize, there are three remedies available:
        1. Cease distributing the offending code (replace it, or withdraw the software from distribution entirely)
        2. Re-license the derivative work under compatible terms (essentially release the entire work as GPLv2)
        3. Buy a suitable license for the copyrighted code under terms compatible with the needs of the derivative work.

        Note that #3 is impossible for some projects, as all copyright holders must agree. For some large projects this is very difficult, especially when some contributors can no longer be found. The Linux Kernel is one example.

        For one-man shows, this is why I always recommend the GPL for released code. This allows you the option to sell commercial licenses should your code prove popular. And you still have the option to add additional open source licenses as others request. Releasing code under a permissive license, such as the BSD or even MIT, shuts the door on some of these options. You can always relicense your own code, but you can't recall code you've previously released under another license.

        • It doesn't even need to be an exclusive copyright. This worked out pretty well for MySQL. They made sure to get at least shared copyright for all contributed code so they could release and licence commercial versions.

        • While I mostly agree, I still see permissive licensed code getting support contracts for the same amounts as the license would cost. For the most part, licensed code is purchased for the support. Especially where other free code exists already, using a permissive license can get your code used, and having code that is used is the first step in selling support contracts to some percent of users.

      • Answer to #3 is clearly NO - that's why GPLv3 was created, which pissed off a large number of companies, and drove them to BSDL solutions, like LLVM/Clang instead of GCC

        Unspecified patent claims on software should be an anathema to BSDL software, given that could result in the patent holder controlling any and all usage of the project despite the license.

        Of course, if this is true then it shows how vile said corporations are. So desperate to undermine the core points of Free Software for the sake of power o

    • by fche ( 36607 )

      The authors of a license, even the GPL, cannot simply redefine the pre-established legal concept of "derivation" (absence of which implies irrelevance of license, no matter what the licensor prefers).

      • So then let's have some fine-print added to the GPL. That way we are able to have legal terms define what we actually mean.

        • by fche ( 36607 )

          Well sure, but the point is that it's not up to the whims of the license or licensor what those legal terms mean. They exist outside.

    • If you were going to ask a "someone" how they meant to define "derived work", you would ask Congress, not the author(s) of one out of a million contracts which happen to make use of that term.

      You're right that it's upsetting that (mostly) people who don't really work with copyright would end up answering it, but that's the nature of law, or at least until you start electing[/appointing/etc] authors. (Cynic: or until those people start funding election campaigns.)

      It's only after you have determined that some

    • by jbolden ( 176878 )

      Either side could call Stallman as a witness. I can't imagine that some percentage of the plethora of materials explaining what a derived work is and what is meant that has been written by open source advocates over the past 3 decades won't make its way into evidence.

  • AWESOME! (Score:5, Interesting)

    by Anonymous Coward on Monday December 15, 2014 @01:11PM (#48602239)

    Versata is essentially a "vulture" company that buys software companies that are down to life-support levels to extract the last bits of value (continuing support contracts, etc.). If you ever see them knocking on your door, find another job ASAP because your company (and your job) is a gonner. Interestingly, one thing that companies do as part of the due diligence when acquiring other companies is determine if there is any open source in the products in order to identify potential liabilities. Btw, companies looking for a buyout often go through a phase to clear out potential open source infringements, so that's a sign if it wasn't clearly spelt out..

    I'm guessing that in this case Infosys, a contractor for Ameriprise, probably got paid to develop some software and made it in the form of a library to be used in other contracts. Having outsourced to India before, I was pretty sure that was going on (that and liberal use of open source). After all, rapid development doesn't happen magically.

    • by rgbscan ( 321794 )

      I did a stint at Ameriprise in the early 2k's. They let most of us go and farmed out the jobs to IBM when American Express divested us. We used to be a division called "American Express Financial Advisors" (or AEFA internally - which was formed from the remnants of IDS Advisors for you Minnesotans familiar with the IDS tower) but when they spun us off into Ameriprise all of IT was axed. Who IBM got to do all the tech work, I don't know. Maybe it was Infosys or maybe they came later. It was all foreign contr

      • I thought that AEFA spun off after it got H&R Block's Investment division - I was an H&R Block customer, and became an AEFA one automatically as a result of that merger. What remained of H&R was the tax advisory arm.
  • Programming Language (Score:4, Interesting)

    by wisnoskij ( 1206448 ) on Monday December 15, 2014 @01:16PM (#48602295) Homepage
    I wonder if it is possible to create a communication language that has the reproducibility and stability of a programming language. There is hardly even any point in writing out legal rights or documents when everytime it is read it can be interpreted differently. It seems that the law has taken the stance that, since reinterpreting it every time is a hassle and inconsistent, we will just use the interpretation of whoever read it first. But since we have no accurate way of storing legal ideas, how do you interpret the interpretation accurately?
    • by oodaloop ( 1229816 ) on Monday December 15, 2014 @01:24PM (#48602383)
      I think you're referring to lojban [wikipedia.org].
      • Re: (Score:2, Interesting)

        by Anonymous Coward

        I view Lojban as fundamentally flawed for general human communication. But it would be excellent for unambiguous technical specifications and legal documents.

        Latin, at one time, was used in law for essentially the same reason. A modern equivalent would be using a controlled language [wikipedia.org] that follows very careful definitions. The common law system, as used in the US, does limit the meaning of many words that have legal significance, but it's an ad hoc system that constantly evolves.

    • I take it you've never seen people arguing about what exactly the C standard means about how "volatile" should behave, or whether the defined memory model is sufficient to reason formally about visibility of variables given specific types of assembly operations, or what optimizations a compiler can legally make (as opposed to what optimizations it would actually make *sense* for it to make).

      Even a reasonably-well-defined language like C can still end up in the weeds once you start looking at edge cases...

      • Because the C spec is written in English. So of course it has the same problem as even the most expertly written law. But the actual interpreter (aka the compiler) is not. Once you have a working compiler, the code you write for it will be interpreted correctly and the same way everytime.
        • How do you let people know what the compiler does? Unless there's a human-readable spec, people can't plan for future code/contracts. Writing random code/contracts and seeing if it "compiles" is not a great way to program/negotiate.

          Doing the same thing every time is only a prerequisite for being correct. What if most people don't like what the current compiler does? After editing, how do you let people know what changed? How do you even know the compiler is correct without a human-verifiable document of exp

        • "Once you have a working compiler, the code you write for it will be interpreted correctly and the same way everytime."

          Rrrrright.

          And then, what's the compiler we should use? GCC, Turbo C, Intel's?

          You may tell "whatever is the first one that came to happen" but then, what would be the difference with common law using precedents?

          • by Rich0 ( 548339 )

            "Once you have a working compiler, the code you write for it will be interpreted correctly and the same way everytime."

            Rrrrright.

            Well, it is one way to eliminate compiler bugs. By definition the compiler doesn't have any.

    • by bws111 ( 1216812 )

      This is an idea that is often posted on here, and it never makes sense to me. What makes you think a programming language would do any better with these kinds of questions? There are basically an infinite number of variables - how do you write a program that handles all those variables, especially when some variables or values were not even known to exist (or were otherwise not considered) when the program (law) was written? The best you could do is throw an exception when the set of variables and values

    • Problem is, people aren't identical. Take 1000 Macbooks, run the same code on them, and you'll get (almost always) identical results. You won't get the same asking 1000 people to interpret a law or contract.

      Lawyers (at least good ones) attempt to deal with this issue by being as precise and comprehensive as possible. Often, they're derided for creating "1000 page contracts in legalese instead of a one page agreement," but 999 of those pages, and the legalese, are usually efforts to explicitly deal with th

      • But you can transfer that codebase to a thousand different aystems, windows, Linux, anything you want, you could even run it on a decemal based computer instead of digital. The program as it is written in an language would run the same every time.
        • by bws111 ( 1216812 )

          It will run the same every time, right up until the point where a new unexpected input is introduced. What then?

        • "you can transfer that codebase to a thousand different aystems, windows, Linux, anything you want, you could even run it on a decemal based computer instead of digital. The program as it is written in an language would run the same every time."

          Except, well, you know, it doesn't.

          I don't know where your computer expertise comes from, but you can bet is not as exhaustive as you think it is.

    • "I wonder if it is possible to create a communication language that has the reproducibility and stability of a programming language."

      You have never faced a bug or a divergence between compilers, do you?

  • by Danathar ( 267989 ) on Monday December 15, 2014 @01:21PM (#48602345) Journal

    From the Article:

    "3. Does the GPLv2 include a patent license?"

    No. And THAT is why we have the GPLv3, to head off patent problems (among other things)

    • by TheRaven64 ( 641858 ) on Monday December 15, 2014 @01:23PM (#48602371) Journal
      Well, sort of. Clause 7 could be interpreted as a patent license, in that if you knowingly distribute code that violates your patents then you are violating the license if you don't also include a patent grant. In v3 it's more explicit precisely because it was ambiguous in v2. It's up to the court to decide whether this ambiguous license is a license.
      • If you distribute a license to use for these terms and you control the patent on them you have an implied license to do so and all it will take is a promissory estoppel motion to squash that claim.

        The entire foundation of contract law would come crashing down if you could write a contract to use some thing you have a patent on (which you didn't mention) but then make them take out a separate patent license. The only way you could do that is explicitly mention in the contract to use that the patent isn't inc

        • If you distribute a license to use for these terms and you control the patent on them you have an implied license to do so and all it will take is a promissory estoppel motion to squash that claim.

          I really wouldn't think so. Sure, it has to be made clear that the license for the software doesn't include the license for the patents. But it would make absolute sense if a patent holder made it easy for companies to actually use their patent by providing open source software to use the patent, instead of everyone having to create their own proprietary software.

          • by Qzukk ( 229616 ) on Monday December 15, 2014 @02:26PM (#48603003) Journal

            it would make absolute sense

            Now back up a second there and consider proprietary software. Imagine if you bought Microsoft Office for your company and a year later Microsoft comes along and starts threatening you with patent infringement lawsuits over your use of their patents. I can't imagine that a court would stand for that at all without at least a fine print "requires additional patent license" and even then there's fitness requirements that the court would have at least a little discussion on.

            Why would it be different for something given to you for free with explicit permission to use and give away? If the patent holder wants to provide a reference implementation to licensors, they can do so with a license tied to their patent, but it should not be the GPL.

            • by jbolden ( 176878 )

              FWIW I've bought software before where the contract has both a license agreement and a limited patent license when the software is designed to throw me into violation of their patents. I've never heard of a company arguing they can sell the one and not the other while claiming both however.

    • No. And THAT is why we have the GPLv3, to head off patent problems (among other things)

      Great, at this rate it should be, what, another 15-20 years before that gets tested in court?

      • by Aighearach ( 97333 ) on Monday December 15, 2014 @03:03PM (#48603447) Homepage

        Contrary to the FUD you may read on the internet, the GPL does not need to be "tested in court" and will not be by these cases. You can't challenge a license you're not a party to; you can't sue to harm yourself; if the Court throws out the contract, it cannot write a new one. So you can't be using GPL code and then challenge the license. If you proved the license to be fatally deficient, you'd lose your right to use the code; you'd be harmed by your own court action, and you'd be the only loser. So the Court wouldn't even let you argue that; it is not a valid basis for a case.

        You can fight over the smaller details of a license and what it means, but you can't actually attack the license here. Because it is free, you can't claim to have been misled, or been a victim of an unfair business practice. So you can only argue the edge cases and how they effect you; you can't both have standing, and also challenge the validity of the license.

        It is mostly just misrepresenting the cases that conflates them with having to do with the GPL. These are cases that revolve around business practices and contracts between companies, where there are issues related to their partnerships and business practices related to each other. It has nothing to do with the GPL itself, except in weighing the reasonable expectations of the different parties. If the rulings go one way or the other, it won't affect the GPL or companies using the GPL. It will only affect companies that engage in whatever practices are found to be unfair or harmful.

        In the cases here, there are three companies involved; company A licensed software from company B under proprietary terms. That license is actually the main one involved in the 5 cases. Company A was allowed to use 3rd party contractors to edit the code, subject to terms. Company C (the 3d party contractor) is accused by Company B of having use Company B's code in a competing project. Company B also accuses Company A of not terminating their contract with Company C when they learned they were in violation of the agreement. Company A accuses Company B of having used GPL code in the code they licensed, and therefore that they didn't have the right to license it, and that the entire codebase is a derivative work of that GPL code.

        So the GPL is "involved" in the sense that the Court has to decide if this code is licensed rightfully under the proprietary license, and/or the GPL. Depending on the answer to that question, some number of these companies may be found to have been naughty, and be made to pay. But the GPL is not being questioned here; and it won't ever be in the sense that people always meant by "testing it in court."

        • by EvilSS ( 557649 )
          Actually there are four companies involved, plus a group of companies other than "Company A" that also use "Company B's" software. That would be the company (Company D?) that wrote the the GPL2 licensed code to begin with, and they are the plaintiff in most of the cases talked about in the article.
        • by bmo ( 77928 )

          There was one direct attack at the GPL that might've had teeth had it not occurred in the fetid imagination of a certain Daniel Wallace.

          Dan Wallace tried to get the GPL considered invalid because it amounted to price fixing and a Sherman Act violation. He claimed the harm was that the Free and free properties of Linux operating systems locked him out of the market, even though he didn't actually have a product to market.

          He was duly struck down hard by a de novo appellate court decision.

          That was probably th

  • by bill_mcgonigle ( 4333 ) * on Monday December 15, 2014 @03:03PM (#48603461) Homepage Journal

    1) What are the remedies for breach of the terms of the GPLv2?

    This one is easy - if there's a breach then the license is void and Copyright is the effective law. Code was copied without permission, which becomes a copyright violation, and remedies are already established for that.

    GPL is entirely based on the teeth of copyright - almost every OSI license is. If you hate imaginary property then you might question your use of licenses that depend on it.

  • I'd note that the 3 points at the end of the article aren't unique to open-source software but apply to all third-party software you use in building your software. And those points are harder to address for proprietary third-party software than for open-source, because any software component may contain other components you aren't directly aware of and without the source code it's a lot harder to scan proprietary libraries to detect those included components (and it may be impossible if the included compone

  • Is fruit salad a derivative work of an apple? I believe that the creators of GPLv2 would say yes, but I don't think it is unreasonable to disagree. A lot hinges on the definition of 'derivative'.

Your program is sick! Shoot it and put it out of its memory.

Working...