1. 39
  1.  

  2. 24

    Although the copyleft licenses were (and in some ways, still are) a bit radical and the founder somewhat controversial, it’s hard to overstate the monumental positive impact that GNU has had on free and open source software and the Internet itself. Happy birthday, GNU.

    1. 2

      My contrarian opinion from my observations in bioinformatics is that people like unencumbered licenses like MIT or Apache better than GPL. These licenses better capture the original hacker spirit of “do what you want, don’t sue”

      Indeed when I cast about for libraries for hobby projects I’m biased towards these permissive ones, just because I don’t have to worry about legal issues and what happens in the future.

      In this respect I view the GPL as a bit like communism: an idea that looked good on paper and in academic circles but missed important aspects of the real world.

      1. 13

        I used to like permissive licenses a lot more than restrictive licenses like this, but now I like copyleft better (for my own projects) because…

        1. You’re still helping education, nonprofits, and individuals benefit from your work.
        2. It’s still open source, so people will still be able to contribute and use your work in their own projects.
        3. Companies that want to use your code to make money can do so, but only if they also help out the other “worthy” causes by contributing changes back.

        In fact, I might even consider something more radical like a YUMMY license (you make money, I make money) - which has all the same benefits of being open source and helping worthy causes, except at least you get to make money when someone uses your work to do something that you might not even want, like selling ads.

        1. 5

          I think a lot of people want the “no-fuss” licenses like MIT or Apache but later misunderstand what some of that no-fuss means when someone else raises a fuss.

          You can grab MIT-licensed code and add all sorts of restrictions on top of it. When someone recently added restrictions that forbade Microsoft and other companies that are helping putting migrant children in detention camps from using this code, many people thought that this was a violation of the MIT license in some way, that you couldn’t add restrictions such as those. Of course, adding restrictions is a right granted by the MIT license. If you wanted a license that could have prevented this, pick a copyleft license.

          People say they don’t want copyleft because it’s too complicated, but it only reflects an inherent complication in the world; it didn’t originate it.

          1. 1

            You can grab MIT-licensed code and add all sorts of restrictions on top of it.

            Yes, that is post modification. The original code remains as is. You added on top of the open sourced code and you can do whatever you want with the derivative work. That is the intent of the MIT/Apache licenses. I think Apache also gives you patent rights. These are truly free licenses.

            1. 2

              You don’t have to modify anything. The MIT license lets you sublicense with or without restriction. I can grab MIT code as-is and redistribute it under different terms as long as I keep the original copyright statement, not necessarily the original terms.

              https://softwareengineering.stackexchange.com/a/189704

          2. 5

            These licenses better capture the original hacker spirit of “do what you want, don’t sue”

            I can’t decide if the call-outs to some imaginary original hacker spirit is a No True Scottsman or Revisionist History.

            In this respect I view the GPL as a bit like communism: an idea that looked good on paper and in academic circles but missed important aspects of the real world.

            I don’t know if you missed it, but:

            • Linux, some BSDs, OSX, and others all use or previously relied upon the GNU userland and/or compiler.
            • GNU was hugely important in shaping the idea of open source and, in turn, the software industry.
            • A whole generation of programmers, academic and commercial, have collaborated on innovative GPL licensed software and made billions of dollars.
            • More to the point, there are successful businesses and business models built around GPL licensing

            What important aspects, in particular, do you think the GPL missed?

            1. 2

              Way before Stallman, people were sharing bits of code quite effectively with extremely permissive licenses (https://en.wikipedia.org/wiki/History_of_free_and_open-source_software#Free_software_before_the_1980s).

              As I have said before, the GPL is born out of idealogical rigidity, mostly from one person, who evangelized this based on his own personal experiences with a company (https://www.free-soft.org/gpl_history/).

              What important aspects, in particular, do you think the GPL missed?

              The level of control the GPL exercises over not just the current software but future versions of the software is impressive and, again, as I have said before, strangles commerce and cooperation.

              1. 5

                I am well aware of that history. If you’re saying that those people embody the “original hacker spirit,” I’d note the exact next section that talks about the decline of free software. It seems like that original hacker spirit sold out immediately after software became copyrightable.

                I sincerely doubt you were active in any of the 1940s - 1960s hacker scenes. So, what source are you getting your definition from? The MIT hacker scene? The Berkeley hacker scene? Very different ethos. Or do you mean the hackers of the ’80s? Or do you mean a Hacker as defined by the Jargon File?

                All software and all software licenses are ideological and political.

                The level of control the GPL exercises over not just the current software but future versions of the software is impressive and, again, as I have said before, strangles commerce and cooperation.

                You’ve said it, but you haven’t demonstrated it.

                Meanwhile, as we type, GPL software is being collaborated upon by a rich tapestry of individuals and organisations; used in the most critical systems of our modern world; having made many, and will again make many more, people rich beyond their dreams.

                1. 1

                  You’ve said it, but you haven’t demonstrated it

                  I have a feeling won’t be able to demonstrate this to your satisfaction. However, I have seen it demonstrated to my satisfaction, and I have understood why. I have heard arguments from legal, corporate and engineering roles. I have also seen this is in operation.

                  1. 5

                    Well, if use of a GPL license prevented a company from using some code, it sounds like Mission Accomplished.

                    1. 1

                      This is interesting. So in your opinion the GPL is explicitly for preventing commercial use? This is not the intent, however. The GPL assumes it’s clauses are sufficient for commercial use, though, as I’ve said, in practice the extreme control it presupposes (the “viral nature”) makes it difficult in practice for companies that are not just providing support for existing software.

                      1. 5

                        So in your opinion the GPL is explicitly for preventing commercial use?

                        My opinion is: if a company decides it can’t use GPL software because they can’t obey its conditions, then Mission Accomplished!

                        Especially since the GPL, in particular, is legally very well understood.

                        The GPL assumes it’s clauses are sufficient for commercial use, though, as I’ve said, in practice the extreme control it presupposes (the “viral nature”) makes it difficult in practice for companies that are not just providing support for existing software.

                        “The GPL” assumes nothing. It’s a legal license. Have you read the GNU manifesto? It speaks, directly, to what kind of commercialisation RMS had in mind, back in ’85.

                        And the GPL makes what difficult in practice? What commercial use, in particular, are you talking about? Because, as I’ve already pointed out, there is heaps of commerical use involving GPL software.

                        You raise the “viral nature.” I defer to a well written article on this very topic, linked from the Philosophy section of the GNU website.

                        1. 1

                          The basic issue is the incorporation of GPL software into proprietary software. If some one writes an efficient math library and releases it permissively then companies using it will freely incorporate it into their own stack.

                          GPL-think says companies will now keep all innovations to themselves and “steal” the code. In practice, the reason why the company used the code is that they did not have sufficient resources in that area of expertise. However, they often will pragmatically realize that it is more efficient to contribute fixes and enhancements back to the original library - which is being actively worked on, often by other companies too - rather than keeping their own separate fork which falls behind and becomes a pain to keep up to date.

                          If the library was written in GPL now there is legal in the loop, and you can’t incorporate the code into proprietary code without exposing your internals. There is also some tip-toeing around by engineering so that the code doesn’t become more embedded in proprietary code (like taking functions etc).

                          1. 2

                            GPL-think says companies will now keep all innovations to themselves and “steal” the code.

                            It’s not about the company keeping changes to the code to themselves, it’s that by building a proprietary software product they have done evil, by using my code to do it they have made me complicit in their evil.

                            1. 2

                              Thank you for a very precise description for why you use the GPL.

                              However, this raises even more questions for me.

                              Can you explain why proprietary software is evil? Is proprietary hardware evil? Can free software run on proprietary hardware? What about books? What about paintings? What about ships and planes? Are buses evil?

                              1. 3

                                This is more than I can answer properly here, and the answers are more than 30 years of work by many people attempting to communicate about software freedom.

                                However, I will try to hit to broad strokes here. As mentioned above I believe a computer exists to serve its owner and that the user has a fundamental human right to control their computing. As such, proprietary software infringes this human right by both legal and (often) technical means.

                                Proprietary hardware is bad, but not so direct an infringement only because we do not yet have the facility for most users to have full control in this area anyway (fabbing silicone being something most could not do in their house, nor reasonably hire anyone to do).

                                Art is different again because proprietary art does not limit its use or the user’s control in the same way as proprietary software (unless it has DRM, which is of course evil). Even so, I believe freedom of expression should trump monopoly rights when it comes to creating new art and support remix culture, etc.

                                Ships/planes/buses are in much the same position as hardware, though of course these days some companies abuse DRM to bad ends even in vehicles like this and I would generally think of “right to repair” as also a human right that should not be infringed.

                                I know these small sentences cannot adequately capture the nuance of a multi-decade movement, but hopefully it shows a glimmer of what is going on.

                                1. 1

                                  Thanks @singployma

                                  Your answer is detailed, but I think the core is here:

                                  I believe a computer exists to serve its owner and that the user has a fundamental human right to control their computing.

                                  I would assume you would extend this to any tool. Since a computer is merely just another tool in a long lineage of tools we’ve had, starting with the stone axe.

                                  As such, proprietary software infringes this human right by both legal and (often) technical means.

                                  Not clear what right this is. If I sell you a stone axe I can actually attach any conditions of use on it I want. I can forbid you from attacking my family with it. I can forbid you from selling it to my rival so he can copy the design. I can forbid you from selling it to a rival tribe who may attack me with it. I can forbid you from selling your own version of it. You, of course can refuse these and not buy my axe.

                                  There is no “right”. Never has been. Always merely a license agreement.

                                  What differs from thing to thing is how easily you can violate our license agreement and how hard it is for me to enforce it.

                                  1. 1

                                    @singpolyma has more patience than me. Your questions are literally responded to in the links I gave earlier.

                                    You have, thus far, entirely refused to do either the basic reading or even demonstrated trying understand the counterarguments and counterexamples given to you.

                                    Whether you intend to or not, you are trolling.

                            2. 2

                              Sounds easy: just use fast middleware or use a fast microkernel or separation kernel. OKL4, which advertised exactly that benefit, was supposedly deployed in phones that sold a billion units or something before General Dynamics bought it. I didn’t track it after that. So, we have that for end-user devices plus the SaaS model as a few examples that have been working fine to generate huge fortunes from FOSS code without giving anything back or giving the tiniest slice of benefit back.

                              I think the GPL holding back proprietary software or forcing sharing like you describe is a myth except in situations where the performance or cost of integrating FOSS in one of these ways prevents management from doing it. It achieved something in those cases, sometimes a lot (esp Linux and Android).

                              Looking at all the high-overhead software out there, I doubt that’s a lot of situations given middleware is often already in use. All kinds of companies are using FOSS without giving anything back. It failed if its goal was what you said. They needed a stronger license.

                              1. 2

                                Hi @nickpsecurity !

                                I think there will always be users who don’t contribute back. It’s the sad truth of the world. Regardless of license. The hope with open source is that a commonly used infrastructure will get distributed maintenance and updation with low cost from individual users. Things like MIT/Apache sweeten the deal by not getting legal worried.

                                I have to understand the solution you propose. The LGPL lets you link libraries without having to give up the company jewels. But it requires returning derived works if used externally (I think). Kind of a half-way, reluctant compromise in response to the popularity of permissive licenses.

                                Would the solution you propose violate the spirit of the GPL? Wouldn’t that be cause for litigation? My understanding of the GPL is if you use GPLd code in your offering (external), you have to GPL your offering. There was some loop-hole to do with micro-services I think (yep: the ASP loophole), for which the AGPL was developed.

                                1. 2

                                  (I’m fairly sleep-deprived right now so just a quick, partial reply with more thorough one later if needed.)

                                  It’s how you link them that seems to determine whether the proprietary work becomes derivative of and/or also GPL. The few things I’ve seen on it say they have to communicate via user-space. The examples I mentioned have GPL software communicate with messaging through middleware or user-space via a microkernel. In those cases, the plumbing that makes messaging work will likely be derivative at least in GPL program but the combo would not. Also, the GPL isn’t enforced as aggressively as proprietary licenses, esp with patents. That means there’s low, overall risk of non-compliance that can hit a company’s balance sheet. The combo of methods that nullify common mechanisms of things going derivative plus lower risk of expensive litigation negates the supposed benefits of the GPL for freedom-loving contributors in these situations. The license essentially was too permissive in its wording if goal is to stop commercial exploitation giving nothing back.

                                  That’s why I found Parity and Prosperity of License Zero interesting in recent research. Parity focuses on contribution sharing by forcing that any changes or extensions that are made are resubmitted back regardless of whether they’re distributed. If you make a change, everyone can use it. Prosperity focuses on commercial angle by saying any commercial use, not distribution, needs a paid license at some point. I like the focus there on “use” for what gain versus distribution since it’s harder to legally cheat around commercial use if you’re provably making money with the software.

                                  I’m not saying they’re the end goal or anything. I just found their terms interesting in that they forced contributions to be shared and more-easily intercepted commercial freeloading. The GPL failed to do both in many cases. Projects aiming to deal with those might want to ditch or update GPL to handle that. AGPL was an attempt. So is Parity. Prosperity is a shared-source, not open, license whose tactics might be useful in open one. Just examples to consider and build on in future experiments.

                                2. 1

                                  All kinds of companies are using FOSS without giving anything back. It failed if its goal was what you said. They needed a stronger license.

                                  I’ve already covered elsewhere in this thread why I think “giving back” is not the point, however any weakness in effectiveness of the license as a strategy I think is less in the strength of the license, and more in the lack of enforcement efforts. There are only one or two small, underfunded non-profits working on community enforcement, and they can only enforce for the very small number of projects where copyright holders are interested in their efforts.

                2. 3

                  Businesses push for MIT and BSD licenses for a reason. Whenever you make a license decision make sure you think about who it ultimately benefits. Seems GPL worked great so far and I prefer to open my code with it. I’m not working for free here!

                  1. 1

                    I find the comparisons to communism interesting. It used to be said, at least here in the birth-country of Linux, that it’s communism because it’s meant to be free. As in free of charge. Linus dad’s political convictions, though they seem to have shifted, probably helped spread the meme here.

                    But it really is meant to be free as in libertarian free.

                    Yet it really isn’t, because it makes private forks harder.

                    There’s also that discrepancy between getting funding for free and open video playback and hardware making illegal forks of the software, for example.

                    I don’t think a BSD/MIT license would have caused as much good in the end, but it is a mixed bag.

                    So the comparisons are interesting, but they don’t really hold up that good. GNU is definitely its own thing and all analogies are lies.

                    1. 0

                      I’m sorry my statement wasn’t clear. I’m not saying GPL is Communism. I’m saying that GPL is another one of those ideas, like Communism, that look good on paper but fall apart in practice because humans.

                      The MIT/BSD/Apache are the original hacker spirit and I am heartened that not only do they make up the bulk of licenses but they make up the bulk of newer licenses.

                      You might say - oh, companies will just take the code and run away and make a profit of your back. That’s GPL-think.

                      In practice what happens is the company sees, god, it’s gonna be SO EXPENSIVE to maintain our own independent fork. Let’s just contribute back to the main code base. So they’ll help out, in fact form consortiums around MIT licensed code, while not touching GPL code because of it’s interesting requirements that makes the legal department twitchy.

                      1. 5

                        I’m saying that GPL is another one of those ideas, like Communism, that look good on paper but fall apart in practice because humans.

                        GPL licensed software is in daily use all over the planet, which can hardly be said about Communism. I’d like to see more examples of how it is “falling apart”.

                        I am aware of the many and deep philosophical differences between GNU and “the BSDs”, especially when it comes to licensing, and I can appreciate both sides, but it’s hardly the case that GNU is losing. Sure, its most visible example is the Linux kernel , but it has been so monumentally successful as to carve out a niche where the BSDs among others can thrive.

                        1. 1

                          If you look at license distributions:

                          https://resources.whitesourcesoftware.com/blog-whitesource/open-source-licensing-trends-2017-vs-2016

                          you can see where people have voted with their feet.

                          A good example of MIT/Apache fostering cooperation and resulting in a common good is Cloud Foundry (Pointed out to me by one of my colleagues as an example of effective cross-company collaboration)

                          Many projects under the Apache foundation are also good examples of collaboration.

                          These are all driven, under the hood, by avoiding legal tangles in licensing and letting development and innovation move ahead.

                          1. 5

                            Right…. but that doesn’t really mean that BSD-like licenses == awesome capitalism and GNU == Godless Communism.

                            Before GNU, it was common for companies to take academic work (traditionally “open”) and shut it down. Stallman reacted to that and created the “viral” license that couldn’t be shut down. In doing so, he expanded the idea of what software licensing could be, how people could work without specifically pecuniary renumeration to write and maintain it, and basically created the ecosystem that enabled the internet.

                            You might say that the pendulum swung too far, and some companies, while totally fine with contributing to open source, would avoid the GPL for precisely its viral features. But without the tireless advocacy of RMS and the GNU foundation, people would still be ridiculing the idea that high quality software could be made “for free”. The BSDs didn’t have that advocacy. They were happy enough to tool around in an academic setting. RMS wanted to take that to the world.

                            1. 1

                              That is an interesting but very speculative hypothesis. The genesis of the internet was via ARPANET. In my view the reason why the internet took off was a common protocol (TCP/IP) which is an open standard.

                              The wide use of the internet outside of academia was driven by private enterprise (the age old theme of offering services for a fee, in an effort to earn revenue).

                              It’s not clear to me that RMS advocated for open source. He advocates for GPL and has strong scriptures on what can and should be distributed via GNU/Linux (including how you can say the name). Note that these scriptures do not prevent people from running the software on closed hardware.

                              Almost every response I’ve gotten in this thread has boiled down to “Companies are evil”. This seems to be RMS’ view based on his experiences. I have no opinion on this - companies are all different and act differently under different stewardship and their own individual circumstances (a lot like people, actually). I do find it fascinating that support for GPL, in this thread, boils down to this aphorism.

                              1. 2

                                The very fact that we can have the discussion of what open-source[1] license is best for business shows that RMS and GNU… maybe didn’t win according to their strict definition of Free Software, but at least pushed very hard to make the idea of open source mainstream and accepted.

                                For that, they deserve recognition and respect.

                                [1] A term RMS despises.

                        2. 3

                          You might say - oh, companies will just take the code and run away and make a profit of your back. That’s GPL-think.

                          I think you have deeply misunderstood the GPL. It isn’t an anti-commercial license – commercial use and “making a profit off your back” is both protected and encouraged by the GPL.

                          The GPL is meant to prevent someone evil from taking my work and using it to lock users in to yet another proprietary oppression. “GPL-think” says a user has a basic human right to control of their computing, and I want no part in the infringement or violation of that right.

                          1. -1

                            The GPL is anti-commercial in effect, which is the distinction I make between an academic hoped for effect and the actual effect in the real world.

                            The legal complexities the GPL introduces act as a barrier in commerce.

                            The GPL is not a concern to service companies, but is a barrier to companies making derivative works.

                            1. 7

                              If they want to make derivative works that remove the user’s control over their computing, then yes. This happens to be a popular practise of commercial entities, but also of some noncommercial entities. Refusing to respect user’s rights isn’t limited to commercial cases – nor must a commercial case choose this path.

                              1. 1

                                Yes, the GPL is driven by an idealogy. There is an “idealogy lock-in” (in this respect it’s very similar to communism). You can take MIT code, modify it and release it under GPL. You can not go the other way. You can form an island of Communism in a free society, but you can not go the other way.

                                I find this idealogy lock-in anti-thetical to my beliefs as a hacker.

                                1. 3

                                  I recently had a user of my AGPL-licensed library ask if I would possibly relicense under a more liberal license… because the user’s clients, including a certain “Goldman Sachs,” might be scared by the AGPL.

                                  I chose not to relicense it, and so they released their derived work under the AGPL, too. The thought of lawyers at Goldman Sachs being scared by my copyleft license deeply appeals to my beliefs as a hacker!

                                  It seems to me that you’re using “ideology” as a pejorative while actually your anti-copyleft advocacy is just a different ideology. But this is such an old and tedious discussion.

                                  Especially the “communism” part is tedious…

                                  1. 1

                                    Hi!

                                    It seems to me that you’re using “ideology” as a pejorative while actually your anti-copyleft advocacy is just a different ideology. But this is such an old and tedious discussion.

                                    Oh yes indeed, but more in the “intolerance of intolerance” manner. I personally support the ideology that is free-er and more permissive. But I don’t object or oppose people using the GPL. To each their own!

                                    I just want to point out some effects it was that may not have been intended. But I am amused by how personally people take my statements. It does suggest a strong cult of personality around the GPL.

                                    Especially the “communism” part is tedious…

                                    I apologize. Perhaps I shouldn’t use that as an example. De-icers on bombers, perhaps? Any idea that seems good initially but then in practice turns out to not work as expected is what I was going for. Especially because of sociological properties.

                                    However, the example I give about freedom in licensing and how the permissive ones give more freedom to everyone I think is correct.

                                    1. 3

                                      It seems like you support both non-copyleft open source and proprietary closed source, with no sympathy for copyleft. I consider copyleft a very reasonable use of copyright—my right as a creator to maintain some control over my creation. When I release my work into the commons, I am generally disinterested—in a rather selfish sense—in enabling others to distribute proprietary derived work, so I usually choose copyleft.

                                      I don’t think ethical judgments can be fully grounded in rationality, but I clearly have at least a personal bias toward wishing for the strategic success of the free software movement, and I prefer to imagine a future where most of the code used in an ordinary user environment is open for study and modification. It seems like this could potentially be extended to a more comprehensive critique of the way business secrecy alienates customers from products, and I do actually feel like that’s a serious problem with capitalism.

                                      The idea of, let’s say, a car whose functioning is transparent to enthusiastic owners, which provides at least basic material for self-study and maintenance, is much more beautiful to me than the idea of a car that’s intentionally locked down as much as possible to prevent the user from opening and understanding it. Same for computers and software systems. So I see the GPL as a strategic/tactical instrument for promoting this latter vision.

                                      One goal being to cause businesses to think “it would be cheaper for us to work within free software and gain access to the copyleft corpus, than to develop proprietary versions of these dependencies,” thus shifting the economic incentives toward freedom of study and modification (which makes for more competent, effective citizens, etc).

                                      You might find all this distasteful and ridiculous… but it’s not a revolutionary plot to seize all private property and centrally control the world economy. It’s just a copyright license made to promote the software commons.

                                      1. 1

                                        Hi @mbrock,

                                        Thank you for that beautiful answer. I do admit that I like the world you envision - it is possibly a world that all tinkerers envision - where all the stuff is out there, open to tinker with.

                                        I’m not convinced that the GPL will help us get there. I don’t think it’ll work with businesses who deal with actual technological innovation - it reduces their freedom to innovate. It fits with businesses who deal in services, but they are not really pushing technology forward, just surfing on existing technology.

                                        Personally, I do like tinkering, but I know the exact moment I gave up on the open world you envision: as a kid I used to love repairing clocks. My uncle had many and I would repair them. Then I opened up a modern LCD clock. There was nothing I could do to help it. I did get something cool out of it though: a polarized plate. Science marches on.

                              2. 1

                                Hard for me to tell on anti-commercial part given all the dollars built on GPL-licensed apps often giving nothing back to developers. That’s part of reason License Zero made a free, shared-source, non-commercial license. That discourages commercial use by freeloaders. The GPL hasn’t.

                            2. 1

                              The MIT/BSD/Apache are the original hacker spirit

                              They really aren’t since two assume only copyright law matters. People using them can still be sued and/or controlled via patents. Apache attempted to correct that. Anything preserving the hacker ethos has to have patent protection. Interestingly, BSD code helped two companies succeed who patent sue their FOSS-using competition the most. One might say those licenses aided and abetted the opponents of hackers even more given their financial and legal model was strong enough to lobby for laws leading to more bad and less good for hacker goals.

                              It’s why I’m against those except when you want those damaging, evil companies using your code in ways that can support further damage. If not, gotta use something that tries to reduce that damage.

                      2. 4

                        It would be easy to dismiss GNU today, given the quality alternatives, and the trend towards minimalism.

                        However, they were pioneers, not only in “creating a free OS” but also in evangelizing about free software. They deserve a lot of credit for their work and they made the work a better place.

                        Nowadays, unless there’s anything I missed, there are no top-tier free software projects sponsored by GNU except maybe for the GIMP and Emacs. We have Libreoffice, Apache, the Linux distros, Firefox, etc, which all have their own foundation supporting them.

                        What I’m trying to say is that they fell “out of fashion” and are devoting their resources more towards advocacy than software development. I don’t know if that’s an official direction but that’s definitely my feeling. But that is not reason enough to throw shit on them like many internet pundits are currently doing.

                        Without GNU we wouldn’t have libre software at today’s scale. There are always the BSDs, but my take is that without the GNU/Linux competition, they wouldn’t have been a viable alternative to the Windows world.

                        Thanks and happy birthday, GNU!

                        1. 5

                          The FSF made a concious decision at some point to stop financially sponsoring software development and focus on advocacy instead. Not because they were “out of fashion” but as a strategy choice.

                          That said, they are still the fiscal sponsor (fancy words for “corporate home”) of the GNU project which contains huge amounts of software you probably depend on, including most of the default Unix commands in a GNU/Linux distro (tail, cat, ls, bash, diff, wdiff, etc), gcc, which on a GNU/Linux distro probably compiled all your base libraries, kernel, etc, emacs, Guix, GNOME (including GTK+, GIMP, etc), and many many more

                          1. 3

                            They still handle donations for a number of packages like Octave:

                            https://my.fsf.org/civicrm/contribute/transact?reset=1&id=10

                        2. 3

                          Happy birthday GNU, to 35 more years of libre software!

                          1. 2

                            Happy birthday GNU, even if you aren’t getting a greeting from Stephen Fry this year.