Do we still need the GPL?

Do we still need the GPL?

Summary: Here's a good topic for debate as we approach Independence Day. As I was doing my morning reading, I came across a link to an interview with Eric S.

SHARE:
TOPICS: Open Source
15

Here's a good topic for debate as we approach Independence Day. As I was doing my morning reading, I came across a link to an interview with Eric S. Raymond, where he says "we don't need the GPL anymore." That certainly caught my attention. Here's part of what Raymond had to say about the GPL:

It's 2005, not 1985. We've learned a lot in the last 20 years. The fears that originally led to the reciprocity stuff in GPL are nowadays, at least in my opinion, baseless. People who do what the GPL tries to prevent (e.g., closed source forks of open source projects) wind up injuring only themselves. They trap themselves unto competing with a small in-house development group against the much larger one in the parent open source project, and failing.

...The pros and cons of "viral licensing" is something I've been thinking about a lot recently. As far back as 1998, I suspected that allegiance to the GPL is actually evidence that open source developers don't really believe their own story. That is, if we really believe that open source is a superior system of production, and therefore that it will drive out closed source in a free market, then why do we think we need infectious licensing? What do we think we gain by punishing defectors?

Stronger virality punishes defectors more effectively, but also has more tendency to scare people away from joining the open source community in the first place. Where the optimum point is all depends on how important punishing defectors really is relative to the economic pressures in favor of open source. My current belief is that the free market will do quite a good job of punishing defectors on its own; thus, increasing virality is a bad move.

There are a couple of problems with Raymond's logic, at least from where I'm sitting. First of all, the argument that "open source developers don't really believe their own story," is based on the assumption that the "story" those developers are trying to tell is Raymond's own "Cathedral and the Bazaar," rather than (for example) espousing Richard Stallman's philosophy that "software should not have owners." Indeed by casting the GPL as merely a business tool, Raymond is highlighting the difference between free software and open source that Stallman has been complaining about for years.

We also have to be realistic, and admit that open source and free software are not really competing in a "free market." FOSS (Free and open source software) competes in a market dominated by entrenched players and monopolies that could easily co-opt open source software as it reaches maturity without most of their customers even realizing the origin of the software.

And, as Stallman writes, the biggest problem that many business types have with free software is that it implies more than business issues:

The main argument for the term "open source software" is that "free software" makes some people uneasy. That's true: talking about freedom, about ethical issues, about responsibilities as well as convenience, is asking people to think about things they might rather ignore. This can trigger discomfort, and some people may reject the idea for that. It does not follow that society would be better off if we stop talking about these things.

Years ago, free software developers noticed this discomfort reaction, and some started exploring an approach for avoiding it. They figured that by keeping quiet about ethics and freedom, and talking only about the immediate practical benefits of certain free software, they might be able to "sell" the software more effectively to certain users, especially business. The term "open source" is offered as a way of doing more of this--a way to be "more acceptable to business." The views and values of the Open Source movement stem from this decision.

...At present, we have plenty of "keep quiet", but not enough freedom talk. Most people involved with free software say little about freedom--usually because they seek to be "more acceptable to business." Software distributors especially show this pattern.

The either/or proposition put forward by Raymond, either developers "believe" in open source as a means of production and therefore can trust the free market to work its magic without the GPL, or they choose the GPL because the open source development method is inferior, is deeply flawed. It presumes that we all share a specific view of the world that many developers do not subscribe to.

Whether we need the GPL or not depends on whether one believes that ethical considerations, or the concept of "free software," are important. From Raymond's point of view, perhaps they're not. However, Raymond's view isn't the only one present in the FOSS community. Look at the Debian project. Yes, Debian GNU/Linux developers are interested in making the software suitable for business -- but without compromising the ethical ideals in the Debian Free Software Guidelines (DSFG).

So, do we still need the GPL? I believe we do. In many ways, Raymond's claim that we don't need the GPL anymore is much like claiming we don't need the Bill of Rights anymore because the free market will ensure that everyone gets a fair shake.

Since Raymond is an avowed fan of guns, I find it particularly odd that he's so quick to dismiss the importance of protecting freedom -- especially since Raymond writes on the Geeks with Guns page, "Geeks and guns are a natural match. Open-source software is about getting freedom; personal firearms are about keeping it." As we approach Independence Day here in the U.S. I'll leave you with that thought -- the GPL is about much more than a software development philosophy: It's about freedom, and we do still need it.

Topic: Open Source

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

15 comments
Log in or register to join the discussion
  • Inclusion vs Forking

    The question is: Does the GPL do more harm than good for FOSS?

    In considering this, it's important to remember that the viral nature of the GPL is not there to simply prevent development of a closed-source fork. It's there to outright block non-GPL developers from including the GPLed software inside there own.

    The LGPL effectively prevents the forking, because it doesn't allow someone to redistribute the LGPLed software under a different license. However, it does allow the LGPLed software to be included in works not licensed under the (L)GPL.

    I would say we need the LGPL, because it's important to prevent closed-source forks of open-source software. But the GPL is based on the idea that closed-sourced software vendors are the enemy. I think this is an excessively paranoid view.
    Erik1234
    • Does the GPL do more harm than good for FOSS?

      Short answer: yes.


      Longer answer: the GPL encourages efficiency in a way that "non-viral" licences don't. This reason for this lies in how proprietry licences are themselves viral in a manner that is a direct mirror to that induced by the GPL: derivative works of proprietry software are themselves proprietry.

      Accordingly, the GPL induces a clean "speciation" between different development models, the result of which is that the more efficient model gets the chance to prove itself, and insofar as the production of free software is efficient (eg. people are motivated to produce it and have the tools and means to develop it, such as a body of existing code), it gets to prevail over proprietry code.

      If the trade of receiving others' code freely and building upon it subject to a condition of reciprocity works for you, then to produce more code in the same vein is to make more prevalent the code-base and development methodology that helped you succeed. Thus, the next developer is in a position to choose the more efficient methodology, which wouldn't have been the case if you were able to close-source it, and you decided that you wouldn't reciprocate.

      In addition for this, there are incentives to produce "first generation" free software in the first place, ranging from idealistic or simple generosity, selling something else (such as hardware), wanting to publish a proof of concept, wanting to receive in kind rather than cash...

      I believe that my "efficiency" argument is appropriate for those wishing to be generous or wishing to receive in kind. If your motive is generosity, you wish to make the work a better place, thus encouraging efficiency through "speciation" can help achieve that; if you wish to receive benefits in kind, your interest as a coder is likely to be more in the form of code rather than a commercial end product, so the route that maximises the pool of free software is the one that you're more likely to wish to take. As it can frequently be worthwhile for someone to produce free software derivatives, even if it would be more worthwhile to them to close-source it, it would appear that a "viral" licence is likely to produce more in the way of benefits in kind than a "non-viral" licence would.
      Morosoph
      • Corrections

        Short answer should be "no": I parsed the question too quickly ("good than harm" => 'yes').

        "If your motive is generosity, you wish to make the work a better place," should instead read as follows:

        "If your motive is generosity, you wish to make the _world_ a better place,"
        Morosoph
  • Who is "we?"

    The GPL is primarily an authors' license; it ensures that an author isn't pouring hir efforts into the sand.

    The BSDish are primarily a consumers' license; it's the license that [b]I[/b] want [b]you[/b] to write under.

    BSD is great for academic work. Not surprisingly, that's where it came from. What's not obvious is why anyone outside of academe would choose to publish under it (barring ideologues, of course. Not that I have anything but the greatest respect for ideologues, mind.)

    Here's the key: for all of the "I don't want to give my preciousssssss work away" rhetoric, it's actually commercially beneficial to have your work merged into the main tree, because it (a) cuts your maintenance work, and (b) gets more quality review. With [u]enough[/u] developers working on the codebase, this suffices to keep the project viable. The x.org, Mozilla, etc. projects demonstrate this.

    The math is the same as for critical mass in nuclear fission: you can sustain a reaction either with a large mass and a small probability of neutron recapture or with a smaller mass and a higher probability of neutron recapture.

    GPL projects have a higher "enhancement recapture" probability than BSD projects, and can therefore be sustainable for smaller projects. What's not so clear is whether (a) BSDish licenses are sufficient for smaller projects, and (b) whether we, collectively, are in turn sustainable without the smaller projects.
    Yagotta B. Kidding
    • I see these assumtions all the time....

      [i]"BSD is great for academic work. Not surprisingly, that's where it came from. What's not obvious is why anyone outside of academe would choose to publish under it (barring ideologues, of course."[/i]

      The assumtion that BSD licensed open source projects are few and far between is something I see all the time. Besides the obvious ones (the *BSD projects), Apache, Perl, Python, Php, bind, X, postgresql, and sendmail are all licensed under "BSD-ish" licenses - that is licenses that allow redistribution, and don't put forth the restrictions of the GPL. MySQL is licensed under the GPL, but is also licensed under an alternate non-gpl license.

      What's common about all of these non-gpl'd projects? They are all in wide use, and the dreaded forking that GPL advocates FUD about all day has not happened.

      [i]GPL projects have a higher "enhancement recapture" probability than BSD projects, and can therefore be sustainable for smaller projects. What's not so clear is whether (a) BSDish licenses are sufficient for smaller projects, and (b) whether we, collectively, are in turn sustainable without the smaller projects.[/i]

      I'm not sure if this is really true, or not. BSD style OS licenses have shown time and time again to lead to wider adoption, which increases the number of potential contributors. Of course adoption requires that the original code woth a crap to begin with.

      Instead of the license, I believe initial code quality, and usefullness is the most important factor in the begining of a project's lifetime, and the choice of license does not factor in until far down the line.

      Starting with php 4.0, php group switched from the GPL license to a Apache-style license. I'm sure all of the free software advocates cried foul when this happened, but the folks at php realized that they were headed for a wall when it came to adoption. Now that anyone can include php in their products without fear of 'infection', php will probably explode in popularity and get a ton of quality review.
      toadlife
      • Just because certain groups don't break the law is not reason to

        say we don't need laws. Maybe you could go to a small town in the middle of Iowa and say there has NEVER been a murder, so why not do away with the law? You examples of projects that have been able to resist forks even though they are legal is not really a good argument. For highly visible projects with an active core of developers, even with BSD style licenses, they will typically not fork, like the examples you gave.

        But, one example you did not mention, is the OS where we have the OpenBSD, NetBSD, FreeBSD, and Darwin forks. On the other hand, we have one Linux kernel.

        You also forgot Microsoft screwing with Keyberos.
        DonnieBoy
        • More ignorance/You are missing the point

          So what if there are forks of the BSDs? They have actually been very benficial, as each fork has benefitted the others by sharing drivers/fixes/improvements. OpenBSD's code auditing discovered serveral bugs which affected all of the BSD's and as a result all of the BSD's are more secure right now. It's been over ten years and nobody has successful closed up the BSD's without giving back. There are Apple exmployees currently working on the FreeBSd development team - if that's not giving back, I don't know what is.

          As for the Microsoft/Kerberos situation, the BSD license had nothing to do with it. I suggest you go and do some research on Microosft and Kerberos before blaming the BSD license.

          I challenge you to find a situation where a proprietary software company "stole" a BSD licensed Open Source project and rendered the Open Source version obsoltete. Countless companies have tried this and failed miserably. Numerous times companies have taken XFree86, closed it up, but XFree86 (and now Xorg) are still alive and kicking. A few companies have even tried to take BSD and sell it, but they have gone under too.

          That's a the point of ESR's rant - the initial fears the GPL advocates had during the 1980's were unfounded.

          [i]"On the other hand, we have one Linux kernel."[/i]

          And 1500 distributions, which are less compatible with eachother than the various BSD forks are with eachother.
          toadlife
  • inclusion

    as has been demonstrated recently by sun, once a business gains enough critical mass, it becomes easier to justify sharing technology and allowing the community to share it. If market forces truly do work, then GPL software will eventually find its way back to FOSS, and if it does not, then it offers some protection for innovation by the inventor. What works in one case may not in another, or it may be a chicken or egg question that will be answered over time by the marketplace. I have to believe that anyone who got their start from an open source would spurn them in the end. (unless your microsoft)
    pesky_z
  • I think the GPL is needed

    One thing I like about the GPL is that it does allow you to take and use software for whatever you wish. You do not have to share your software or its modifications with anyone, as long as you keep them to yourself. If, however, you attempt to take someone else's work and try to sell it, or take the work, modify it, then sell it, then you must give back, not only recognition of the original work, but also surrender your derivative work as well.

    Notice that you don't have to use the original code. You can always start from scratch and make whatever you want and license it however you want. What the GPL protects are software freedoms. If we are going to share, then share means give and take.

    The primary difference between the GPL and other licenses, such as the BSD license or the similar Mozilla license is that you can extract stuff out of the BSD license without ever giving back a thing except for an acknowledgement that you used the code ... at least that's the way that I interpret it.

    Projects large and small have done much with the BSD and the Mozilla licenses, but not all such efforts have given back. Apple and Microsoft, for example, have taken quite a bit of BSD licensed code. I won't say that they haven't given back anything, but if they have contributed, it hasn't been real obvious. I understand that Apple is beginning to make moves to share some of their technology in the OS X --> Darwin/Fink projects and their derivatives. I've not specifically heard of any Microsoft efforts to share back, but perhaps there have been quiet efforts, and perhaps this will change.

    I have no problem with companies trying to earn money, I need money to live myself. I do have problems with individuals or companies taking what is not theirs to begin with, modifying it, and then holding the resultant work as if it is exclusively their own work.

    My personal opinion is that the GPL is an excellent vehicle for assuring and maintaining software freedoms. I think that we need to promote those freedoms so that there is more collaborative sharing of software.

    Personally, I am greatly disappointed that there isn't a much higher degree of sharing than what we have seen over the past twenty years. Though the GPL has resulted in many fine applications that now comprise almost complete systems, it is obvious that a much greater level of cooperation needs to exist. Clearly there isn't much trust between companies, and that, in itself, speaks volumes for why the GPL needs to exist and continue to exist. I think we need more collaboration on core OS elements, beginning with drivers for devices, then moving a bit further up the stack to the common office tools. I think that there is plenty of room for system and application vendors to build upon those foundations, and that's where I think that the proprietary work ought to be focused, not at the OS or commodity applications level.

    There's still hope for that, but I think we need the GPL more than ever.
    masinick@...
    • Resolving the contradiction.

      If existing work is taken and adapted for profit, how should the prior work be acknowledged?

      As you put the problem:
      I have no problem with companies trying to earn money, I need money to live myself. I do have problems with individuals or companies taking what is not theirs to begin with, modifying it, and then holding the resultant work as if it is exclusively their own work.

      The first solution you favor is giving back in return.
      For example:

      Apple and Microsoft, for example, have taken quite a bit of BSD licensed code. I won't say that they haven't given back anything, but if they have contributed, it hasn't been real obvious.

      They can obviously "give back" without reducing the money-making potential of what they've done with the code used.


      But you also approve exactly such a devaluing of potentially profitable software (via the GPL):

      If, however, you attempt to take someone else's work and try to sell it, or take the work, modify it, then sell it, then you must give back, not only recognition of the original work, but also surrender your derivative work as well.

      "Surrender your derivative work" obviously requires a reduction in value. Perhaps you meant some other word than surrender(?).


      I'd say you have it right the first time. The BSD license concerns work that is essentially proof of concept, academic. The concepts can then be transformed into a commercially saleable work, with acknowledgement.

      In return, the recipients of this research can provide money and assistance.

      Everyone can gain potentially because the basic work is available to all. The real competition becomes ways to make money from it. Just like medicines and genetic discoveries.


      By the way, I expect you do realize that operating systems and basic office tools are not commoditized, that some versions are worth a premium (to those who pay for them) because of their special features and capabilities.
      Be careful of applying Unix-style definitions heedlessly of practical realities, please.
      Anton Philidor
    • Another thing I see all the time...

      ...is the ignorance of GPL advocates regarding Microsoft and it's use of BSD code.


      [i]"Projects large and small have done much with the BSD and the Mozilla licenses, but not all such efforts have given back. Apple and Microsoft, for example, have taken quite a bit of BSD licensed code."[/i]

      There is hardly a shred of BSD code in Microsft's operating systems. If you had done some research into the begginings of Windows NT, and specifically Dave Cutler, you would realize this. NT bases OS's are derived from VMS, which have nothing to do with Unix or BSD.

      [i]"I understand that Apple is beginning to make moves to share some of their technology in the OS X --> Darwin/Fink projects and their derivatives. I've not specifically heard of any Microsoft efforts to share back, but perhaps there have been quiet efforts, and perhaps this will change."[/i]

      What is Microsoft going to "share back"..their improvements to the command line ftp client?
      toadlife
  • I release all of the custom programs

    that I write for my clients under the GPL.

    These programs are likely not going to be 'released' anywhere, but I do it for other reasons.

    First, I develop using GPL products (like Qt) and I include GPL products (like MySQL) when I need them. This requires that I distribute the source code and GPL the product. Nothing prevents me from making money doing it, though.

    Second, there is nothing stopping my clients from using someone else for their IT work (except customer satisfaction ;-)), and anyone can modify my code later on. Hopefully, they will respect the GPL and not steal my work, but build on it.

    Third, my client does not 'own' the code that I wrote, so if I have done something I might find useful when working for another client, I am free to re-use what I have done previously.

    Everyone involved benefits one way or another.



    Have a great Fourth, those of you in the U.S.!!
    Hugh Jass
  • Underneath Joe's Great Post

    Underneath Joe's great post is the fact that the GPL is in the process of being rewritten.

    I look at what was said here as one "leak" from within what's currently a Kremlin-like system of re-creation.

    For an open source system, the GPL is conducting its review of licensing in a very closed-source way.

    The information here, I think, may be an attempt to open things up.
    DanaBlankenhorn
    • They should get it right this time...

      Good to know that GPL is being re-written, but they should get it right this time. No doubt, GPL has served a great deal for the open source community, but there must be a way for participation by commercial software. This attitude of "holier than thy" doesn't work.

      Although I still see a lot of challanges ahead. Lets say companies are allowed to just release the object code and an API to further extend that code. But nobody should be able to Patent or copyright the derived work. How will businesses sell their products without a copyright ? May be there is way, but I am not an IP lawyer and have no clue how it could be done....
      low-life
  • Yes

    The GPL will be here as long as it's needed and wanted. It doesn't matter what one person or a group thinks. I guess one can lobby but at this point I doubt it will go anywhere.

    We wouldn't have some of the packages we have today without it. I was looking over a list of licens and for the most part GPL seems to be a safe bet for now.

    It was a great idea at a time when we needed to protect the free tools we have to use to create nice applications.

    Also, anyone can compile and or assemble whatever they want and copy protect it. It's not so easy to do now but I happen to like the pressure it puts on software giants.

    Software giants like MS went their own way off in left field. What makes Open source secure in one way is the fact that it's built on 30+ years of Unix. The concept of Unix programs was to build a program to do one thing really well!

    This is a great universal standard that a business model can be built on.
    xstep