1. 11
  1. 11

    Stallman’s co-opting of the word “freedom” was a stroke of genius. To those uneducated about the specific definition of freedom used by the FSF, the way he uses it makes the LLVM folks seem like they’d rather not share their code at all. Of course, that’s not the case. For example, I put my code into the public domain precisely because I think that’s the fullest expression of freedom possible. But not according to Stallman.

    I certainly don’t agree with Stallman’s values, but I’m all for freedom. Just not the kind that has stipulations on distribution of derivative works.

    1. 7

      “The fullest expression of freedom” is a ridiculous goal; it means whatever you want it to mean. You see the same fight over the definition of freedom in debates about closed shops vs. Right to Work legislation, and it’s just as pointless there.

      I don’t care about what freedom means in the abstract sense. I care about results, and the FSF’s principles have done a pretty good job of obtaining the results I want (ensuring the continued existence of a modern software system that I can study and modify).

      1. 5

        I care about results

        Me too… Specifically, I care about maximizing the number of people using my software. Which is one reason why I use copyfree licenses. I’ve never had someone tell me they won’t use my software because it’s permissively licensed, but I’ve bumped up against that with GPL licenses.

        “The fullest expression of freedom” is a ridiculous goal; it means whatever you want it to mean.

        To each their own, I guess. I care about it and I don’t think it’s ridiculous.

        1. 3

          Specifically, I care about maximizing the number of people using my software.

          That’s something actual that we can look at, discuss and determine what effect different licenses have. (And I agree that it is a totally valid goal, and that if that is your goal you should not license your software under the GPL.)

          But we can’t discuss what effect different licenses have on “freedom”, because we’re not going to agree on what that word means.

          1. 4

            because we’re not going to agree on what that word means

            I think that’s because of Stallman and the FSF co-opting the word “freedom”. Almost all dictionary definitions are about the ability to act without restraint or restriction, which is opposite of the FSF “freedom”, which explicitly puts restrictions on what you can do.

            I have no with the FSF existing, and attempting to get people to share their code, forcibly if necessary. I have a problem with their redefinition of “freedom” and moralistic stance on the availability of code. They’re the reason we can’t have a discussion on the effects of licenses on freedom, because they took a word and mutated it’s meaning, to something almost opposite of it’s original definition, like the word “literally”, but with a cohesive group actively working to redefine it.

            1. 3

              They didn’t mutate its meaning, they’re just looking at a larger scope than you are.

              If I’m restrained from assaulting you, I’m less free. But if we expand the scope of our search so that you’re included…

              (This debate over the meaning of freedom is much older than the FSF.)

              1. 3

                I’ve been mulling that over for the past eight minutes trying to think of a way to refute that point, or wheedle in something in without being perceived as overly argumentative for the purpose of being argumentative. I’ve got nothing, however it’s lead me to some thoughts I’ve been having for a while…

                Mainly, and I have no idea if you have any insights on this, that it seems odd that the FSF doesn’t explicitly come out and say “that capitalism thing, it’s not all that great now, is it?”. The goals are inherently anti-free market, but nobody affiliated with the FSF is willing to say that. It just seems odd to me.

                1. 4

                  The goals are inherently anti-free market

                  What makes you say that? Even if one takes the extreme position that proprietary software is morally wrong (and you don’t have to believe that to support the FSF), there are lots of opportunities for free enterprise in software that don’t involve selling licenses.

                  (P.S. This is exactly the kind of meaty discussion I think the subject deserves. It’s way better than the usual “lol rms is weird”.)

                  1. 2

                    I had a big long thing on how if the world were economically efficient that with Free Software the price of everything would eventually be driven down to effectively nothing but it wasn’t really going anywhere, and that’s kinda true with most things, minus the fact that in the “real world” things aren’t created out of nothing, and the resources have to be paid for, also collusion is typically frowned upon…

                    The core of what makes me say Free Software is inherently anti-free market is that with a minimum of collusion it isn’t difficult to drive down the prices as much as possible, and the moralistic viewpoint pushes collectivism and production of software not for capital gains but communal gains.

                    Edit: Also, the GNU Manifesto states:

                    The real reason programmers will not starve is that it will still be possible for them to get paid for programming; just not paid as much as now.

                    In a free market wouldn’t those programmers would be free to keep the enhancements they make private and sell them to the highest bidder, instead of having to supply them back for the collective good?

                    Edit2: (also: lol rms is silly)

                    1. [Comment removed by author]

                      1. 2

                        Isn’t it more that the free market will help society reach an equilibrium, without any coercion? Because copyleft is entirely coercive in nature. It looks to me (and I could be wrong) that Free Software requires a non-free market.

                        I didn’t say they were anti-capitalist, but anti-free market (which I suppose could be equated to the same thing). Because the source code changes have to be released back to the community, that makes it seem much more like social ownership than private ownership, which is not a hallmark of capitalism but socialism, which still makes it seem anti-free market.

                        1. 1

                          How is the private model any better? Are you arguing that one should be able to modify non-free software (windows for instance) and redistribute it? I fail to see your point. Copyleft only describes what you are allowed to do with software that someone else created, and if you believe in intellectual property, should control. FSF has argued against the Pirate Party’s plan to eliminate software copyright, indicating that they are strong supporters of ownership of software. I’d like to understand how you think software should be distributed that would result in better “social equilibrium without any coercion”.

                          1. 2

                            What if I said intellectual property was incompatible with freed markets? (Stated differently, intellectual property requires a non-free institution to enforce, like a government.)

                            1. 1

                              I’m sorry I used the term “Intellectual Property”. It’s pretty meaningless. What I meant was specifically copyright. All rights require a non-free institution to enforce, don’t they?

                              1. 2

                                Copyright falls under the umbrella of intellectual property. I see intellectual property as the body of law that governs the ownership of ideas. I reject that ideas can be owned, and therefore reject all IP (trademarks, copyright, patents, etc).

                                All rights require a non-free institution to enforce, don’t they?

                                I don’t really want to get in a big debate on this, but no, I don’t believe so. Mostly because I believe all human association ought to be voluntary. (One conclusion you can draw from this is that we are therefore all responsible for our own defense, which includes appointing someone else to act on your behalf.)

                            2. 1

                              Did I say the private model is better? Did I say anything about software distribution and creating a social equilibrium without coercion? I think you’re either selectively reading what’s been posted, or trying to twist my words to gain arbitrary internet points.

                              Social ownership can coexist with intellectual property, it just means that the “intellect is own by society”, or controlled by it, just like any other form of socially controlled property. It especially doesn’t mean the revocation of intellectual property rights (though that could be a part of it, but that’s true of private ownership too). I’ve even referenced the FSF’s position against the Pirate Party’s shortening of copyright length in this thread.

                              The “social equilibrium without any coercion” thing you’re trying to quote, but failing to grasp, is the ideal free market. The goal is an equilibrium created without any coercion by outside entities. That seems contrary to the FSF’s goals of collectively making better software, and using copyleft as a means to foster the collective building of better software.

                              The GPL can coexist with the free market, but that doesn’t mean it inherently supports it. The way the Free Software communities exist is basically as a gift economy, while there are people that get paid to support Free Software, the community around it and a good deal of contributions are “given without an explicit agreement for immediate or future rewards” (which is in the first sentance on Wikipedia’s Gift Economy entry. Copyleft is even referenced in the Gift Economy entry).

                              1. 2

                                I assure you that I don’t keep track of internet points, lol. One thing you’ve misunderstood is the goal of the FSF. It is not intended on creating better software, it is intended on preserving freedom. From a few of your misconceptions I’m starting to think that you don’t understand what you are arguing against. RMS has said time and time again that he would rather see inferior, but free software. You can disagree with that, if you want, but you can’t misrepresent the view you are arguing against. This is the second popular fallacy that you have espoused on this thread. It seems you haven’t even had a cursory read of the FSF FAQ.

                                As far as your last paragraph goes, I’d say it’s purely academic. Just because I agree to paint someones house for free doesn’t mean painting is a gift economy. I don’t know where you get the idea that most people are working for free, that isn’t my experience at all. Quite the contrary. Most of the important contributions to open source have been paid for and paid for very well. The key difference is that you must get paid to do the work, rather than own the copyright to the work. In most cases, this is exactly the way it works anyway, with or without free software. Most engineers I know don’t get royalties on all the software they create. Instead, some company needed software written, and they paid the engineer to write it. For most engineers there is an explicit and immediate reward for creating free software.

                                1. 1

                                  While RMS would rather use inferior but free software, that doesn’t make their goal to not have high quality free software. The two are not mutually exclusive. They would prefer to get free software globally first (which is what I disagree with), but that doesn’t mean high quality isn’t also a goal. Especially because they acknowledge that it’s difficult to get average consumers to switch with lower quality software.

                                  It’s not purely academic. Painting someones house for free is acting in a gift economy. Quite right, just because you paint one house as a gift doesn’t mean the entire industry of painting houses (or even painting) becomes a gift economy. Similarly, just because a subset of software is being developed using a gift economy doesn’t mean that all software is developed using a gift economy. And while there are people getting paid to develop Free Software, almost every piece of Free Software exists because the original creators gifted it to the community. Without the initial gift there wouldn’t be people making money. They can coexist.

                                  Following, even if people are getting paid to make contributions by their company, the company is acting inside of the gift economy by gifting the changes they paid for to the community.

                                  You’re also once again confusing Free Software and Open Source. Which is really silly to me, because Open Source is a counter movement to Free Software, and you seem to be arguing for Free Software.

                        2. 1

                          Nothing in the GPL says you can’t sell your changes and nothing says that you have to share them with anyone. The only thing you aren’t allowed to do is control it after it leaves your hands. In other words, you can’t control how your customers use it. I’ve been paid quite a bit to work on open source software in my lifetime, much of which was never shared back to the community.

                          1. 1

                            From my understanding, section 10 of the GPL states that the downstream (the guys who made the software you’ve modified) are given a license to “run, modify, and propagate that work”, meaning they can ask you for your source at any time, and are free to put it back into the main software, or share changes provided they follow the procedures in the GPL.

                            Also, open source is a counter movement to Free Software, working on open source doesn’t automatically mean you’re working with Free Software, and because open source is more encompassing, it doesn’t require you share your changes…

                            1. 2

                              I’m sorry to say that your understanding is wrong. RMS has pointed out this fallacy on many occasions. Sorry to conflate the term “Open Source” with Free Software. There are many people running modified versions of GPL software as web services that haven’t shared their modifications. The GPL is not about protecting developers, it’s about protecting consumers of software. It just so happens that we are all consumers, even developers.


                              [edit] added link

                              1. 1

                                It also says that if the web services can be used by the public the source needs to be available to those users. It’s only acceptable to not release the sources if they’re used entirely internally, where the source is essentially available to the parent entity (the company) already.

                                1. 3

                                  Again, you are dead wrong. Please stop. If that were the case every major company in the world would have to release all their code. There is a license you can use that says that, but it’s not the GPL and I haven’t seen it actually used anywhere.


                                  [edit] Added link to license that NOONE uses.

                                  1. 1

                                    The link you posted says that. You’re selectively reading again to only support your viewpoint, which is flawed. Cite where it says you’re okay to not convey the source.

                                    an organization can make a modified version and use it internally without ever releasing it outside the organization.

                                    But if you release the modified version to the public in some way, the GPL requires you to make the modified source code available to the program’s users, under the GPL.

                                    The modified version is released to the public by being used by the public. It’s no longer private. Your link says that.

                                    The GPL also says so in sections 2 and 6 (2: source must be conveyed; 6: if you have object code running on a server, you need to have the source available somewhere).

                                      1. 2

                                        That’s the FAQ about the GPLv2. It even says that they wanted to address it in v3 (not that it was).

                                        Both the v2 and v3 FAQ have the following:

                                        It is essential for people to have the freedom to make modifications and use them privately, without ever publishing those modifications. However, putting the program on a server machine for the public to talk to is hardly “private” use, so it would be legitimate to require release of the source code in that special case.

                                        Which basically says they don’t like it, but they don’t have the wording for it (v2 explicitly says they don’t have the wording for it).

                                        Similarly, releasing those changes back is in the spirit of the license, if not the wording.

                                        1. 3

                                          That is specifically what the AGPL is meant to address. No one uses it. There are multiple licenses people can use. The FSF doesn’t mandate using one license. Many people use the LGPL which allows linking with proprietary code. All those licenses are considered “free” licenses and good licenses by the FSF.

                                          The fact remains that what you said is completely false. No company that uses GPLv2 or GPLv3 code is compelled to release their source simply by using it on a public web server. This should be fairly obvious since almost every company on earth is using some GPL code on their production stack. I don’t know whether you generally don’t know this, or if you are spreading FUD, but I suggest you learn more about FSF before speaking about it with such authority. Everything I’ve corrected you on is in the FAQ.

                                          1. 1

                                            You seem to be mistaken about exactly how the GPL infects. Because there’s a lot of times when you’re permitted to use GPL software and have it not infect your software. Using GPL code in your stack most of the time doesn’t infect your application. The FAQ says that. So maybe you should read your own FAQ and the license’s text.

                                            Also, fascinating that you’ve decided to only go down the nitpicking road, instead of the more interesting and meaty discussion involving the Free Software being not a market economy.

                      2. 3

                        Mainly, and I have no idea if you have any insights on this, that it seems odd that the FSF doesn’t explicitly come out and say “that capitalism thing, it’s not all that great now, is it?”

                        I’ve certainly had those thoughts too. But I’m not sure if they see it that way.

                        I’ve personally had many long discussions with extremely ardent supporters of the GPL, the FSF and its mission. There’s an ungodly amount of back and forth, but eventually, we get to the topic of intellectual property and whether coercion ought to be used to impose it. Interestingly, most, if not all such supporters agreed with me that intellectual property was unjust and should be abolished. But how could this be? Copyleft fundamentally requires intellectual property to work. But it turns out, they see the use of copyleft as a necessary evil to fight back against intellectual property. I think they see it as a sort of leveling the playing field.

                        At this point, we come together in (some?) agreement over philosophy but a disagreement over how it should be achieved.

                        I guess I don’t have any real point here, but just trying to share an experience. It’s obviously impossible to extrapolate it across all FSF supporters (or even the FSF itself), but I found it to be an interesting similarity between the hardcore FSF supporters.

                        1. 1

                          I’m not entirely sure that RMS is against intellectual property, because copyleft is dependent on copyright. He wrote an essay against the Swedish Pirate Party’s position on shortened copyrights because they would put things in the public domain, where they can no longer be copyleft.

                          It seems silly to me. Copyleft is a total hack on copyright, but because the FS movement took the moralistic stance they couldn’t exist without copyright, and copyright lasting for a good long time. It also means there needs to be someone actively maintaining the software to make sure people know about it and have relatively easy access to it.This is one of the main reasons I prefer Open Source’s ideas. Plus it aligns nicely with my views on copyright and IP in general.

                          1. 2

                            Sorry for the late comment, but funnily enough, I just ran into exactly what I described in my GP comment.

                      3. 1

                        I don’t think the comparison of copyleft with “restraining from assaulting someone else” is relevant. I agree that in both cases my liberty is reduced, but there is still a big difference.

                        By assaulting someone, I cause harm to this person and I reduce his/her liberty to be safe and healthy. This is the reason why it looks necessary to reduce my liberty to assault a third person to guarantee the liberty of that third person.

                        By reusing a free software in a non-free software, I’m not reducing the liberty of free software people to do what they want.

                        1. 4

                          You’re technically reducing the liberty of anyone else who might want to use those improvements.

                          I went down the same line of thinking myself (and several other lines), there’s a response from the “Free Software” viewpoint, and I don’t completely disagree: it is a broadened look on freedoms, and restricting use to keep it open and available keeps everyone else’s freedom to modify and play with it open.

                          I don’t think it’s necessary, bit it’s more free, from a certain point of view.

                          1. 2

                            I suspect bct’s link to negative freedom elsewhere in this thread is relevant here. The FSF version of freedom is very much more in line with the notion of positive liberties, whereas the copyfree version of freedom is more in line with the notion of negative liberties.

                            I don’t believe either one correspond that well to the colloquial definition of freedom which simply means “less/without restriction” without any other qualifiers.

                            With that said, in my view, there are at least three definitions of freedom at play here.

                            1. 2

                              I can’t tell if you’re agreeing with me or not because we have too many meaning associated with “freedom”… It’s… a conundrum.

                              1. 2

                                I think I was just adding some extra clarification. But based on your other comments in this thread, we seem to be in general agreement.

                                1. 2

                                  Okay, cool.

                      4. 2

                        I don’t think the problem is with having different definitions of “freedom”. I think that Stallman agrees with freedom being “the ability to act without restraint or restriction”.

                        The problem is that Stallman promotes user freedom while non-copyleft license like BSD/MIT promotes programmer freedom. Stallman wants a future where every computer programs are free. But to achieve this goal, he wants to prevent non-free software from using free software, in order to slow down as much as possible the development of non-free software.

                        This is paradoxical because to promote freedom, he has to withdraw the liberty of using a free software with a non-free one :)

                        1. 2

                          It’s not paradoxical. No freedoms are removed or restricted, it’s entirely his (and anyone else who follows that philosophy’s) choice to not use non-free software.

                          I’m also not entirely sure that non-copyleft licenses like the BSD/MIT promote programmer freedom while copyleft licenses promote user freedom, because the entire purpose of copyleft is to let programmers play with other programmer’s changes/improvements.

                          I can see the FSF/other Free Software peoples' point, I just disagree…

                          1. 5

                            That’s a programmer-centric viewpoint. The fundamental goal of FSF to guarantee that users have control over the software they run. They may use that control directly if they are programmers, or they may hire programmers if needed, but either way the actual benefit is aimed at users. A social movement that only improved the lives of programmers would be pretty sadly limited…

                            1. 2

                              It’s not paradoxical. No freedoms are removed or restricted, it’s entirely his (and anyone else who follows that philosophy’s) choice to not use non-free software.

                              What I meant is that a GPL licensed software removes my “freedom” to combine this software with another non-free software. This is what I consider paradoxical. The goal of the GPL is “to guarantee that I have control over the software I run” (like wrs wrote in another comment), but to achieve this goal I’m forbidden to combine this software with a non-free one. I’m not saying this is good or bad, I just think this is an interesting situation. I can do anything I want with a GPLed software, except combining it with a non-free software. Whatever definition of freedom is used, this clearly reduces my freedom on the short term, even if I understand the long term goal of the FSF (to give a kind of competitive advantage to free software and level the playing field).

                              I’m also not entirely sure that non-copyleft licenses like the BSD/MIT promote programmer freedom while copyleft licenses promote user freedom, because the entire purpose of copyleft is to let programmers play with other programmer’s changes/improvements.

                              I agree. Copyleft licenses promote user and programmer freedom (not just the user). What I wrote in my previous comment was ambiguous.

                        2. 2

                          We can’t discuss what effect different licenses have on “freedom”, because we’re not going to agree on what that word means.

                          Indeed, you’re 100% correct. I suppose I was being too coy in my initial comment. :-) It’s similar to the criticism I was lodging against Stallman.

                          1. 2

                            Sure, and I think that his insistence on the word freedom is a mistake that encourages his detractors to engage on an abstract level instead of discussing concrete goals & results.

                    2. 3

                      “They hate us for our freedom.”

                    3. 5

                      Good ol' RMS… Freedom is only freedom if you prevent people from using it non-freely.

                      1. 2

                        Freedom as such is a difficult concept. Paradoxically restrictions are needed to maintain freedom, exactly for the same reason we need fair rules and just governments.

                      2. 2

                        I’m confused, ESR says “They want to build IDEs and other tools that share the compiler’s code. GCC policy won’t let them do that.”

                        Why can’t the GNU project support GCC being built to enable such tools, but license it such that tools based on it have to in-turn be free?

                        (I’m using free in the FSF sense here.)

                        1. 4

                          We’re talking about technical architecture here, not licensing. If you provide an interface through which things outside of GCC can use all the information GCC has internally, you can (Stallman fears) use that interface to write a new front-end, or optimizer, or whatever. That means that you can write a proprietary front-end for a new language to GCC and claim that it’s a separate program from GCC. This would allow replacing GCC piecemeal with non-copyleft or even proprietary components. And, of course, once it’s non-copyleft, you can make a proprietary derivative.

                          I think it’s probably possible to provide the information that IDEs need without opening GCC up to proprietary frontends.

                          1. 1

                            I hadn’t considered the piecemeal replacement angle, it’s a fair point.

                            1. 1

                              It also actually happened: NeXT tried to build their Objective-C compiler as a proprietary frontend on top of the GCC backend, claiming that the two were separate programs. Under FSF legal threats they backed down, and that’s why GCC supports Objective-C today. It’s probably also why Apple backed Clang.