3 ms·

> GPL is ironically one of the least free licenses This is a common misunderstanding as you are talking from the perspective of a publisher, rather than that

by zizee 10y ago

> GPL is ironically one of the least free licenses This is a common misunderstanding as you are talking from the perspective of a publisher, rather than that of an End User. GPL is about enforcing the rights of End Users, not Publishers. So, from a End Users perspective, GPL is more free than say MIT as it gives the end user the freedom to look at and change the code of software that they are running. This takes away the freedom of the publisher to keep code secret. MIT provides more rights to a Publisher than the End User as it allows the publisher to do pretty much whatever they want, but removes freedom to look and modify code from the end user. It is all a matter of perspective, wherein lies the confusion.

yellowapple 10y ago
> MIT [...] removes freedom to look and modify code from the end user. Er, what? What's stopping the end user from just downloading the source code? The MIT license applies to them, too. Sure, they aren't guaranteed the source code of a derived work, but they are informed about the origin of the MIT-licensed source code (per the terms of the MIT license). The problem with distinguishing between "publisher" and "end user" here is that - under the free software and open source philosophies - the two roles are one and the same. Anyone end user can be a publisher. It's thus nonsensical to claim that the MIT licenses (and other non-copyleft free software licenses) themselves somehow "remove" freedom from anyone.
- zizee 10y ago
  > What's stopping the end user from just downloading the source code? Ah, the ambiguities of the english language! You are talking about the original code, whereas I am talking about the derived work. If the Publisher of a derived work chooses to not grant permission, the End User does not have the freedom to look/modify/redistribute the code of software derived from MIT licensed work. > "publisher" and "end user" ... under the free software and open source philosophies - the two roles are one and the same This is incorrect. Whilst it is true that sometimes the same person is performing both roles, the the roles can be mutually exclusive. I can use a piece of GPL'd software without publishing it. I do it all the time with Linux. The GPL is about protecting a person using the software as an End User (not "use" as in using it to create a derived work). The GPL gives freedoms to one role, and takes from the other. Your statement: > Sure, they aren't guaranteed the source code of a derived work Is not compatible with: > It's thus nonsensical to claim that the MIT licenses ... "remove" freedom from anyone. It is clear to me that the ability for an End User to inspect, modify and redistribute the software that they are using is more free than not being able to inspect modify or redistribute at all. Edit: It is true that the end user hasn't "lost" anything (or had it "removed"), as they never had the thing before the derived work was published, but they are certainly less free under the two hypothetical situations. Once again, the english language proves to be a tricky thing for me to wield, not matter how much care I take!
  - yellowapple 10y ago
    > Ah, the ambiguities of the english language! You are talking about the original code, whereas I am talking about the derived work. Right, but when talking about freedoms regarding a specific work, the derivative works aren't really relevant, since my (as an end user) freedom to use/modify/redistribute/etc. the original work remains unhindered. Likewise, my choice to release software under an MIT-like license does not in and of itself detract from end users' freedom; what downstream projects do with my code is not my concern so long as my in-code copyright notices are maintained. For all I know, my code is being used in projects with GPL or MPL or CDDL or Artistic License 2.0 or BSD-style or Apache or what have you licensing terms. > the roles can be mutually exclusive. That's not what "mutually exclusive" means. "Mutually exclusive" means that you can either be an end user or a publisher - never both. The whole philosophy behind FOSS dictates the precise opposite of mutual exclusion; the "Four Freedoms" (https://en.wikipedia.org/wiki/The_Free_Software_Definition https://en.wikipedia.org/wiki/The_Free_Software_Definition) make this very clear. Sure, you might happen to be one or the other, but the point is that free software empowers you to be both at the same time. By restricting one's freedom as a "publisher" (as you say), you're inherently restricting one's freedom as an "end user".
    - zizee 10y ago
      > the derivative works aren't really relevant Um... why not? If I am the end user of a work derived from some GPL code, then having the freedom to see/modify and re-distribute that code is valuable to me. From the very start of this thread I have been talking about protecting the freedom of the end user. Perhaps an example can help? But first, let me give an example of what I mean by "End User". I do not mean a programmer who "uses" GPL'd code by making changes to that code to make a derived work to redistribute. I mean someone who uses a piece of software to perform a task. e.g. Using LibreOffice to type up a document. Now for an example of a derivative work where GPL adds freedom for the end user, but takes it away from the publisher: I use Ubuntu on my desktop as an End User. Ubuntu is a derived from Debian, which is derived from GNU/Linux, a GPL licensed work. Because Debian and Ubuntu are derived from GNU/Linux, they must also be licensed under the GPL. As an End User of Ubuntu, I have the freedom to examine/modify Ubuntu's source code. The Ubuntu Team has to let me do it because of the GPL. If GNU/Linux was licensed under MIT, Ubuntu could be published as a closed source, proprietary software. I could still use Ubuntu, but I would not be free to look at the source code, modify it or redistribute it. True, with Ubuntu licensed as GPL, I am not free to package up my own closed source variant of Ubuntu, but neither would I have the freedom to do so if Ubuntu was closed source. > That's not what "mutually exclusive" means. "Mutually exclusive" means that you can either be an end user or a publisher - never both I concede that I have used that term incorrectly. Perhaps I am wrong in thinking this, but I suspect most people would have been able to glean what I was attempting to convey, but perhaps my use of english is too clumsy for me to communicate what I wish to convey. > By restricting one's freedom as a "publisher" (as you say), you're inherently restricting one's freedom as an "end user". What I am saying, and have been from the start of this thread, is that GPL is designed to put some freedoms above other freedoms. It's like the saying "Your Liberty To Swing Your Fist Ends Just Where My Nose Begins". Unlimited freedom cannot exist for everybody, under all conditions. GPL puts more importance on the freedom to operate/examine/tinker than to publish. The fact that an individual can have both their freedom increased and decreased at the same time does seem paradoxical, but when you realise different freedoms can be in conflict with each other, it does make sense. Ok, I think I've probably written enough. I do understand what you are saying. The GPL takes away some freedoms. I hope that what I have written has convinced you that it grants you some freedoms at the same time. Cheers and have a great night!
pluma 10y ago
GPL is not about the end user. GPL is about the software. The end user doesn't care about modifying the source code. And if the end user did care, the end user would also care about being able to chose a permissive license for their modifications. Because then the end user would be the same as the "publisher" you're talking about. This means there is no dichotomy between the "end user" vs the "publisher". The GPL is about the freedom of the software, not the rights of the user. It effectively proclaims software as a common good that is inherently unownable. It does this for a philosophical ideal of a greater good (the popular comparisons with communism are obviously hyperbolic but not entirely unreasonable). Making this about the "end user" vs the "publisher" is incredibly intellectually dishonest. MIT and friends grant the user of the code more rights by definition because the user can re-publish their modifications under a different license (without impacting the original code so nobody "loses" anything). The irony is that this actually makes the GPL a useful tool for commercial publishers because it prohibits competitors from creating private extensions to the GPL code without making them public -- while the original copyright holders can still keep their non-GPL copy of the code around and make private extensions to that without having to obey the GPL (as long as they don't include any modifications others have made to the GPL code directly). Heck, the common standard practice of corporate open source projects is to have all external contributors sign a CLA that effectively grants the company direct rights to the modifications outside of the project's license, allowing them to redistribute the modifications under any license they see fit. The GPL is a restrictive license. It makes more restrictions on how the licensee can use the licensed work. This serves a moral ideal, yes, but don't kid yourself that this isn't restrictive. It's like saying democracy is freer than anarchy. You might be correct in a philosophical sense (because in a democracy there is a system in place protecting the liberties you are afforded) but saying that democracy affords less restrictions would be absurd.

by zizee 10y ago

yellowapple 10y ago

> MIT [...] removes freedom to look and modify code from the end user. Er, what? What's stopping the end user from just downloading the source code? The MIT license applies to them, too. Sure, they aren't guaranteed the source code of a derived work, but they are informed about the origin of the MIT-licensed source code (per the terms of the MIT license). The problem with distinguishing between "publisher" and "end user" here is that - under the free software and open source philosophies - the two roles are one and the same. Anyone end user can be a publisher. It's thus nonsensical to claim that the MIT licenses (and other non-copyleft free software licenses) themselves somehow "remove" freedom from anyone.

zizee 10y ago
> What's stopping the end user from just downloading the source code? Ah, the ambiguities of the english language! You are talking about the original code, whereas I am talking about the derived work. If the Publisher of a derived work chooses to not grant permission, the End User does not have the freedom to look/modify/redistribute the code of software derived from MIT licensed work. > "publisher" and "end user" ... under the free software and open source philosophies - the two roles are one and the same This is incorrect. Whilst it is true that sometimes the same person is performing both roles, the the roles can be mutually exclusive. I can use a piece of GPL'd software without publishing it. I do it all the time with Linux. The GPL is about protecting a person using the software as an End User (not "use" as in using it to create a derived work). The GPL gives freedoms to one role, and takes from the other. Your statement: > Sure, they aren't guaranteed the source code of a derived work Is not compatible with: > It's thus nonsensical to claim that the MIT licenses ... "remove" freedom from anyone. It is clear to me that the ability for an End User to inspect, modify and redistribute the software that they are using is more free than not being able to inspect modify or redistribute at all. Edit: It is true that the end user hasn't "lost" anything (or had it "removed"), as they never had the thing before the derived work was published, but they are certainly less free under the two hypothetical situations. Once again, the english language proves to be a tricky thing for me to wield, not matter how much care I take!
- yellowapple 10y ago
  > Ah, the ambiguities of the english language! You are talking about the original code, whereas I am talking about the derived work. Right, but when talking about freedoms regarding a specific work, the derivative works aren't really relevant, since my (as an end user) freedom to use/modify/redistribute/etc. the original work remains unhindered. Likewise, my choice to release software under an MIT-like license does not in and of itself detract from end users' freedom; what downstream projects do with my code is not my concern so long as my in-code copyright notices are maintained. For all I know, my code is being used in projects with GPL or MPL or CDDL or Artistic License 2.0 or BSD-style or Apache or what have you licensing terms. > the roles can be mutually exclusive. That's not what "mutually exclusive" means. "Mutually exclusive" means that you can either be an end user or a publisher - never both. The whole philosophy behind FOSS dictates the precise opposite of mutual exclusion; the "Four Freedoms" (https://en.wikipedia.org/wiki/The_Free_Software_Definition https://en.wikipedia.org/wiki/The_Free_Software_Definition) make this very clear. Sure, you might happen to be one or the other, but the point is that free software empowers you to be both at the same time. By restricting one's freedom as a "publisher" (as you say), you're inherently restricting one's freedom as an "end user".
  - zizee 10y ago
    > the derivative works aren't really relevant Um... why not? If I am the end user of a work derived from some GPL code, then having the freedom to see/modify and re-distribute that code is valuable to me. From the very start of this thread I have been talking about protecting the freedom of the end user. Perhaps an example can help? But first, let me give an example of what I mean by "End User". I do not mean a programmer who "uses" GPL'd code by making changes to that code to make a derived work to redistribute. I mean someone who uses a piece of software to perform a task. e.g. Using LibreOffice to type up a document. Now for an example of a derivative work where GPL adds freedom for the end user, but takes it away from the publisher: I use Ubuntu on my desktop as an End User. Ubuntu is a derived from Debian, which is derived from GNU/Linux, a GPL licensed work. Because Debian and Ubuntu are derived from GNU/Linux, they must also be licensed under the GPL. As an End User of Ubuntu, I have the freedom to examine/modify Ubuntu's source code. The Ubuntu Team has to let me do it because of the GPL. If GNU/Linux was licensed under MIT, Ubuntu could be published as a closed source, proprietary software. I could still use Ubuntu, but I would not be free to look at the source code, modify it or redistribute it. True, with Ubuntu licensed as GPL, I am not free to package up my own closed source variant of Ubuntu, but neither would I have the freedom to do so if Ubuntu was closed source. > That's not what "mutually exclusive" means. "Mutually exclusive" means that you can either be an end user or a publisher - never both I concede that I have used that term incorrectly. Perhaps I am wrong in thinking this, but I suspect most people would have been able to glean what I was attempting to convey, but perhaps my use of english is too clumsy for me to communicate what I wish to convey. > By restricting one's freedom as a "publisher" (as you say), you're inherently restricting one's freedom as an "end user". What I am saying, and have been from the start of this thread, is that GPL is designed to put some freedoms above other freedoms. It's like the saying "Your Liberty To Swing Your Fist Ends Just Where My Nose Begins". Unlimited freedom cannot exist for everybody, under all conditions. GPL puts more importance on the freedom to operate/examine/tinker than to publish. The fact that an individual can have both their freedom increased and decreased at the same time does seem paradoxical, but when you realise different freedoms can be in conflict with each other, it does make sense. Ok, I think I've probably written enough. I do understand what you are saying. The GPL takes away some freedoms. I hope that what I have written has convinced you that it grants you some freedoms at the same time. Cheers and have a great night!

pluma 10y ago

GPL is not about the end user. GPL is about the software. The end user doesn't care about modifying the source code. And if the end user did care, the end user would also care about being able to chose a permissive license for their modifications. Because then the end user would be the same as the "publisher" you're talking about. This means there is no dichotomy between the "end user" vs the "publisher". The GPL is about the freedom of the software, not the rights of the user. It effectively proclaims software as a common good that is inherently unownable. It does this for a philosophical ideal of a greater good (the popular comparisons with communism are obviously hyperbolic but not entirely unreasonable). Making this about the "end user" vs the "publisher" is incredibly intellectually dishonest. MIT and friends grant the user of the code more rights by definition because the user can re-publish their modifications under a different license (without impacting the original code so nobody "loses" anything). The irony is that this actually makes the GPL a useful tool for commercial publishers because it prohibits competitors from creating private extensions to the GPL code without making them public -- while the original copyright holders can still keep their non-GPL copy of the code around and make private extensions to that without having to obey the GPL (as long as they don't include any modifications others have made to the GPL code directly). Heck, the common standard practice of corporate open source projects is to have all external contributors sign a CLA that effectively grants the company direct rights to the modifications outside of the project's license, allowing them to redistribute the modifications under any license they see fit. The GPL is a restrictive license. It makes more restrictions on how the licensee can use the licensed work. This serves a moral ideal, yes, but don't kid yourself that this isn't restrictive. It's like saying democracy is freer than anarchy. You might be correct in a philosophical sense (because in a democracy there is a system in place protecting the liberties you are afforded) but saying that democracy affords less restrictions would be absurd.