55 ms·

Licensing in a Post Copyright World

129 points by llambda 13y ago

azeemk 13y ago
great post.
zeckalpha 13y ago
This is a must read for anyone licensing software. I wish there would have been a discussion of the ISC license, as I find it even clearer than the MIT/BSD licenses.
- jtome 13y ago
  I agree about the ISC, that's why OpenBSD, the most free OS (as in you can do whatever you want with a larger percentage of it than any other OS), prefers using the ISC over the BSD license now.
  - graue 13y ago
    The FSF's site[1] points out what looks like a serious problem with this license: > This license does have an unfortunate wording choice: it provides recipients with "Permission to use, copy, modify, and/or distribute this software…" This is roughly the same language from the license of Pine that the University of Washington later claimed prohibited people from distributing modified versions of the software. To me, it sounds like "modify and/or distribute" means it should be okay to both modify and then distribute the modified version. But I'm not a lawyer, and if the same language has been used to argue you can't distribute modified versions, that's pretty scary. Enough reason to prefer the slightly longer MIT license. [1] https://www.gnu.org/licenses/license-list.html#ISC https://www.gnu.org/licenses/license-list.html#ISC
    - zeckalpha 13y ago
      I was under the impression that that website is outdated. The ISC used to just say "and distribute", which is similar to older versions of the Pine license, and had problems. However, the new language is more similar to the MIT license ("and/or ..."), but the FSF does not have an issue with the MIT license. https://en.wikipedia.org/wiki/ISC_license https://en.wikipedia.org/wiki/ISC_license https://www.washington.edu/pine/faq/legal.html https://www.washington.edu/pine/faq/legal.html https://www.gnu.org/licenses/license-list.html#X11License https://www.gnu.org/licenses/license-list.html#X11License
      - graue 13y ago
        The FSF quotes the new version with "and/or" in the blurb claiming problems with wording, but that may just be an editing mistake; maybe someone went through and corrected the quotation without realizing it resolved the concern. Either way, though, the OpenBSD folks didn't change "and" to "and/or" in their template, which ranks high for a search for "ISC license". Someone could easily copy the older, problematic version. Given the MIT license works fine, why take even a small risk?
    - dragonwriter 13y ago
      > To me, it sounds like "modify and/or distribute" means it should be okay to both modify and then distribute the modified version. In fact, and/or is exactly the way to specify that options can either be used together or separately. The FSF's claimed problem, you'll note, isn't that this language was ever held to restrict distributing modified versions, only that similar language was once argued by a party with a financial interest in the outcome to have that effect. Which is a pretty flimsy basis to assert a problem. I'm sure you can find passages in the GPL that are "roughly the same" language to something that someone, somewhere, argued in a court case meant something that would be frightening if it was what the similar language in the GPL passage was actually treated as meaning under the law, but the fact that someone once argued that a similar piece of the language means something doesn't mean that the language that they argued about (and even less, the actual license language that it is "roughly the same as") has ever been or would ever be given that effect in court.
      - graue 13y ago
        Ultimately I care more about avoiding the increased risk of the ISC license than about saving a few lines. It would be silly if trying to make my license statement a few insignificant lines shorter prevented someone from using my open-source code.
        dragonwriter 13y ago
        My point was that the argument that there is increased risk is based on a flimsy premise which could equally well be applied to just about anything; there's very few pieces of commonly used legal language where someone, somewhere hasn't at one time or another argued that something phrased in "roughly the same" manner means something that you'd want to avoid.
        graue 13y ago
        And my point is that even going through this legal reasoning is more work than just saying, "Fuck it, I'll use the MIT license."
desmondrd 13y ago
This is awesome.
jdmitch 13y ago
Very informative - here is a license comparison tool I've found useful: http://www.tldrlegal.com/compare http://www.tldrlegal.com/compare
__david__ 13y ago
One license I think gets overlooked is the Mozilla Public License (MPL). I think it's particularly good for libraries and I've published some of my simple javascript code under it. The interesting part about it is that it says modifications must be published, but that they are file based and not project or linker based. This omits many of the "onerous" (from the library user's perspective) restrictions of the LGPL, while still keeping the spirit of "this is free software, please give your enhancements to others to make the world better" which is sort of the whole point of GNU.
- JoshTriplett 13y ago
  Note, though, that older versions of the MPL (before version 2.0) are incompatible with the GPL, so you end up with things like MPL/GPL dual-licensing. MPL 2.0 is OK in that regard, though I still prefer LGPL or GPL.
  - synchronise 13y ago
    You can use someone else's proprietary modules in MPL code to better your program or library, you can't do the same with (L)GPL.
    - JoshTriplett 13y ago
      In many cases you can with LGPL, and you can with GPL if you explicitly add a licensing exception. Also, regarding "use ... proprietary modules ... to better your program or library": depends on your definition of "better", but if that's one of your goals, then I agree that you don't want the GPL.
      - synchronise 13y ago
        To add onto it in a beneficial way to the user. Also, MPL allows static linking, the LGPL doesn't unless you explicitly state it. For the record, would 'GPL + Linking exception' be incompatible with the GPL and LGPL?
qznc 13y ago
My try of a licence wizard: http://beza1e1.tuxen.de/licences/ http://beza1e1.tuxen.de/licences/
- Tobu 13y ago
  Thoughts on LGPL vs MPL vs GPL with linking exception?
  - qznc 13y ago
    I opted for MPL, because it is the most liberal licence that I could find, which is still "copyleft". The spectrum between MPL and GPL (covering LGPL etc) is complex.
pessimizer 13y ago
>People that license software under the BSD or MIT license probably would not mind that much if copyrights would be abolished or greatly restricted. Richard Stallman's world on the other hand would would fall apart. I'm pretty sure that Stallman would love a post-copyright world, unless it led to a world of rented locked down single-purpose appliances run on gifted code - an outcome that MIT/BSD would be totally fine with. AGPL and GPLv3 was a line in the sand to separate OSI types from people who feel that it is a problem that people are legally and physically alienated from their own tools. IMO, releasing the source code that operates people's property is equivalent to putting a list of ingredients on a packaged food product.
- rayiner 13y ago
  > IMO, releasing the source code that operates people's property is equivalent to putting a list of ingredients on a packaged food product. No, it's like putting the secret recipe to Coke on the bottle, or the food science process that makes Rice Crispies so crispy.
  - pessimizer 13y ago
    The GPL doesn't force you to tell people how to build your appliance, it forces you to tell them what's in it. That's not comparable to having to share the manufacturing process of a good, but is comparable to having to share what's in it.
    - rayiner 13y ago
      That takes the view that the software is just an ingredient, instead of a crucial part of the "secret sauce" that makes the product special. In many cases, the software is the interesting part, not the hardware.
      - pessimizer 13y ago
        I'm not saying that software is just an ingredient, I'm saying that software is an ingredient. I don't know what it means to be just an ingredient. Rice isn't just an ingredient, nor water - unless I'm misunderstanding you.
MikeTaylor 13y ago
This post is in most respects excellent, but it suffers from a fundamental problem. It starts from the assumption that the goal of free software is to be as widely used as possible, and that anything that prevents this is therefore Bad. But widest possible use was never the goal of the Free Software Foundation. They have always been about freedom first and software second, and Stallman has many times stated that he'd rather a program be used less if it results in more freedom, than vice versa. So the goal of the various restrictions imposed by the GPL (in either of the main versions) is to increase freedom. We can usefully have a debate about how well they achieve that goal, and whether the increase in freedom is worth what has to be spent in order to achieve it. But if we do that, let's do it from a position of first having understood why the GPL says the things it does. I saw this summed up an admirably concise way some years ago, but stupidly forgot to note whose soundbite this is: "Saying that the GPL is less free because it forbids proprietary derivatives is like saying the USA is less free because it forbids slavery". Yes, of course it's hyperbole to hold up freedom of software alongside the freedom of people; but this quote shows by example that more freedom doesn't always get achieved by less regulation but by more.
- baddox 13y ago
  > They have always been about freedom first and software second, and Stallman has many times stated that he'd rather a program be used less if it results in more freedom, than vice versa. I think I disagree with your interpretation, but I think my disagreement hinges on a semantic disagreement about the word "freedom." To me, the GPL is a legal limitation on freedom. It prohibits certain use cases, like distributing that software without its source code. To me, a piece of software is "more free" if its source code is available, but people are still able to do literally anything they want with the source code, including distributing binaries without providing access to the source code. I think the point of the GPL is to utilize the copyright system to ensure access to source code for everyone, not to simply increase freedom. The slavery quip doesn't really make sense, because in that analogy the source code is the slave. So unless we're talking about increasing freedom for source code (which is a strange notion, since most source code is not conscious) rather than increasing freedom for people, my disagreement still applies. The reason forbidding slavery is not generally considered a decrease in freedom is that both slave owners and slaves are people.
null_ptr 13y ago
"What if it will become illegal to sell consumer devices without signature checks for all software running on it." Well that would be a tragedy. As if computing isn't moving towards closed systems fast enough. I really hope general-purpose computing devices will not become something only found in museums. Few movements in humanity's history were as revolutionary.
er0k 13y ago
This is why I use the WTFPL (DO WHAT THE FUCK YOU WANT TO PUBLIC LICENSE) http://www.wtfpl.net/txt/copying/ http://www.wtfpl.net/txt/copying/
- LeafStorm 13y ago
  I'm pretty sure suing you if it causes damage counts as "what the f--k they want to."
  - droithomme 13y ago
    Can you post a list of all lawsuits filed against WTFPL licensed software?
    - LeafStorm 13y ago
      Yes: [] But that's only because it hasn't been used for anything worth suing over.
      - droithomme 13y ago
        Are you aware that disclaimers of liability are not considered valid contract terms in most scenarios and legal jurisdictions in western countries where software is likely to be used? Do you believe that you can disclaim liability for negligence such as carelessness or a failure to thoroughly test? How about for intentional malicious acts? Courts in the US and commonwealth nations at least don't allow such disclaimers. What about other sorts of liability? Do you know the differences between direct, consequential, special, incidental and indirect damages? The reality is that in enough jurisdictions to matter, authors can be held legally liable for bad code they create no matter whether they have a liability clause as a condition of licensing or not. Forming an LLC or corporation won't shield them either if someone wants to sue. The only protection in these cases is maintaining E&O insurance. If I use GPL, BSD, or any other licensed code that disclaims liability, and the code is flawed because the author was negligent, I can sue the author for damages and I can win. The disclaimer does absolutely nothing. What purpose then is there in cluttering up licenses with useless clauses?
egh 13y ago
If you want to misunderstand licensing, this article is here to help. To pick out just one of the many errors of fact and interpretation: linux has historically been GPL v2 only; this decision was not made after the release of GPL v3. It's great that his "generation sees copyrights [as] a concept that should be much more restricted and have a smaller lifetime": I'm glad that he's so surrounded by the FOSS that makes this possible. But this possibility came about because of the work of free software advocates and developers. Reveling in your ignorance and disinterest really isn't going helping. And if you think that it's a problem that Apple and the FSF (among others!) are clashing on issues like restricted bootloaders, tivoization, etc. -- well, I don't really know what to say. There's a reason that Apple and advocates like the FSF are on different sides here. I know which side I'm on: do you?
- nkurz 13y ago
  There's a reason that Apple and advocates like the FSF are on different sides here. I know which side I'm on: do you? In the US the Democrats and Republicans clash on many issues. There are reasons for this as well, which have mostly to do with self-interest. Does this mean that I have to choose one of them, even if my interests align with neither? Might there be a better choice beyond those two? I'm pretty certain I'm not on either side.
  - mcguire 13y ago
    Not making a choice is still a choice.
stormbrew 13y ago
An interesting thing, to me, about the GPL is that it's possible for companies to use it as an anti-competitive weapon against each other. A company who owns or has been assigned the copyright to an entire codebase is able to hold the exclusive right to profit from hidden modifications of it. Oracle is in this position right now with mysql, for example. GNU itself, were it so inclined (though it obviously isn't), would be as well. This becomes even more significant when talking about the AGPL and its limitations on using it as a service. 10gen is the only company that could run a proprietary uber-version of MongoDB, and that right persists effectively forever. When it comes to more open licenses like BSD no one holds this as an exclusive right.
- zanny 13y ago
  I think if the FSF could get away with a clause akin to "once licensed under the GPL, this software can not be licensed under any other license for distribution, and it's license can not be changed by the copyright holder". But that wouldn't fly. Ever. At the end of the day, the copyright holder still gets to change their license whenever they want, and the license itself can't deny that, because it would never hold up in court.
  - icebraining 13y ago
    It wouldn't fly, because to the copyright holder, the license simply doesn't apply. At all. You only need a license if you're restricted by copyright law from doing what you're doing, but copyright law doesn't restrict the holder themselves.
- spacelizard 13y ago
  Arguably that problem is worse with permissively licensed software. In that case you don't just have the original copyright holder peddling a proprietary uber-version, you have every company and his dog doing it and every version is expensive and/or incompatible with each other.
  - stormbrew 13y ago
    I don't see how it can be considered worse. I could see even, but not worse. Does it matter if it's one or ten organizations that are taking advantage of the ability to make proprietary software out of open source code? At least with ten there might actually be competition. But realistically, it's not like permissive licenses are a new thing and this apocalypse doesn't seem to be a very common occurrence in practice.
nkurz 13y ago
This is a wonderful article, and a great summary of the current open-source software licensing issues, but I feel uneasy about the apparent conclusion. After dozens of paragraphs about the intricacies and uncertainties of licensing, it seems to suggest that if instead of asking people to choose from a drop-down list of licenses, things would be greatly improved if they were given a 10 minute overview before being forced to choose. Any overview, even if vetted by vast teams of international copyright lawyers, can only brush the surface of the issues. At best, you could end up with a list of reasons to eliminate certain licenses based on your personal goals, with the wordings chosen by the ideologies of the lawyers involved. Would this be any better than GitHub deciding internally which license(s) they want to promote, and putting these at the top of the list? I think we should start to seriously consider simplifying our software licensing environment This I can agree with. Having seen the amount of time wasted by open-source projects on licensing issues, copyright worries, and patent fears (and the resulting reinvention of the wheel) I think simplification is the way to go. Personally, I don't think this simplification can be achieved by piling more layers of abstraction on top of copyright and contracts. Most books and papers seem to get along alright without licenses. I'd love to see "fair use" defined more broadly, but I don't think that adding additional licensing would be an improvement. Most electronic parts can be used as building blocks for whatever project you want without worrying whether one part will "contaminate" the others or "infect" the project. Music is licensed more frequently, but I don't think amateur musicians would benefit if they had to agree to terms of service before attending a concert. Please tell me again why we want every project on GitHub to choose and stick to a license? And who is this "we", and what do "we" stand to gain?
- spacelizard 13y ago
  Yes, the morass of software licensing sucks. But unfortunately another side effect of that is that not choosing a license or making uninformed decisions about licensing creates situations that end up with users becoming extremely susceptible to copyright trolling. It is true though that the solution is obviously to ease up on some of the powers granted by copyrights.
  - nkurz 13y ago
jtome 13y ago
Did anyone notice the line: What if it will become illegal to sell consumer devices without signature checks for all software running on it. I find it pretty weird that he's complaining that any country foolish enough to enact such a draconian law would also criminalize all gplv3 software. I personally think the gplv3 would be wonderful in this case since it would create a high cost for enacting such an evil law.
lifeisstillgood 13y ago
Most people are ignoring licenses, because most developers think they have won. There is a generation of developers for whom software freedom is in the bones. Why bother putting a license on code, or releasing code under compatible licenses -its all free and open. This attitude may seem naive - but i see it as wondrous - any attempts to put the genie back in the box will not be met by court cases led by Lessig, but simple mass global disobedience - software is free when everyone simply thinks it is free.
- icebraining 13y ago
  Most developers think software is free for them to use, while completely proprietary if anyone tries to use theirs. It's not software freedom in their bones, it's self-interest running in their veins.
  - lifeisstillgood 13y ago
    Well logically that still means everyone thinks all software is free. :-) Excellent metaphor aikido there though.
- mcguire 13y ago
  ...unless you make money using software, in which case you can be disobedient all you want, from the cardboard box in which you live.
- gillianseed 13y ago
  I don't know the original study in detail, but given my time on github I see a ton of 'projects' which are just small scraps of test code, heck even dot files. I can certainly see why people don't bother putting a licence on these types of 'projects'. However on all projects I've come across which are actually something substantial there has been a licence declared, so I'm wondering if this 'lack of licencing' is a 'real' problem or if it's just the result of github becoming a convenient dumping ground/backup for tons of small snippets/settings etc which used to just sit on people's harddrives.
binarycrusader 13y ago
Sun liked the whole concept of not being compatible with GPL... This old chestnut again? I'm assuming this is based on Danese Cooper's comments. Well, Danese was wrong: She is speaking this way because she lost an argument inside Sun, not because her view is representative of the views of Sun or its staff in the way she claims. She, along with many actual engineers, was an advocate of using GPL or OpenSolaris but the need to release rather than wait for one of {GPL v3, Mozilla license revision, encumbrance removal} meant that this was not possible. I am still furious with her for the statement she made at DebConf, which was spiteful and an obstacle to a united FOSS movement. http://web.archive.org/web/20110716130648/http://www.opensolaris.org/jive/message.jspa?messageID=55013 http://web.archive.org/web/20110716130648/http://www.opensol...
drewcrawford 13y ago
This is a great article. One thing I would add to this is that nobody really knows what licenses like the GPL actually do. Jacob Kaplan-Moss published a fantastic piece called "20 questions about the GPL"[0] , and my attempts to clarify a few of them with the FSF and SFLC have so far revealed that not even two developer advocate organizations can agree on even fundamentally basic concepts about the most popular license, so isn't that a problem? The reason that I contribute to the "projects on GitHub with no license" problem is that I cannot find a license that unambiguously does what I want, after spending many, many weekends on the problem. I could write one, or get my lawyer to, but license proliferation is grounds for expulsion from the cool kids club. Since I have no good options, I take no action. Fundamentally, I believe we need a new license steward that isn't afraid to update regularly to address 1990s questions like dynamic linking and optional dependencies, and 2010s questions like app stores. Something like Creative Commons for code, where you have some options to choose from. [0] http://jacobian.org/writing/gpl-questions/ http://jacobian.org/writing/gpl-questions/
- tjaerv 13y ago
  Creative Commons for code would be CC0: http://creativecommons.org/weblog/entry/27081 http://creativecommons.org/weblog/entry/27081
yk 13y ago
The post ignores, why there are so many different FOSS licenses, the thing is people have interests and sometimes these interests are not compatible. So there is a place for MIT licenses, namely if you want as much adoption as possible, and there is a place for copy left licenses, if you want to protect the associated rights of copyleft. This is probably most visible in the “anti-tivoization” section, which concludes The license restrictions of the AppStore, Google's Play Store and similar distribution systems and the GPLv3 are just not compatible. The reason for this is Apple, Google and others have their interests, namely controlling a platform and enforcing whichever restrictions are mandated by their interests.[1] On the other hand, the GPL[v3] tries to protect the freedom to tinker for the owner of the device, the ability not only to see the source code, but also to modify and run it. So the absence of GPLv3 licensed software on iOS is actually proof, that the license is working as intended. [1] Some examples of these interests are: Enforcement of DRM, trying to create a 'more secure' environment, having a consistent UX, etc.
danbmil99 13y ago
Great article. My favorite example of these Talmudic complexities is opensim (opensimulator.org), an MIT-licensed Second Life-compatible server that has a twisted, Black Widow-like relationship with Linden's GPL-licensed client code. Before 2011, there was a nutty "6 month cool-off period" before any SL client contributor could contribute to Opensim: http://justincc.org/blog/2011/06/30/announcing-the-overte-foundation-for-opensimulator/ http://justincc.org/blog/2011/06/30/announcing-the-overte-fo...

129 points by llambda 13y ago

great post.

This is a must read for anyone licensing software. I wish there would have been a discussion of the ISC license, as I find it even clearer than the MIT/BSD licenses.

jtome 13y ago
I agree about the ISC, that's why OpenBSD, the most free OS (as in you can do whatever you want with a larger percentage of it than any other OS), prefers using the ISC over the BSD license now.
- graue 13y ago
  The FSF's site[1] points out what looks like a serious problem with this license: > This license does have an unfortunate wording choice: it provides recipients with "Permission to use, copy, modify, and/or distribute this software…" This is roughly the same language from the license of Pine that the University of Washington later claimed prohibited people from distributing modified versions of the software. To me, it sounds like "modify and/or distribute" means it should be okay to both modify and then distribute the modified version. But I'm not a lawyer, and if the same language has been used to argue you can't distribute modified versions, that's pretty scary. Enough reason to prefer the slightly longer MIT license. [1] https://www.gnu.org/licenses/license-list.html#ISC https://www.gnu.org/licenses/license-list.html#ISC
  - zeckalpha 13y ago
    I was under the impression that that website is outdated. The ISC used to just say "and distribute", which is similar to older versions of the Pine license, and had problems. However, the new language is more similar to the MIT license ("and/or ..."), but the FSF does not have an issue with the MIT license. https://en.wikipedia.org/wiki/ISC_license https://en.wikipedia.org/wiki/ISC_license https://www.washington.edu/pine/faq/legal.html https://www.washington.edu/pine/faq/legal.html https://www.gnu.org/licenses/license-list.html#X11License https://www.gnu.org/licenses/license-list.html#X11License
    - graue 13y ago
      The FSF quotes the new version with "and/or" in the blurb claiming problems with wording, but that may just be an editing mistake; maybe someone went through and corrected the quotation without realizing it resolved the concern. Either way, though, the OpenBSD folks didn't change "and" to "and/or" in their template, which ranks high for a search for "ISC license". Someone could easily copy the older, problematic version. Given the MIT license works fine, why take even a small risk?
  - dragonwriter 13y ago
    > To me, it sounds like "modify and/or distribute" means it should be okay to both modify and then distribute the modified version. In fact, and/or is exactly the way to specify that options can either be used together or separately. The FSF's claimed problem, you'll note, isn't that this language was ever held to restrict distributing modified versions, only that similar language was once argued by a party with a financial interest in the outcome to have that effect. Which is a pretty flimsy basis to assert a problem. I'm sure you can find passages in the GPL that are "roughly the same" language to something that someone, somewhere, argued in a court case meant something that would be frightening if it was what the similar language in the GPL passage was actually treated as meaning under the law, but the fact that someone once argued that a similar piece of the language means something doesn't mean that the language that they argued about (and even less, the actual license language that it is "roughly the same as") has ever been or would ever be given that effect in court.
    - graue 13y ago
      Ultimately I care more about avoiding the increased risk of the ISC license than about saving a few lines. It would be silly if trying to make my license statement a few insignificant lines shorter prevented someone from using my open-source code.
      - dragonwriter 13y ago
        My point was that the argument that there is increased risk is based on a flimsy premise which could equally well be applied to just about anything; there's very few pieces of commonly used legal language where someone, somewhere hasn't at one time or another argued that something phrased in "roughly the same" manner means something that you'd want to avoid.
        graue 13y ago
        And my point is that even going through this legal reasoning is more work than just saying, "Fuck it, I'll use the MIT license."

desmondrd 13y ago

This is awesome.

jdmitch 13y ago

Very informative - here is a license comparison tool I've found useful: http://www.tldrlegal.com/compare http://www.tldrlegal.com/compare

__david__ 13y ago

One license I think gets overlooked is the Mozilla Public License (MPL). I think it's particularly good for libraries and I've published some of my simple javascript code under it. The interesting part about it is that it says modifications must be published, but that they are file based and not project or linker based. This omits many of the "onerous" (from the library user's perspective) restrictions of the LGPL, while still keeping the spirit of "this is free software, please give your enhancements to others to make the world better" which is sort of the whole point of GNU.

JoshTriplett 13y ago
Note, though, that older versions of the MPL (before version 2.0) are incompatible with the GPL, so you end up with things like MPL/GPL dual-licensing. MPL 2.0 is OK in that regard, though I still prefer LGPL or GPL.
- synchronise 13y ago
  You can use someone else's proprietary modules in MPL code to better your program or library, you can't do the same with (L)GPL.
  - JoshTriplett 13y ago
    In many cases you can with LGPL, and you can with GPL if you explicitly add a licensing exception. Also, regarding "use ... proprietary modules ... to better your program or library": depends on your definition of "better", but if that's one of your goals, then I agree that you don't want the GPL.
    - synchronise 13y ago
      To add onto it in a beneficial way to the user. Also, MPL allows static linking, the LGPL doesn't unless you explicitly state it. For the record, would 'GPL + Linking exception' be incompatible with the GPL and LGPL?

qznc 13y ago

My try of a licence wizard: http://beza1e1.tuxen.de/licences/ http://beza1e1.tuxen.de/licences/

Tobu 13y ago
Thoughts on LGPL vs MPL vs GPL with linking exception?
- qznc 13y ago
  I opted for MPL, because it is the most liberal licence that I could find, which is still "copyleft". The spectrum between MPL and GPL (covering LGPL etc) is complex.

pessimizer 13y ago

>People that license software under the BSD or MIT license probably would not mind that much if copyrights would be abolished or greatly restricted. Richard Stallman's world on the other hand would would fall apart. I'm pretty sure that Stallman would love a post-copyright world, unless it led to a world of rented locked down single-purpose appliances run on gifted code - an outcome that MIT/BSD would be totally fine with. AGPL and GPLv3 was a line in the sand to separate OSI types from people who feel that it is a problem that people are legally and physically alienated from their own tools. IMO, releasing the source code that operates people's property is equivalent to putting a list of ingredients on a packaged food product.

rayiner 13y ago
> IMO, releasing the source code that operates people's property is equivalent to putting a list of ingredients on a packaged food product. No, it's like putting the secret recipe to Coke on the bottle, or the food science process that makes Rice Crispies so crispy.
- pessimizer 13y ago
  The GPL doesn't force you to tell people how to build your appliance, it forces you to tell them what's in it. That's not comparable to having to share the manufacturing process of a good, but is comparable to having to share what's in it.
  - rayiner 13y ago
    That takes the view that the software is just an ingredient, instead of a crucial part of the "secret sauce" that makes the product special. In many cases, the software is the interesting part, not the hardware.
    - pessimizer 13y ago
      I'm not saying that software is just an ingredient, I'm saying that software is an ingredient. I don't know what it means to be just an ingredient. Rice isn't just an ingredient, nor water - unless I'm misunderstanding you.

MikeTaylor 13y ago

This post is in most respects excellent, but it suffers from a fundamental problem. It starts from the assumption that the goal of free software is to be as widely used as possible, and that anything that prevents this is therefore Bad. But widest possible use was never the goal of the Free Software Foundation. They have always been about freedom first and software second, and Stallman has many times stated that he'd rather a program be used less if it results in more freedom, than vice versa. So the goal of the various restrictions imposed by the GPL (in either of the main versions) is to increase freedom. We can usefully have a debate about how well they achieve that goal, and whether the increase in freedom is worth what has to be spent in order to achieve it. But if we do that, let's do it from a position of first having understood why the GPL says the things it does. I saw this summed up an admirably concise way some years ago, but stupidly forgot to note whose soundbite this is: "Saying that the GPL is less free because it forbids proprietary derivatives is like saying the USA is less free because it forbids slavery". Yes, of course it's hyperbole to hold up freedom of software alongside the freedom of people; but this quote shows by example that more freedom doesn't always get achieved by less regulation but by more.

baddox 13y ago
> They have always been about freedom first and software second, and Stallman has many times stated that he'd rather a program be used less if it results in more freedom, than vice versa. I think I disagree with your interpretation, but I think my disagreement hinges on a semantic disagreement about the word "freedom." To me, the GPL is a legal limitation on freedom. It prohibits certain use cases, like distributing that software without its source code. To me, a piece of software is "more free" if its source code is available, but people are still able to do literally anything they want with the source code, including distributing binaries without providing access to the source code. I think the point of the GPL is to utilize the copyright system to ensure access to source code for everyone, not to simply increase freedom. The slavery quip doesn't really make sense, because in that analogy the source code is the slave. So unless we're talking about increasing freedom for source code (which is a strange notion, since most source code is not conscious) rather than increasing freedom for people, my disagreement still applies. The reason forbidding slavery is not generally considered a decrease in freedom is that both slave owners and slaves are people.

null_ptr 13y ago

"What if it will become illegal to sell consumer devices without signature checks for all software running on it." Well that would be a tragedy. As if computing isn't moving towards closed systems fast enough. I really hope general-purpose computing devices will not become something only found in museums. Few movements in humanity's history were as revolutionary.

er0k 13y ago

This is why I use the WTFPL (DO WHAT THE FUCK YOU WANT TO PUBLIC LICENSE) http://www.wtfpl.net/txt/copying/ http://www.wtfpl.net/txt/copying/

LeafStorm 13y ago
I'm pretty sure suing you if it causes damage counts as "what the f--k they want to."
- droithomme 13y ago
  Can you post a list of all lawsuits filed against WTFPL licensed software?
  - LeafStorm 13y ago
    Yes: [] But that's only because it hasn't been used for anything worth suing over.
    - droithomme 13y ago
      Are you aware that disclaimers of liability are not considered valid contract terms in most scenarios and legal jurisdictions in western countries where software is likely to be used? Do you believe that you can disclaim liability for negligence such as carelessness or a failure to thoroughly test? How about for intentional malicious acts? Courts in the US and commonwealth nations at least don't allow such disclaimers. What about other sorts of liability? Do you know the differences between direct, consequential, special, incidental and indirect damages? The reality is that in enough jurisdictions to matter, authors can be held legally liable for bad code they create no matter whether they have a liability clause as a condition of licensing or not. Forming an LLC or corporation won't shield them either if someone wants to sue. The only protection in these cases is maintaining E&O insurance. If I use GPL, BSD, or any other licensed code that disclaims liability, and the code is flawed because the author was negligent, I can sue the author for damages and I can win. The disclaimer does absolutely nothing. What purpose then is there in cluttering up licenses with useless clauses?

egh 13y ago

If you want to misunderstand licensing, this article is here to help. To pick out just one of the many errors of fact and interpretation: linux has historically been GPL v2 only; this decision was not made after the release of GPL v3. It's great that his "generation sees copyrights [as] a concept that should be much more restricted and have a smaller lifetime": I'm glad that he's so surrounded by the FOSS that makes this possible. But this possibility came about because of the work of free software advocates and developers. Reveling in your ignorance and disinterest really isn't going helping. And if you think that it's a problem that Apple and the FSF (among others!) are clashing on issues like restricted bootloaders, tivoization, etc. -- well, I don't really know what to say. There's a reason that Apple and advocates like the FSF are on different sides here. I know which side I'm on: do you?

nkurz 13y ago
There's a reason that Apple and advocates like the FSF are on different sides here. I know which side I'm on: do you? In the US the Democrats and Republicans clash on many issues. There are reasons for this as well, which have mostly to do with self-interest. Does this mean that I have to choose one of them, even if my interests align with neither? Might there be a better choice beyond those two? I'm pretty certain I'm not on either side.
- mcguire 13y ago
  Not making a choice is still a choice.

stormbrew 13y ago

An interesting thing, to me, about the GPL is that it's possible for companies to use it as an anti-competitive weapon against each other. A company who owns or has been assigned the copyright to an entire codebase is able to hold the exclusive right to profit from hidden modifications of it. Oracle is in this position right now with mysql, for example. GNU itself, were it so inclined (though it obviously isn't), would be as well. This becomes even more significant when talking about the AGPL and its limitations on using it as a service. 10gen is the only company that could run a proprietary uber-version of MongoDB, and that right persists effectively forever. When it comes to more open licenses like BSD no one holds this as an exclusive right.

zanny 13y ago
I think if the FSF could get away with a clause akin to "once licensed under the GPL, this software can not be licensed under any other license for distribution, and it's license can not be changed by the copyright holder". But that wouldn't fly. Ever. At the end of the day, the copyright holder still gets to change their license whenever they want, and the license itself can't deny that, because it would never hold up in court.
- icebraining 13y ago
  It wouldn't fly, because to the copyright holder, the license simply doesn't apply. At all. You only need a license if you're restricted by copyright law from doing what you're doing, but copyright law doesn't restrict the holder themselves.
spacelizard 13y ago
Arguably that problem is worse with permissively licensed software. In that case you don't just have the original copyright holder peddling a proprietary uber-version, you have every company and his dog doing it and every version is expensive and/or incompatible with each other.
- stormbrew 13y ago
  I don't see how it can be considered worse. I could see even, but not worse. Does it matter if it's one or ten organizations that are taking advantage of the ability to make proprietary software out of open source code? At least with ten there might actually be competition. But realistically, it's not like permissive licenses are a new thing and this apocalypse doesn't seem to be a very common occurrence in practice.

nkurz 13y ago

This is a wonderful article, and a great summary of the current open-source software licensing issues, but I feel uneasy about the apparent conclusion. After dozens of paragraphs about the intricacies and uncertainties of licensing, it seems to suggest that if instead of asking people to choose from a drop-down list of licenses, things would be greatly improved if they were given a 10 minute overview before being forced to choose. Any overview, even if vetted by vast teams of international copyright lawyers, can only brush the surface of the issues. At best, you could end up with a list of reasons to eliminate certain licenses based on your personal goals, with the wordings chosen by the ideologies of the lawyers involved. Would this be any better than GitHub deciding internally which license(s) they want to promote, and putting these at the top of the list? I think we should start to seriously consider simplifying our software licensing environment This I can agree with. Having seen the amount of time wasted by open-source projects on licensing issues, copyright worries, and patent fears (and the resulting reinvention of the wheel) I think simplification is the way to go. Personally, I don't think this simplification can be achieved by piling more layers of abstraction on top of copyright and contracts. Most books and papers seem to get along alright without licenses. I'd love to see "fair use" defined more broadly, but I don't think that adding additional licensing would be an improvement. Most electronic parts can be used as building blocks for whatever project you want without worrying whether one part will "contaminate" the others or "infect" the project. Music is licensed more frequently, but I don't think amateur musicians would benefit if they had to agree to terms of service before attending a concert. Please tell me again why we want every project on GitHub to choose and stick to a license? And who is this "we", and what do "we" stand to gain?

spacelizard 13y ago
Yes, the morass of software licensing sucks. But unfortunately another side effect of that is that not choosing a license or making uninformed decisions about licensing creates situations that end up with users becoming extremely susceptible to copyright trolling. It is true though that the solution is obviously to ease up on some of the powers granted by copyrights.
- nkurz 13y ago

jtome 13y ago

Did anyone notice the line: What if it will become illegal to sell consumer devices without signature checks for all software running on it. I find it pretty weird that he's complaining that any country foolish enough to enact such a draconian law would also criminalize all gplv3 software. I personally think the gplv3 would be wonderful in this case since it would create a high cost for enacting such an evil law.

lifeisstillgood 13y ago

Most people are ignoring licenses, because most developers think they have won. There is a generation of developers for whom software freedom is in the bones. Why bother putting a license on code, or releasing code under compatible licenses -its all free and open. This attitude may seem naive - but i see it as wondrous - any attempts to put the genie back in the box will not be met by court cases led by Lessig, but simple mass global disobedience - software is free when everyone simply thinks it is free.

icebraining 13y ago
Most developers think software is free for them to use, while completely proprietary if anyone tries to use theirs. It's not software freedom in their bones, it's self-interest running in their veins.
- lifeisstillgood 13y ago
  Well logically that still means everyone thinks all software is free. :-) Excellent metaphor aikido there though.
mcguire 13y ago
...unless you make money using software, in which case you can be disobedient all you want, from the cardboard box in which you live.
gillianseed 13y ago
I don't know the original study in detail, but given my time on github I see a ton of 'projects' which are just small scraps of test code, heck even dot files. I can certainly see why people don't bother putting a licence on these types of 'projects'. However on all projects I've come across which are actually something substantial there has been a licence declared, so I'm wondering if this 'lack of licencing' is a 'real' problem or if it's just the result of github becoming a convenient dumping ground/backup for tons of small snippets/settings etc which used to just sit on people's harddrives.

binarycrusader 13y ago

Sun liked the whole concept of not being compatible with GPL... This old chestnut again? I'm assuming this is based on Danese Cooper's comments. Well, Danese was wrong: She is speaking this way because she lost an argument inside Sun, not because her view is representative of the views of Sun or its staff in the way she claims. She, along with many actual engineers, was an advocate of using GPL or OpenSolaris but the need to release rather than wait for one of {GPL v3, Mozilla license revision, encumbrance removal} meant that this was not possible. I am still furious with her for the statement she made at DebConf, which was spiteful and an obstacle to a united FOSS movement. http://web.archive.org/web/20110716130648/http://www.opensolaris.org/jive/message.jspa?messageID=55013 http://web.archive.org/web/20110716130648/http://www.opensol...

drewcrawford 13y ago

This is a great article. One thing I would add to this is that nobody really knows what licenses like the GPL actually do. Jacob Kaplan-Moss published a fantastic piece called "20 questions about the GPL"[0] , and my attempts to clarify a few of them with the FSF and SFLC have so far revealed that not even two developer advocate organizations can agree on even fundamentally basic concepts about the most popular license, so isn't that a problem? The reason that I contribute to the "projects on GitHub with no license" problem is that I cannot find a license that unambiguously does what I want, after spending many, many weekends on the problem. I could write one, or get my lawyer to, but license proliferation is grounds for expulsion from the cool kids club. Since I have no good options, I take no action. Fundamentally, I believe we need a new license steward that isn't afraid to update regularly to address 1990s questions like dynamic linking and optional dependencies, and 2010s questions like app stores. Something like Creative Commons for code, where you have some options to choose from. [0] http://jacobian.org/writing/gpl-questions/ http://jacobian.org/writing/gpl-questions/

tjaerv 13y ago
Creative Commons for code would be CC0: http://creativecommons.org/weblog/entry/27081 http://creativecommons.org/weblog/entry/27081

yk 13y ago

The post ignores, why there are so many different FOSS licenses, the thing is people have interests and sometimes these interests are not compatible. So there is a place for MIT licenses, namely if you want as much adoption as possible, and there is a place for copy left licenses, if you want to protect the associated rights of copyleft. This is probably most visible in the “anti-tivoization” section, which concludes The license restrictions of the AppStore, Google's Play Store and similar distribution systems and the GPLv3 are just not compatible. The reason for this is Apple, Google and others have their interests, namely controlling a platform and enforcing whichever restrictions are mandated by their interests.[1] On the other hand, the GPL[v3] tries to protect the freedom to tinker for the owner of the device, the ability not only to see the source code, but also to modify and run it. So the absence of GPLv3 licensed software on iOS is actually proof, that the license is working as intended. [1] Some examples of these interests are: Enforcement of DRM, trying to create a 'more secure' environment, having a consistent UX, etc.

danbmil99 13y ago

Great article. My favorite example of these Talmudic complexities is opensim (opensimulator.org), an MIT-licensed Second Life-compatible server that has a twisted, Black Widow-like relationship with Linden's GPL-licensed client code. Before 2011, there was a nutty "6 month cool-off period" before any SL client contributor could contribute to Opensim: http://justincc.org/blog/2011/06/30/announcing-the-overte-foundation-for-opensimulator/ http://justincc.org/blog/2011/06/30/announcing-the-overte-fo...