I also think that a lot of people have no clue about what is free software/open source at its root.
Nothing was said to make money, be rich or avoid someone else (being big companies) becoming rich.
At the origin, the greatness of the concept was that you were free on your own computer to do whatever you wanted, how you wanted it. For yourself. And more particularly, FOSS was allowing anyone to have access to tools and server software that was restricted to companies with very expensive and complicated licenses before that.
See Linux that gave access to an Unix-like OS at a time were it was unreachable to hobbyist.
In my own time, I remember this incredible pleasure of being able to host his own web server with WAMP or LAMP stack, at a time where you were forced to buy and use windows Pro with corporate license IIS to host even a simple website...
Nowadays, the greatness of FOSS is that anyone can still access for free, without string attached, to all the software needed to train and run LLM models.
> See Linux that gave access to an Unix-like OS at a time were it was unreachable to hobbyist.
This is revisionist history. Linux eventually supplanted FreeBSD, which was widely used and popular in the 1990s. FreeBSD was also objectively the better OS back then, a clean and reliable system.
What made Linux successful over the long-term IMO is that they evolved the kernel much faster and earlier than FreeBSD, particularly when it came to hardware scalability and performance, which was advantageous with the progression of Moore’s Law. The cluster-style supercomputers were developed almost entirely on Linux as a result, which created a positive feedback cycle. Second, the Linux user-land quickly became really pleasant to use compared to most other UNIX flavors at the time because they weren’t too concerned with purity or being strictly interoperable with some other UNIX.
Most people in the industry today don’t remember the era of at least a dozen different UNIX-style OS being widely used. I probably used half of them in anger at some point. In that environment Linux won by having a superior UX out of the box for a lot of use cases, even if you paid the absurd money for one of the proprietary UNIX flavors. Later Linux started demonstrating better performance too. FreeBSD was definitely tidier and cleaner with its much slower evolution but sometimes worse is better.
While Linux was technically released a year or two earlier, the contemporaneous versions of FreeBSD were much more mature. Linux was pretty unusable as a production server for the first several years but FreeBSD was more than serviceable.
I used them both almost from the beginning of their existence. Linux was a novelty early on but FreeBSD was serving a lot of the load in Silicon Valley for a big chunk of the 1990s. Linux didn’t really hit its stride until the 2000s. Linux was pretty actively bad early on if you just needed a UNIX server.
We will have to disagree, by 1993 we (University) used it all over the place for production with either Slackware or Debian. Which where rock solid. FreeBSD barely worked around the time, and even with the FreeBSD 2.0 release which nuked a bunch of things due to the whole AT&T mess, people eschewed it like the plauge.
That GNU/Linux hit its stride "in the 2000s" is ... what is revisionism of history here to the point of it being an absurd thing to say.
> Nothing was said to make money, be rich or avoid someone else (being big companies) becoming rich.
Sometimes many people don't realise the value of information is not there in individual blogpost but when you can combine them into a holistic approach. A drop of water by itself is useless, only a body of water becomes useful.
> With Open Source/Free Software as well as Creative Commons we have build pipelines to contribute to the commons (great!) but we never thought about how to defend those commons against appropriation.
Perhaps the CC people have not thought about this, but it is precisely why GPL exists, and later AGPL.
If courts find that derivative work comes from AI systems, the AGPL will become a very interesting legal landscape for even hosted models.
> Perhaps the CC people have not thought about this, but it is precisely why GPL exists, and later AGPL.
Considering the next sentence is "This is why Open Source ends up massively benefitting corporations who don't give much back.", I think this is more along the lines of the usual complaints about for example cloud providers not hiring the upstream devs for oss projects they sell "as-a-service" versions of.
Surely GPL'd code was swept up with ai training like the rest of the internet. Perhaps some of that code incidentally has licensing that works with this, but surely not all of it....
CC SA(ShareAlike) clause is basically the equivalent of GPL: any derivations must be licensed the same as the original (either CC BY-SA or CC BY-NC-SA, latter only permitting noncommercial use).
People should just modify the GPL and add a clause that explicitly forbids training AI and similar things. Instead of hoping a court will follow the logic, just settle it for good.
I feel like this keeps coming up again and again and I can't help thinking that they just don't get it. Or maybe I don't.
There's two prevailing sentiments IMO - (1) FSF-like: the greatest freedom is achieved when we empower the users of the software to improve or fix the very software that they are using and (2) BSD-like: the greatest freedom is achieved when we permit licensees to freely use the software how they see fit.
In my opinion, nothing in either scenario risks "appropriation". There's only different takes on "freedom." But if you fear appropriation then clearly licenses like AGPL and GPLv3 are more suitable.
Making money is not incompatible with Open Source nor Free Software. Making lots and lots of money is not incompatible with it either.
If you don't like what might happen when you license your code liberally, you might just prefer proprietary software. And that's okay too. You can even publish the source but add a restrictive license (or explicitly forbid any use). This is a source available proprietary approach. I wouldn't contribute to a project like this, but there's no reason you couldn't make yours like this.
Free (beer/kostenloss, not freiheit) is magic and does magical things to people's brains. Also, public software work is really valuable to attract higher quality results.
If you seek to balance a ledger, then it's not free (in either sense of the word).
I wonder to myself: these articles, are they really written by open source software contributors who regret their contributions? Or someone on the outside looking in, wondering how they can "fix" things?
My view on free software has changed over the course of my life. When I was in my 20s, my big concern was being able to do things like afford the mortgage on my house and be able to maybe be able to afford to get married and have kids or at the least pay off my credit card debt. Time spent on free software was unlikely to help with that and I focused my development time on things that would bring in money.
Now, in my 50s, I’m looking at retirement somewhere between 4 and 14 years away (although given the ongoing demolition of the United States, it might be more like 24 years away). I have little interest in building a side-business in software and, frankly, the flow of money in software is such that it wouldn’t be likely to be a viable concern anyway (to give just one example, in the 90s, I spent hundreds of dollars on compilers and database software that today are expected to be given away for free). As a consequence, the code that I write on the side I release with the least restrictive license possible because, frankly, I don’t really care if some corporation takes it and incorporates it into their product (in fact, I’d be kind of honored), and while it would be nice if they would kick me a few dollars, given my experience working as a corporate developer, my guess is that the few dollars that have come my way aren‘t coming out of corporate coffers but out of other developers’ wallets (for which I’m truly grateful, even if the actual dollar amounts likely don’t even cover my marginal expenses on the work).
There's a reason why FOSS is by hackers for hackers—it's software written by people to solve their own needs. They release that software as Free/Libre so that others can benefit from it if they would like to, but first and foremost it's written because the author themselves saw a problem and wanted a solution. What's changed is that as startup culture infected hacker culture a few subcultures developed within FOSS that thought that that wasn't good enough.
One of those subcultures decided that it was an imperative that they make more money off of their FOSS projects than anyone else does—this is how you get the Matt Mulenwegs of the FOSS world, who diss on other FOSS developers because "most of the value" of their projects was "captured by others" [0]. They see FOSS as a revenue stream and if it's not a revenue stream (or if it ends up being a bigger revenue stream for someone else than it is for you) it's a failure.
Another subculture thinks that FOSS should be mainstream. That's the idea typified here: by "hackers for hackers" is selfish and we should instead "reshape our thinking towards more political goals and values".
The fact is that both of these perspectives are fundamentally not about adapting FOSS to the times—they're both about building two entirely different philosophies. FOSS is not a business model, and what the Matt Mulenwegs of the world actually want is to be a tech startup CEO. And FOSS is not and cannot be mainstream because to become mainstream requires a bucketload of UX design and user support work that no one in their right mind wants to take on for free as a hobbyist.
FOSS is by hackers for hackers because it must be. If you want to make more profit than anyone else does off your project or if you want to pursue "political goals and values" you're looking for an entirely different type of organization.
The answer is not that complicated but people seem to make it so out of desire to either cover all the corner cases and/or tackle a more tangential issue related to free software (such as compensation). While valid areas of exploration, it can make one miss the forest for the trees.
I feel the question can be answered as simply as once upon a time a question like "how many angels can dance on the head of a pin?"
Free Software is "for hackers" because, quite frankly, most other people just don't care about the goals of the movement. No amount of "getting out of our comfort zones and modes of operation" is going to change that. And the political angle generally makes it less palatable to general public, not more.
"Meeting people where they are" would require a drastic change of goals, to the point where the result would be unrecognizable, and wouldn't prioritize the things that hackers care about. Which isn't wrong per se, but then why expect hackers to drive such a movement?
> Free Software is "for hackers" because, quite frankly, most other people just don't care about the goals of the movement.
Until their car dashboard starts showing them ads. Or they can't even screenshot streamed movies to comment on for their friends. Or the app they were using to organize protests gets removed from 'their' phones [1]. Or they can't fix their tractor [2]. Or their printer refuses to print with 3rd party ink, or includes tracking dots to betray its user. Or their videoconferencing program gags them at government request [3].
I think the author is confused because he doesn't realize that hackers and muggles are fundamentally different.
Hackers come into foss because they get fed up about how shitty proprietaryware is. Shitty because it gets in the way of their hacking. Muggles don't care about hacking, so to them it's not shitty.
The author thinks that 10x people = 10x better so we have to bring the muggles on board. But actually it doesn't matter, non-hacker users don't contribute much to foss. They don't send pull requests, and don't submit useful bugs. Most they'll ever do is nag you into making the thing less hackable.
What non-hacker users contribute, in theory, is widespread acceptance. This comes with its own benefits, such as F/OSS solutions becoming de facto standards, which in turn means things like e.g. Linux drivers for more software, content providers not trying to lock it out etc.
However, this all can only happen if there are a lot of non-hacker users. Unless and until you actually get to that point, there are no benefits, but the support costs are there regardless.
And I have serious doubts about getting to that point, ever. I remember being much more... optimistic - or perhaps naive - back in early 00s. Many of us really thought back then that F/OSS would surely win, gradually, slowly, but surely. Now? Things are better in some ways, but overall I'd say that the environment is more hostile towards openness than it ever had been before, with widespread adoption DRM and lock-ins throughout the entire software ecosystem, massive centralization of online services, and pervasive surveillance being pretty much the accepted social norm. On top of all this, smartphones became the dominant personal computing platform, and DRM and restrictions are even worse there.
So, I don't know. Maybe you're right and this kind of "secret knowledge" for the initiated, those willing enough to dedicate significant time and effort to make it work, is the best we can have. It sure is a depressing thought, though. And I'm not so sure we can keep even this much, to be honest. E.g. at this point I fully expect to see bans on things like backdoor-less end-to-end encryption in my lifetime. Or, privacy aside, how about protocols that are straight up non-implementable because of DRM? Free software has limited usefulness if it can't meaningfully interoperate with things outside of the bubble.
The beauty of free software is having your cake and eating it.
It seems like today catering to hackers and normies with slightly different builds or packages would be easier than ever.
The author really doesnt craft a path forward, but nothing he implies would indicate that hackers lose anything (other than like you say, lack of nagging)
> Free Software is "for hackers" because, quite frankly, most other people just don't care about the goals of the movement
Sure. But, like, the world wide web. There is a playbook for making this work. Where Silicon Valley's mythmaking bites hackers in the ass is in pretending to write out the government's role out of the history of the internet.
Free software is a public good. Public goods have basically one good funding model.
Free Software isn't for hackers. It's for users. Hackers are also users, but they're not special.
The purpose of Free Software isn't to teach people how to program. It's not a scholarship program for gifted kids to generate startups. It's to institutionalize the right to access and change that software that defines the behavior of the machines that control our lives. You may say that only a programmer can take advantage of that right, so that means that Free Software is meant to create hackers. But I can hire a hacker just like I hire a plumber, or like my condo association hires a roofer, or my town hires a construction firm.
There should not be machines in your life that run on secret commands that you are not allowed to know about. They are not working for you.
edit: it is best not to talk about Free Software Licenses and Open Source/Creative Commons licenses at the same time if you're not talking as a consumer. They don't have much to do with other, other than that their development models are similar and that Open Source software will always be available over the same channels as Free Software.
Copyleft licenses like the GPL are very restrictive licenses that try to make sure that access to the code that they cover will never be restricted to a user of that code. They are an attempt to grant the rights that people should have to all code to enough code that one can accomplish one's goals without having to touch a locked-down black box, and to forbid the makers of locked-down black boxes from taking advantage of, embracing, extending, or extinguishing that code. The copyright holders of copyleft code are granting you that right of access and modification instead of the government. Open Source/Creative Commons has zero interest in that.
Open Source/Creative Commons are liberal licenses that allow anyone to do what they want, and make no (or very few and trivial) demands. You can take it and lock it behind any license. They are only compatible with copyleft licenses because copyleft licenses are part of the class "any license," not for any other reason.
>Free Software isn't for hackers. It's for users. Hackers are also users, but they're not special.
This is wrong. Hackers like to hack, which requires having control over their tech. So you get foss which gives you lots of control. You get things like GPL which tries to protect that control.
The control and freedom to hack is not relevant to regular users. Regular users don't want to modify and hack their software, they're happy to consume it passively. To them, the control is not a factor and foss is just another product. The pro is that they can maybe save a few bucks, unless the proprietary alternative is also free (they don't care about privacy or ads). The con is that it's less polished, the website is ugly, the docs are confusing and don't have cutesy cartoons saying to press the shiny button to make thing go brrr.
Lumping both of these in the same category just confuses things. Hackers and non hackers want very different and fundamentally incompatible things.
While I mostly agree with this, it's inaccurate to say that Open Source Software is more liberal than Free Software. The issue of copyleft vs. permissive licenses is orthogonal to that classification. The difference between Free Software and Open Source is mostly that of intent: Free Software is primarily political, with Freedom being the ultimate goal, while Open Source is primarily practical, with high quality software being the ultimate goal. In practice, almost all software that meets the definition of one meets the definition of the other.
Rather than "political" vs "practical", I prefer to say, "social" vs "economic". Free Software is about helping your neighbor. Open Source is about cheaply producing a good.
> Free Software isn't for hackers. It's for users. Hackers are also users, but they're not special.
This is exactly the dated attitude that I think the author is talking about.
Free software is for programmers. Once upon a time that was a distinction without a difference. Today, the fraction of users who can take advantage of source code rounds to zero.
Yet, FOSS advocates keep beating this drum to “users” while entirely ignoring the reality that today’s users are interested an entirely different value proposition than they’re offering.
Computers aren’t professional tools anymore. They are commodities. Computer users want source code like grocery store shoppers want a tractor and a plot of land.
It's not about source code access, it's about guaranteeing the four freedoms. As a user, with proprietary software you have to worry about whether you can share a copy with a friend, sell it on a CD to a guy on the street for $5, or use it at your work place. With free software you can do all of that without worry. For better or for worse, I am not a programmer, and free software is important to me. I have often wished I were a programmer in order to better help and contribute, but have long struggled to learn. I doubt I'm the only one in such a situation.
Yes, but the difference between freeware and FOSS is rights to the source code. Most users today would only want to exercise freedom 0, so FOSS really isn't offering them anything that appealing. And in fact, freeware is more popular than ever today.
>Most users today would only want to exercise freedom 0, so FOSS really isn't offering them anything that appealing.
I was aiming to address this in my previous post, as a non-programmer user, listing realistic examples that the other freedoms give you. Being able to use software at both home and work, or to share software with a friend without worrying about legal trouble is a benefit I think anyone can appreciate. Lots of software will charge [extra] money for commercial use, limit installs to x number of machines, etc. If you get used to using free software, you can avoid these issues.
Also, while not related to the license exactly, using free software, you'll tend to not run into ads like you do with a lot of freeware. I recommend apps from F-Droid to people with Android phones for this reason a lot. While grabbing a popular file manager or music player from the Play Store often seems to get you an ad-ridden mess, I never run into that with anything from F-Droid.
> Computer users want source code like grocery store shoppers want a tractor and a plot of land.
You entirely ignore ecosystem effects. Users don't want source code, but they do want programs that won't betray them. Open access to source code is how such programs come about. See my other post on what happens when that access is taken away: https://news.ycombinator.com/item?id=43385000
Non-experts are neither aware of this nor have no way of independently verifying that themselves. If you observe today's software marketplace, people are not switching to FOSS in response to privacy concerns. They're not even aware of what FOSS licensing is.
And regardless, the issues you list in your linked comment aren't even anything FOSS has a solution for -- all of them are only partially software products. You can't replace something that is more than software, with software. At some point an IP licensing agreement just simply isn't enough can't plow a field. You need a factory, some steel, and some diesel fuel. Farmers don't plow their field with software, they do it with a tractor.
Most people don't really care. In your link you mention cars locking people out, well, they still use them don't they? And they even pay huge sums of money for the privilege.
At the same time, FOSS software is also historically filled with corporations and slimy individuals rugpulling their users (many examples over the years, Emby jumps to mind, so does the Simple Mobile Tools set of apps). It's so common on mobile in particular that a major benefit of F-Droid[0] is that it just chooses to sit between you and the developer to make sure that the developer doesn't get up to any shenanigans.
Open source/Free software isn't a panacea against that and if anything has led to the formation of a development moat, where AWS et al. can profit from FOSS, but it can do so by muscling out all profits from the smaller developers[1]. FOSS by design can't offer any legal recourse against that, which is a pretty major mark against it for most people.
[1]: A less aggressive, but no less insidious version of this is the "this piece of software has a ton of extra services it relies on tech stack", where to replace one single service, you're essentially setting yourself up to rehost an entire corporate network and said hosting usually requires resources not easily accessible to individuals (ie. Multiple servers). This usually happens with organizations that want to maintain some level of control over their users; Firefox Sync for example isn't feasible to truly selfhost, since the Mozilla accounts service also needs to be selfhosted and it's a massive PITA to do that.
Not sure why you are bring downvoted. Your argument is clear, and drives the discussion forward.
I would agree that Free Software can (and is) used by everyone but only a very tiny fraction are able to exercise any of the Freedoms it advances. But that's OK.
My mom doesn't know the difference between Windows and Linux, and the license distinctions are meaningless to her. But she could still get meaningful value choosing one over another.
FOSS beats the drum, not to turn people into programmers, but rather so that some tiny fraction of the next generation understand the building blocks thst got us here, and carry the game forward.
Personally I think there is a lot of ignorance and misunderstanding of both OSS and Free. Lots of people want something outside the scope of these, and rather than create some new category, try and morph one of these.
Complaints like "give nothing back" or "AWS is eating our lunch" etc are all irrelevant to OSS and Free. They are topics which are not interesting to those licenses.
I would say this; if an OSS or Free license does not suit your need, use another license. Or write your own. There's no problem with that. (For 25 years I've been shipping source, but without distribution rights- "source supplied", not "open source".) I freely use OSS and obviously the OSS parts keep their OSS license.
99% of my customers won't ever look at the source. That's fine. I found a license that works for me, and them.
I think end users can certainly benefit from the downstream effects that FOSS has on developers, integrators, and other technology professionals. But most FOSS advocacy ignores this in their messaging, their strategy, and in the spaces where they operate.
I've only ever seen FOSS outreach to industry groups, academic groups, hobbyists, and other professional or professional-adjacent groups.
They very well could make their case to the modern end user that their approach is better than closed source SaaS solutions, or other freeware. But they don't. It's a missed opportunity, and they're going to lose relevance and influence if they don't appeal to those who drive market demand for software products.
>> They very well could make their case to the modern end user that their approach is better than closed source SaaS solutions, or other freeware.
Lots of people benefit from Free Software, yes. But asking the consumer to understand licensing and the implications thereof would be a fantastically ambitious and expensive project.
Ultimately yes, FOSS outreach is targeted, because the first-order benefits are targeted. Outside the internet itself (where the message is there for anyone to find) only very narrow groups would get any value from a discussion on licenses.
For example, say I make software for small-town bakers. My software might be open source, or closed source. My bakers couldn't care less. They care only about the first order benefits - does the planner work etc.
Ironically "no-one" chooses Linux because of the license. They choose it because of the price. And consequently they dominate where "number of machines" is the key issue (aka servers).
Consumers choose Windows or Android because of the first order benefits there. They are better than Linux Desktop or Linux on the phone.
(And yes, I'm aware Android uses the Linux kernel, but Linux is not sufficient.)
The whole "year of the Linux desktop" argument is for Joe public to care about License. 25 years in, and I think it's safe to assume that approach has failed.
>> to industry groups, academic groups, hobbyists, and other professional or professional-adjacent groups.
FOSS exists today precisely because these are yhe groups that matter. And they are, I think, the groups that will continue to matter.
> For example, say I make software for small-town bakers. My software might be open source, or closed source. My bakers couldn't care less. They care only about the first order benefits - does the planner work etc.
And in today's business landscape, the model that the baker and the software developer both like the best is permissive-FOSS licensing with proprietary extensions and SaaS delivery. Or, an entirely proprietary application that runs on top of FOSS server components. Which is not looking good for the future of FOSS in terms of contributions back to the community. In this way, it looks like FOSS is becoming less for users and more for venture capitalists and startup founders :)
I feel like perhaps you are seeing a change in this?
Obviously the baker doesn't care about the license- but certainly a SAAS approach has gained momentum. Whether this is because the baker wants it that way, or because its easier, and more profitable for providers that way, is debatable.
Does FOSS offer less for end users than before? I'm not sure. The benefits of getting source code (reading, editing, compiling) has always demanded a skillset far removed from most users. Doing SAAS as FOSS just adds "hosting" to that list of skills.
Yes, lots of startups advertise "Open Source" but that's just a marketing thing. They pivot the license when someone else starts getting commercial value from it. Folks contributing to VC projects know what they're getting.
But there are lots of modern Open Source successes as well. Chromium has spawned a whole slew of browsers, Linux continues to grow (code wise), and most programming languages are now open.
Writing Open Source has always been difficult from a commercial point of view, simply because its cheaper to commercialize work that didn't cost development $. But again, that's always been true. Red Hat didn't write Linux.
Is the future for new Open Source projects rosy? I don't know. But folk have been predicting its demise since, well, it started.
The most important thing these articles don't understand is that it doesn't matter that they don't understand.
"How can this possibly work?" It doesn't matter, it already is. Almost every example of a failure or problem they point to is usually no failure or problem at all.
When someone doesn't understand or can't imagine why anyone should work on gpl software, all I say to that is "OK".
Ok. You don't think it makes any sense. Ok. My participation does not hinge on you understanding, and the quality and utility of the software and the garden that it grows in does not hinge on you understanding. Go forth and have nothing to do with this nonsense, with my honest blessing. I can do a thing without needing to try to convince you to do exactly as I do. We can be different.
Your participation hinges on you understanding, but if you aren't still a child and don't understand after 15 minutes with google, then it's because you don't want to understand and the whole deal is just not for you and you were never going to participate anyway, and that's ok. There is no problem to resolve there.
Open source, in the traditional simple gpl form without any further "fixes", already works exactly the way it was always intended to, regardless if you understand how it can possibly work. It's working perfectly well right now. All the big companies making billions while using gpl software is perfectly fine and they didn't steal anything unless they actually violated the gpl. (Which does happen but that's a seperate issue. People steal commercial software by violating it's license too.)
All the fixes actually just break it. As another commenter said about trying to add special modifications for commercial use etc: "You can do that. I won't contribute to a project with those terms but you can do that." And there's the breakage right there.
Free Software is for the people who build it. Great closed source companies become great through their ability to sell to many customers. Great free software projects become great through their ability to coordinate and integrate the contributions of many people. The larger the percentage of "people who have contributed something" to "people who use it", the healthier the project is (think of this as [C/(C+U)] where C=Contributors and U=Non-Contributing Users; every Contributor is a user).
The reality that you can use many free software projects without much contribution at all is a happy and good accident of the system; but a dangerous one that can and has been often-accidentally weaponized to burn out the actual contributors.
>> The reality that you can use many free software projects without much contribution at all is a happy and good accident of the system; but a dangerous one that can and has been often-accidentally weaponized to burn out the actual contributors.
I'm not downvoting you, but I expect people are disagreeing with this last paragraph. (There's a metaphor between people voting up and down based on what they agree or disagree with - something this site doesn't advocate for, and the way some interpret Free licenses, and their goals.)
I agree you can use any Free Software without any contribution at all. And that only the tiniest fraction contribute any code. In most projects the contributers past the owner are zero.
This is not "by accident" though. It's very much by design. We can assume RMS thought a bit about his license and he most certainly expected most of his users not to need to read the code, much less add to the program.
The Freedoms are thus Freedoms, not Requirements.
Burnout is an issue completely unrelated to the license. It happens when an employee doesn't find a balance in their life and work. It happens with free, and non-free software alike.
Yes, lots of programmers burn out because they cave to the demands of unreasonable bosses, or users. But that's unrelated to license and I'd certainly not a "weapon".
Yup, this article hits it straight on the head. Both Free Software and Open Source (the later is just a more corporate-friendly version of the former) are fundamentally 80s ideologies, concerned about software you run on your computer and the whole thing was created because Stallman couldn't fix his Xerox printer. 40 years later, many things have happened - the Internet, widespread usage of embedded devices and SaaS - and the movements just.... didn't adapt to them at all.
The whole culture is so enmired in the original definitions that it simply can't see anything outside it - so everyone just drains from the creative commons but doesn't really expand it. There's a reason why most prominent FOSS applications are old pieces of software. It's just not a model which works in the 2020s, except for software libraries.
Ironically, many Free Software projects are almost indistinguishable from "look but no touch" software - while in theory you can make changes, in practice you can't because contributing to it is next to impossible and the project structure basically spells the definition of NIMBYism out.
There's also a fundamental tension here - anyone who argues that the current tradeoffs of open source/free software just aren't good enough almost instantly gets dismissed with "well well well.... if you don't like it, don't use it/pick a proprietary licence lol". Or they go straight to victim blaming with "you should have known what you got yourself into, no takebacksies~"... There's a strange dynamic here where if you want to earn money from your project, then FOSS is not a business model, but if a megacorp wants to earn money from your project, then it's okay because everyone wins and you don't lose anyway. Or something like that.
The Free Software movement rejected all ideas which could have contributed to a solution to the ecosystem thriving - non-commercial licences, usage-based licences, dual-licences (in the sense of different terms applying to different users) and countless other things in favour of strict textual interpretation of founding principles. Religions didn't survive thousands of years only on textualism so Free Software won't survive this way, either.
>Ironically, many Free Software projects are almost indistinguishable from "look but no touch" software - while in theory you can make changes, in practice you can't because contributing to it is next to impossible and the project structure basically spells the definition of NIMBYism out.
These projects still allow you to make those changes locally - and that's the entire point. FOSS has never guaranteed that your changes would be incorporated by anyone else, distributed by anyone else, or recognized as legitimate or valuable by anyone else.
We're just vastly more conscious of that in the 2020s, in our era of internet celebrities.
>There's a strange dynamic here where if you want to earn money from your project, then FOSS is not a business model, but if a megacorp wants to earn money from your project, then it's okay because everyone wins and you don't lose anyway. Or something like that.
You're perfectly permitted to charge for the software, just as the megacorp is. The fact that people are willing to pay the megacorp for it and not you is unrelated to FOSS philosophy.
>The Free Software movement rejected all ideas which could have contributed to a solution to the ecosystem thriving - non-commercial licences, usage-based licences, dual-licences (in the sense of different terms applying to different users) and countless other things in favour of strict textual interpretation of founding principles
There is actual philosophy behind these principles, not just literal dogma. For example, applying different terms to different users requires them to agree to a different license because that's kinda just how law works, and you can't withhold the option of the OSI-approved licenses from anyone because non-discrimination is part of the philosophy.
The logical step here is what's alluded to at the end of the article: self hosting and doing things with FLOSS is impossible unless you're sufficiently one-of-us to get into the guts of software. We can only make this situation change when we recognise that including a wider audience is necessary, and for that we need to run things and support them for other people. Cooperatives are the way that strikes me as being the right one: giving everyone a share and a vote instead of the code itself. Nobody is going to vote to have the contents of their emails mined or their attention bought and sold.
The legal infrastructure we would need for this is a Cooperative General License. The "four freedoms" in the GPL are born from academia and they show it. The "open" licenses are libertarianism and opportunistic IP laundering (from one failed startup to another) for the most part. A CGL could incorporate a restriction on use: personal or cooperative, but not corporate, not unless the form is the right kind.
This is an old debate in tech but as the author says, one worth revisiting. We shouldn't be working for bastards, and certainly not giving them the gift of our incredibly valuable work.
There's always going to be some tension between people who push free and open source software for ethical or political reasons, and those who push free software for technical and practical reasons. This article seems to give a very unfavorable view of the latter, suggesting that the primary purpose of the "open source" movement was to make free software more palatable to corporations. Well, I believe that it was largely for that purpose, but when you put it that way, it misses the point, I think. The average enthusiast of the "open" movement sees openness as a major advantage in and of itself, for a variety of practical reasons, and the potential benefits in terms of user freedoms are largely just a consequence of that, whereas free software enthusiasts generally seem to be mostly concerned about the movement as a solution to the problem of software that doesn't respect the user's freedoms.
I don't have any issues with the concept of free software, or having strong ethical convictions, or anything like that; however, I am not convinced by the arguments the FSF and Richard Stallman have made on a philosophical level, and frankly, as a programmer, I largely don't even really care to spend much more time worrying about the philosophy of it all. I am glad somebody does, but it's not for me, at least insofar as much as I can avoid having to do it.
And that's okay, because I don't really have to. The fact is that many of us have always preferred open source software primarily due to the side effects that having software with permissively licensed source code available has: the freedom to fork, the ability to audit code, fix bugs, re-use code for new purposes the author did not foresee, and of course, the fact that it is also typically free-as-in-beer. The funny thing about this ad-hoc list of benefits is that it is exactly the same for corporations as it is for users; the reason why so many startups release open source software is because it is a good way to gain trust and goodwill early on. Companies and their employees are more willing to adopt open source solutions because of these things, even if they don't necessarily intend to exercise all of the options it gives them.
The reason why open source is somewhat lopsided, e.g. it's obviously better for corporations than users in many cases, is because the arrangement in which there is no discrimination on who can use something for what yields the maximum utility for everyone. The moment you add even a single restriction, such as "People named Rick are not allowed to use this software", it opens up a world of complexity. Open source licenses can get a little complicated, but the broad strokes are so much simpler and easier to reason about. Think of the complexity that unfree packages place on Linux distributions.
And honestly, the discussion of the "for hackers, by hackers" mentality doesn't really seem fully baked. What does it even mean to be a "hacker" anymore? What identity are we asking people to assume?
I have another theory for why the Linux desktop and Firefox are not winning: it's because they kinda suck. Don't shoot just yet; I use Firefox and Linux all day, every day, and Firefox is in relatively good shape nowadays despite Mozilla seeming to always piss away any opportunity and goodwill they have. But, for-profit corporations have more money, more people and move faster. Competing with Google and Microsoft is hard. Open source can never beat Microsoft and Google at their own game.
However, what open source can do is beat them at games they're not playing. While corporations move fast, open source endures. In the longer term, I believe many prominent open source projects will rise to the top, as we've seen with a select few. Look, I don't expect a "year of the Linux desktop", but if it does happen, it won't be what most people were expecting in the 2000s. Microsoft's enshittification will most likely just lead to a situation where Linux wins by doing nothing, so to speak... I think we're going to see a lot more of that as the software industry slows down.
But honestly, the fact that these behemoth projects exist is a testament to it's success. I don't expect an offshoot to be able to replicate its success, less surpass it. Before people jump on board, I think there really needs to be some more convincing arguments for why we should abandon a movement this successful for an unproven alternative. It's not unfathomable... but personally, I remain unconvinced.
Nah. OSS is low quality because the feedback loops do not include its users unless they are also developers.
There are exceptions, but the ideology of OSS resists attempts to make good products. Or, rather, people who are associated with the OSS ideology fail to build products.
AI tools are removing coding barriers so everyone benefits except those who depend on restrictions, but even they benefit in the long run.
A couple minor issues: the AGPL and similar licenses exist and have been effective at stopping classes of appropriation, and the cake quote is the common misconception. It's referring to the nutritious leftovers usually discarded, and free software means the corporate code deemed nonessential enough to release under free licenses gets spread around, so let's eat that delicious free software cake.
It claims FOSS isn't winning because it's stuck in the hackers for hackers mindset.
But the two big examples it gives for FOSS not winning are Firefox vs. Chrome and the Linux desktop.
But Firefox and Chrome are more or less the same product. The reason Chrome won is because it is faster. The difference is smaller now, but for a long time it was significantly faster, and that trained a lot of people to automatically install Chrome.
The Linux desktop is a different discussion. But even here, it's mostly a network effect. If you put a layman in front of a fresh Linux Mint install, and in front of a fresh Windows install, I don't think they'd have much of a problem with Mint.
But most laymen aren't laymen, they're Windows power users, to a certain extent, because that's what they've been using their whole life.
So I don't see how having a less hacker-for-hacker mindset could help either of these fights.
I also think that a lot of people have no clue about what is free software/open source at its root.
Nothing was said to make money, be rich or avoid someone else (being big companies) becoming rich.
At the origin, the greatness of the concept was that you were free on your own computer to do whatever you wanted, how you wanted it. For yourself. And more particularly, FOSS was allowing anyone to have access to tools and server software that was restricted to companies with very expensive and complicated licenses before that.
See Linux that gave access to an Unix-like OS at a time were it was unreachable to hobbyist. In my own time, I remember this incredible pleasure of being able to host his own web server with WAMP or LAMP stack, at a time where you were forced to buy and use windows Pro with corporate license IIS to host even a simple website...
Nowadays, the greatness of FOSS is that anyone can still access for free, without string attached, to all the software needed to train and run LLM models.
> See Linux that gave access to an Unix-like OS at a time were it was unreachable to hobbyist.
This is revisionist history. Linux eventually supplanted FreeBSD, which was widely used and popular in the 1990s. FreeBSD was also objectively the better OS back then, a clean and reliable system.
What made Linux successful over the long-term IMO is that they evolved the kernel much faster and earlier than FreeBSD, particularly when it came to hardware scalability and performance, which was advantageous with the progression of Moore’s Law. The cluster-style supercomputers were developed almost entirely on Linux as a result, which created a positive feedback cycle. Second, the Linux user-land quickly became really pleasant to use compared to most other UNIX flavors at the time because they weren’t too concerned with purity or being strictly interoperable with some other UNIX.
Most people in the industry today don’t remember the era of at least a dozen different UNIX-style OS being widely used. I probably used half of them in anger at some point. In that environment Linux won by having a superior UX out of the box for a lot of use cases, even if you paid the absurd money for one of the proprietary UNIX flavors. Later Linux started demonstrating better performance too. FreeBSD was definitely tidier and cleaner with its much slower evolution but sometimes worse is better.
> This is revisionist history. […]
FreeBSD came quite a bit later than GNU and Linux. The BSDs were also plagued by the whole AT&T mess.
As someone who was around at the time… I cannot really relate to your comment.
While Linux was technically released a year or two earlier, the contemporaneous versions of FreeBSD were much more mature. Linux was pretty unusable as a production server for the first several years but FreeBSD was more than serviceable.
I used them both almost from the beginning of their existence. Linux was a novelty early on but FreeBSD was serving a lot of the load in Silicon Valley for a big chunk of the 1990s. Linux didn’t really hit its stride until the 2000s. Linux was pretty actively bad early on if you just needed a UNIX server.
We will have to disagree, by 1993 we (University) used it all over the place for production with either Slackware or Debian. Which where rock solid. FreeBSD barely worked around the time, and even with the FreeBSD 2.0 release which nuked a bunch of things due to the whole AT&T mess, people eschewed it like the plauge.
That GNU/Linux hit its stride "in the 2000s" is ... what is revisionism of history here to the point of it being an absurd thing to say.
> Nothing was said to make money, be rich or avoid someone else (being big companies) becoming rich.
Sometimes many people don't realise the value of information is not there in individual blogpost but when you can combine them into a holistic approach. A drop of water by itself is useless, only a body of water becomes useful.
> With Open Source/Free Software as well as Creative Commons we have build pipelines to contribute to the commons (great!) but we never thought about how to defend those commons against appropriation.
Perhaps the CC people have not thought about this, but it is precisely why GPL exists, and later AGPL.
If courts find that derivative work comes from AI systems, the AGPL will become a very interesting legal landscape for even hosted models.
> Perhaps the CC people have not thought about this, but it is precisely why GPL exists, and later AGPL.
Considering the next sentence is "This is why Open Source ends up massively benefitting corporations who don't give much back.", I think this is more along the lines of the usual complaints about for example cloud providers not hiring the upstream devs for oss projects they sell "as-a-service" versions of.
Surely GPL'd code was swept up with ai training like the rest of the internet. Perhaps some of that code incidentally has licensing that works with this, but surely not all of it....
CC SA(ShareAlike) clause is basically the equivalent of GPL: any derivations must be licensed the same as the original (either CC BY-SA or CC BY-NC-SA, latter only permitting noncommercial use).
People should just modify the GPL and add a clause that explicitly forbids training AI and similar things. Instead of hoping a court will follow the logic, just settle it for good.
I feel like this keeps coming up again and again and I can't help thinking that they just don't get it. Or maybe I don't.
There's two prevailing sentiments IMO - (1) FSF-like: the greatest freedom is achieved when we empower the users of the software to improve or fix the very software that they are using and (2) BSD-like: the greatest freedom is achieved when we permit licensees to freely use the software how they see fit.
In my opinion, nothing in either scenario risks "appropriation". There's only different takes on "freedom." But if you fear appropriation then clearly licenses like AGPL and GPLv3 are more suitable.
Making money is not incompatible with Open Source nor Free Software. Making lots and lots of money is not incompatible with it either.
If you don't like what might happen when you license your code liberally, you might just prefer proprietary software. And that's okay too. You can even publish the source but add a restrictive license (or explicitly forbid any use). This is a source available proprietary approach. I wouldn't contribute to a project like this, but there's no reason you couldn't make yours like this.
Free (beer/kostenloss, not freiheit) is magic and does magical things to people's brains. Also, public software work is really valuable to attract higher quality results.
If you seek to balance a ledger, then it's not free (in either sense of the word).
I wonder to myself: these articles, are they really written by open source software contributors who regret their contributions? Or someone on the outside looking in, wondering how they can "fix" things?
> Making money is not incompatible with Open Source nor Free Software. Making lots and lots of money is not incompatible with it either.
Well not at a philosophical level, but there appear to be some practical difficulties.
I still prefer Louigi Verona's deep ruminations on free/libre software:
https://louigiverona.com/?page=projects&s=writings&t=philoso...
My view on free software has changed over the course of my life. When I was in my 20s, my big concern was being able to do things like afford the mortgage on my house and be able to maybe be able to afford to get married and have kids or at the least pay off my credit card debt. Time spent on free software was unlikely to help with that and I focused my development time on things that would bring in money.
Now, in my 50s, I’m looking at retirement somewhere between 4 and 14 years away (although given the ongoing demolition of the United States, it might be more like 24 years away). I have little interest in building a side-business in software and, frankly, the flow of money in software is such that it wouldn’t be likely to be a viable concern anyway (to give just one example, in the 90s, I spent hundreds of dollars on compilers and database software that today are expected to be given away for free). As a consequence, the code that I write on the side I release with the least restrictive license possible because, frankly, I don’t really care if some corporation takes it and incorporates it into their product (in fact, I’d be kind of honored), and while it would be nice if they would kick me a few dollars, given my experience working as a corporate developer, my guess is that the few dollars that have come my way aren‘t coming out of corporate coffers but out of other developers’ wallets (for which I’m truly grateful, even if the actual dollar amounts likely don’t even cover my marginal expenses on the work).
There's a reason why FOSS is by hackers for hackers—it's software written by people to solve their own needs. They release that software as Free/Libre so that others can benefit from it if they would like to, but first and foremost it's written because the author themselves saw a problem and wanted a solution. What's changed is that as startup culture infected hacker culture a few subcultures developed within FOSS that thought that that wasn't good enough.
One of those subcultures decided that it was an imperative that they make more money off of their FOSS projects than anyone else does—this is how you get the Matt Mulenwegs of the FOSS world, who diss on other FOSS developers because "most of the value" of their projects was "captured by others" [0]. They see FOSS as a revenue stream and if it's not a revenue stream (or if it ends up being a bigger revenue stream for someone else than it is for you) it's a failure.
Another subculture thinks that FOSS should be mainstream. That's the idea typified here: by "hackers for hackers" is selfish and we should instead "reshape our thinking towards more political goals and values".
The fact is that both of these perspectives are fundamentally not about adapting FOSS to the times—they're both about building two entirely different philosophies. FOSS is not a business model, and what the Matt Mulenwegs of the world actually want is to be a tech startup CEO. And FOSS is not and cannot be mainstream because to become mainstream requires a bucketload of UX design and user support work that no one in their right mind wants to take on for free as a hobbyist.
FOSS is by hackers for hackers because it must be. If you want to make more profit than anyone else does off your project or if you want to pursue "political goals and values" you're looking for an entirely different type of organization.
[0] http://web.archive.org/web/20241014235025/https://ma.tt/2024...
The answer is not that complicated but people seem to make it so out of desire to either cover all the corner cases and/or tackle a more tangential issue related to free software (such as compensation). While valid areas of exploration, it can make one miss the forest for the trees.
I feel the question can be answered as simply as once upon a time a question like "how many angels can dance on the head of a pin?"
Who is free software for? Anyone and everyone.
Free Software is "for hackers" because, quite frankly, most other people just don't care about the goals of the movement. No amount of "getting out of our comfort zones and modes of operation" is going to change that. And the political angle generally makes it less palatable to general public, not more.
"Meeting people where they are" would require a drastic change of goals, to the point where the result would be unrecognizable, and wouldn't prioritize the things that hackers care about. Which isn't wrong per se, but then why expect hackers to drive such a movement?
> Free Software is "for hackers" because, quite frankly, most other people just don't care about the goals of the movement.
Until their car dashboard starts showing them ads. Or they can't even screenshot streamed movies to comment on for their friends. Or the app they were using to organize protests gets removed from 'their' phones [1]. Or they can't fix their tractor [2]. Or their printer refuses to print with 3rd party ink, or includes tracking dots to betray its user. Or their videoconferencing program gags them at government request [3].
[1] https://www.nytimes.com/2019/10/09/technology/apple-hong-kon...
[2] https://www.vice.com/en/article/john-deere-hit-with-class-ac...
[3] https://apnews.com/article/2ba80f30ecaf5aa37852164c3d149514
You're right that many people are fed up with this kind of stuff. But, to them, the obvious solution is to have laws that prohibit it, not copyleft.
I think the author is confused because he doesn't realize that hackers and muggles are fundamentally different.
Hackers come into foss because they get fed up about how shitty proprietaryware is. Shitty because it gets in the way of their hacking. Muggles don't care about hacking, so to them it's not shitty.
The author thinks that 10x people = 10x better so we have to bring the muggles on board. But actually it doesn't matter, non-hacker users don't contribute much to foss. They don't send pull requests, and don't submit useful bugs. Most they'll ever do is nag you into making the thing less hackable.
What non-hacker users contribute, in theory, is widespread acceptance. This comes with its own benefits, such as F/OSS solutions becoming de facto standards, which in turn means things like e.g. Linux drivers for more software, content providers not trying to lock it out etc.
However, this all can only happen if there are a lot of non-hacker users. Unless and until you actually get to that point, there are no benefits, but the support costs are there regardless.
And I have serious doubts about getting to that point, ever. I remember being much more... optimistic - or perhaps naive - back in early 00s. Many of us really thought back then that F/OSS would surely win, gradually, slowly, but surely. Now? Things are better in some ways, but overall I'd say that the environment is more hostile towards openness than it ever had been before, with widespread adoption DRM and lock-ins throughout the entire software ecosystem, massive centralization of online services, and pervasive surveillance being pretty much the accepted social norm. On top of all this, smartphones became the dominant personal computing platform, and DRM and restrictions are even worse there.
So, I don't know. Maybe you're right and this kind of "secret knowledge" for the initiated, those willing enough to dedicate significant time and effort to make it work, is the best we can have. It sure is a depressing thought, though. And I'm not so sure we can keep even this much, to be honest. E.g. at this point I fully expect to see bans on things like backdoor-less end-to-end encryption in my lifetime. Or, privacy aside, how about protocols that are straight up non-implementable because of DRM? Free software has limited usefulness if it can't meaningfully interoperate with things outside of the bubble.
I agree with a massive caveat.
The beauty of free software is having your cake and eating it.
It seems like today catering to hackers and normies with slightly different builds or packages would be easier than ever.
The author really doesnt craft a path forward, but nothing he implies would indicate that hackers lose anything (other than like you say, lack of nagging)
> Free Software is "for hackers" because, quite frankly, most other people just don't care about the goals of the movement
Sure. But, like, the world wide web. There is a playbook for making this work. Where Silicon Valley's mythmaking bites hackers in the ass is in pretending to write out the government's role out of the history of the internet.
Free software is a public good. Public goods have basically one good funding model.
Free Software isn't for hackers. It's for users. Hackers are also users, but they're not special.
The purpose of Free Software isn't to teach people how to program. It's not a scholarship program for gifted kids to generate startups. It's to institutionalize the right to access and change that software that defines the behavior of the machines that control our lives. You may say that only a programmer can take advantage of that right, so that means that Free Software is meant to create hackers. But I can hire a hacker just like I hire a plumber, or like my condo association hires a roofer, or my town hires a construction firm.
There should not be machines in your life that run on secret commands that you are not allowed to know about. They are not working for you.
edit: it is best not to talk about Free Software Licenses and Open Source/Creative Commons licenses at the same time if you're not talking as a consumer. They don't have much to do with other, other than that their development models are similar and that Open Source software will always be available over the same channels as Free Software.
Copyleft licenses like the GPL are very restrictive licenses that try to make sure that access to the code that they cover will never be restricted to a user of that code. They are an attempt to grant the rights that people should have to all code to enough code that one can accomplish one's goals without having to touch a locked-down black box, and to forbid the makers of locked-down black boxes from taking advantage of, embracing, extending, or extinguishing that code. The copyright holders of copyleft code are granting you that right of access and modification instead of the government. Open Source/Creative Commons has zero interest in that.
Open Source/Creative Commons are liberal licenses that allow anyone to do what they want, and make no (or very few and trivial) demands. You can take it and lock it behind any license. They are only compatible with copyleft licenses because copyleft licenses are part of the class "any license," not for any other reason.
>Free Software isn't for hackers. It's for users. Hackers are also users, but they're not special.
This is wrong. Hackers like to hack, which requires having control over their tech. So you get foss which gives you lots of control. You get things like GPL which tries to protect that control.
The control and freedom to hack is not relevant to regular users. Regular users don't want to modify and hack their software, they're happy to consume it passively. To them, the control is not a factor and foss is just another product. The pro is that they can maybe save a few bucks, unless the proprietary alternative is also free (they don't care about privacy or ads). The con is that it's less polished, the website is ugly, the docs are confusing and don't have cutesy cartoons saying to press the shiny button to make thing go brrr.
Lumping both of these in the same category just confuses things. Hackers and non hackers want very different and fundamentally incompatible things.
While I mostly agree with this, it's inaccurate to say that Open Source Software is more liberal than Free Software. The issue of copyleft vs. permissive licenses is orthogonal to that classification. The difference between Free Software and Open Source is mostly that of intent: Free Software is primarily political, with Freedom being the ultimate goal, while Open Source is primarily practical, with high quality software being the ultimate goal. In practice, almost all software that meets the definition of one meets the definition of the other.
Rather than "political" vs "practical", I prefer to say, "social" vs "economic". Free Software is about helping your neighbor. Open Source is about cheaply producing a good.
Either one can produce high quality software.
> Free Software isn't for hackers. It's for users. Hackers are also users, but they're not special.
This is exactly the dated attitude that I think the author is talking about.
Free software is for programmers. Once upon a time that was a distinction without a difference. Today, the fraction of users who can take advantage of source code rounds to zero.
Yet, FOSS advocates keep beating this drum to “users” while entirely ignoring the reality that today’s users are interested an entirely different value proposition than they’re offering.
Computers aren’t professional tools anymore. They are commodities. Computer users want source code like grocery store shoppers want a tractor and a plot of land.
It's not about source code access, it's about guaranteeing the four freedoms. As a user, with proprietary software you have to worry about whether you can share a copy with a friend, sell it on a CD to a guy on the street for $5, or use it at your work place. With free software you can do all of that without worry. For better or for worse, I am not a programmer, and free software is important to me. I have often wished I were a programmer in order to better help and contribute, but have long struggled to learn. I doubt I'm the only one in such a situation.
Yes, but the difference between freeware and FOSS is rights to the source code. Most users today would only want to exercise freedom 0, so FOSS really isn't offering them anything that appealing. And in fact, freeware is more popular than ever today.
>Most users today would only want to exercise freedom 0, so FOSS really isn't offering them anything that appealing.
I was aiming to address this in my previous post, as a non-programmer user, listing realistic examples that the other freedoms give you. Being able to use software at both home and work, or to share software with a friend without worrying about legal trouble is a benefit I think anyone can appreciate. Lots of software will charge [extra] money for commercial use, limit installs to x number of machines, etc. If you get used to using free software, you can avoid these issues.
Also, while not related to the license exactly, using free software, you'll tend to not run into ads like you do with a lot of freeware. I recommend apps from F-Droid to people with Android phones for this reason a lot. While grabbing a popular file manager or music player from the Play Store often seems to get you an ad-ridden mess, I never run into that with anything from F-Droid.
> Computer users want source code like grocery store shoppers want a tractor and a plot of land.
You entirely ignore ecosystem effects. Users don't want source code, but they do want programs that won't betray them. Open access to source code is how such programs come about. See my other post on what happens when that access is taken away: https://news.ycombinator.com/item?id=43385000
Non-experts are neither aware of this nor have no way of independently verifying that themselves. If you observe today's software marketplace, people are not switching to FOSS in response to privacy concerns. They're not even aware of what FOSS licensing is.
And regardless, the issues you list in your linked comment aren't even anything FOSS has a solution for -- all of them are only partially software products. You can't replace something that is more than software, with software. At some point an IP licensing agreement just simply isn't enough can't plow a field. You need a factory, some steel, and some diesel fuel. Farmers don't plow their field with software, they do it with a tractor.
>but they do want programs that won't betray them
Most people don't really care. In your link you mention cars locking people out, well, they still use them don't they? And they even pay huge sums of money for the privilege.
At the same time, FOSS software is also historically filled with corporations and slimy individuals rugpulling their users (many examples over the years, Emby jumps to mind, so does the Simple Mobile Tools set of apps). It's so common on mobile in particular that a major benefit of F-Droid[0] is that it just chooses to sit between you and the developer to make sure that the developer doesn't get up to any shenanigans.
Open source/Free software isn't a panacea against that and if anything has led to the formation of a development moat, where AWS et al. can profit from FOSS, but it can do so by muscling out all profits from the smaller developers[1]. FOSS by design can't offer any legal recourse against that, which is a pretty major mark against it for most people.
[0]: https://f-droid.org/
[1]: A less aggressive, but no less insidious version of this is the "this piece of software has a ton of extra services it relies on tech stack", where to replace one single service, you're essentially setting yourself up to rehost an entire corporate network and said hosting usually requires resources not easily accessible to individuals (ie. Multiple servers). This usually happens with organizations that want to maintain some level of control over their users; Firefox Sync for example isn't feasible to truly selfhost, since the Mozilla accounts service also needs to be selfhosted and it's a massive PITA to do that.
Not sure why you are bring downvoted. Your argument is clear, and drives the discussion forward.
I would agree that Free Software can (and is) used by everyone but only a very tiny fraction are able to exercise any of the Freedoms it advances. But that's OK.
My mom doesn't know the difference between Windows and Linux, and the license distinctions are meaningless to her. But she could still get meaningful value choosing one over another.
FOSS beats the drum, not to turn people into programmers, but rather so that some tiny fraction of the next generation understand the building blocks thst got us here, and carry the game forward.
Personally I think there is a lot of ignorance and misunderstanding of both OSS and Free. Lots of people want something outside the scope of these, and rather than create some new category, try and morph one of these.
Complaints like "give nothing back" or "AWS is eating our lunch" etc are all irrelevant to OSS and Free. They are topics which are not interesting to those licenses.
I would say this; if an OSS or Free license does not suit your need, use another license. Or write your own. There's no problem with that. (For 25 years I've been shipping source, but without distribution rights- "source supplied", not "open source".) I freely use OSS and obviously the OSS parts keep their OSS license.
99% of my customers won't ever look at the source. That's fine. I found a license that works for me, and them.
I think end users can certainly benefit from the downstream effects that FOSS has on developers, integrators, and other technology professionals. But most FOSS advocacy ignores this in their messaging, their strategy, and in the spaces where they operate.
I've only ever seen FOSS outreach to industry groups, academic groups, hobbyists, and other professional or professional-adjacent groups.
They very well could make their case to the modern end user that their approach is better than closed source SaaS solutions, or other freeware. But they don't. It's a missed opportunity, and they're going to lose relevance and influence if they don't appeal to those who drive market demand for software products.
>> They very well could make their case to the modern end user that their approach is better than closed source SaaS solutions, or other freeware.
Lots of people benefit from Free Software, yes. But asking the consumer to understand licensing and the implications thereof would be a fantastically ambitious and expensive project.
Ultimately yes, FOSS outreach is targeted, because the first-order benefits are targeted. Outside the internet itself (where the message is there for anyone to find) only very narrow groups would get any value from a discussion on licenses.
For example, say I make software for small-town bakers. My software might be open source, or closed source. My bakers couldn't care less. They care only about the first order benefits - does the planner work etc.
Ironically "no-one" chooses Linux because of the license. They choose it because of the price. And consequently they dominate where "number of machines" is the key issue (aka servers).
Consumers choose Windows or Android because of the first order benefits there. They are better than Linux Desktop or Linux on the phone.
(And yes, I'm aware Android uses the Linux kernel, but Linux is not sufficient.)
The whole "year of the Linux desktop" argument is for Joe public to care about License. 25 years in, and I think it's safe to assume that approach has failed.
>> to industry groups, academic groups, hobbyists, and other professional or professional-adjacent groups.
FOSS exists today precisely because these are yhe groups that matter. And they are, I think, the groups that will continue to matter.
> For example, say I make software for small-town bakers. My software might be open source, or closed source. My bakers couldn't care less. They care only about the first order benefits - does the planner work etc.
And in today's business landscape, the model that the baker and the software developer both like the best is permissive-FOSS licensing with proprietary extensions and SaaS delivery. Or, an entirely proprietary application that runs on top of FOSS server components. Which is not looking good for the future of FOSS in terms of contributions back to the community. In this way, it looks like FOSS is becoming less for users and more for venture capitalists and startup founders :)
I feel like perhaps you are seeing a change in this?
Obviously the baker doesn't care about the license- but certainly a SAAS approach has gained momentum. Whether this is because the baker wants it that way, or because its easier, and more profitable for providers that way, is debatable.
Does FOSS offer less for end users than before? I'm not sure. The benefits of getting source code (reading, editing, compiling) has always demanded a skillset far removed from most users. Doing SAAS as FOSS just adds "hosting" to that list of skills.
Yes, lots of startups advertise "Open Source" but that's just a marketing thing. They pivot the license when someone else starts getting commercial value from it. Folks contributing to VC projects know what they're getting.
But there are lots of modern Open Source successes as well. Chromium has spawned a whole slew of browsers, Linux continues to grow (code wise), and most programming languages are now open.
Writing Open Source has always been difficult from a commercial point of view, simply because its cheaper to commercialize work that didn't cost development $. But again, that's always been true. Red Hat didn't write Linux.
Is the future for new Open Source projects rosy? I don't know. But folk have been predicting its demise since, well, it started.
Free software is definitely not for programmers
Source code is not useful unless someone is capable of using it.
The most important thing these articles don't understand is that it doesn't matter that they don't understand.
"How can this possibly work?" It doesn't matter, it already is. Almost every example of a failure or problem they point to is usually no failure or problem at all.
When someone doesn't understand or can't imagine why anyone should work on gpl software, all I say to that is "OK".
Ok. You don't think it makes any sense. Ok. My participation does not hinge on you understanding, and the quality and utility of the software and the garden that it grows in does not hinge on you understanding. Go forth and have nothing to do with this nonsense, with my honest blessing. I can do a thing without needing to try to convince you to do exactly as I do. We can be different.
Your participation hinges on you understanding, but if you aren't still a child and don't understand after 15 minutes with google, then it's because you don't want to understand and the whole deal is just not for you and you were never going to participate anyway, and that's ok. There is no problem to resolve there.
Open source, in the traditional simple gpl form without any further "fixes", already works exactly the way it was always intended to, regardless if you understand how it can possibly work. It's working perfectly well right now. All the big companies making billions while using gpl software is perfectly fine and they didn't steal anything unless they actually violated the gpl. (Which does happen but that's a seperate issue. People steal commercial software by violating it's license too.)
All the fixes actually just break it. As another commenter said about trying to add special modifications for commercial use etc: "You can do that. I won't contribute to a project with those terms but you can do that." And there's the breakage right there.
Free Software is for the people who build it. Great closed source companies become great through their ability to sell to many customers. Great free software projects become great through their ability to coordinate and integrate the contributions of many people. The larger the percentage of "people who have contributed something" to "people who use it", the healthier the project is (think of this as [C/(C+U)] where C=Contributors and U=Non-Contributing Users; every Contributor is a user).
The reality that you can use many free software projects without much contribution at all is a happy and good accident of the system; but a dangerous one that can and has been often-accidentally weaponized to burn out the actual contributors.
>> The reality that you can use many free software projects without much contribution at all is a happy and good accident of the system; but a dangerous one that can and has been often-accidentally weaponized to burn out the actual contributors.
I'm not downvoting you, but I expect people are disagreeing with this last paragraph. (There's a metaphor between people voting up and down based on what they agree or disagree with - something this site doesn't advocate for, and the way some interpret Free licenses, and their goals.)
I agree you can use any Free Software without any contribution at all. And that only the tiniest fraction contribute any code. In most projects the contributers past the owner are zero.
This is not "by accident" though. It's very much by design. We can assume RMS thought a bit about his license and he most certainly expected most of his users not to need to read the code, much less add to the program.
The Freedoms are thus Freedoms, not Requirements.
Burnout is an issue completely unrelated to the license. It happens when an employee doesn't find a balance in their life and work. It happens with free, and non-free software alike.
Yes, lots of programmers burn out because they cave to the demands of unreasonable bosses, or users. But that's unrelated to license and I'd certainly not a "weapon".
being free it’s probably for anyone including everyone blogging about it. Enjoy
“Winning” is not for everyone because in duality you need losers to define winners but that’s more like politics.
Yup, this article hits it straight on the head. Both Free Software and Open Source (the later is just a more corporate-friendly version of the former) are fundamentally 80s ideologies, concerned about software you run on your computer and the whole thing was created because Stallman couldn't fix his Xerox printer. 40 years later, many things have happened - the Internet, widespread usage of embedded devices and SaaS - and the movements just.... didn't adapt to them at all.
The whole culture is so enmired in the original definitions that it simply can't see anything outside it - so everyone just drains from the creative commons but doesn't really expand it. There's a reason why most prominent FOSS applications are old pieces of software. It's just not a model which works in the 2020s, except for software libraries.
Ironically, many Free Software projects are almost indistinguishable from "look but no touch" software - while in theory you can make changes, in practice you can't because contributing to it is next to impossible and the project structure basically spells the definition of NIMBYism out.
There's also a fundamental tension here - anyone who argues that the current tradeoffs of open source/free software just aren't good enough almost instantly gets dismissed with "well well well.... if you don't like it, don't use it/pick a proprietary licence lol". Or they go straight to victim blaming with "you should have known what you got yourself into, no takebacksies~"... There's a strange dynamic here where if you want to earn money from your project, then FOSS is not a business model, but if a megacorp wants to earn money from your project, then it's okay because everyone wins and you don't lose anyway. Or something like that.
The Free Software movement rejected all ideas which could have contributed to a solution to the ecosystem thriving - non-commercial licences, usage-based licences, dual-licences (in the sense of different terms applying to different users) and countless other things in favour of strict textual interpretation of founding principles. Religions didn't survive thousands of years only on textualism so Free Software won't survive this way, either.
>Ironically, many Free Software projects are almost indistinguishable from "look but no touch" software - while in theory you can make changes, in practice you can't because contributing to it is next to impossible and the project structure basically spells the definition of NIMBYism out.
These projects still allow you to make those changes locally - and that's the entire point. FOSS has never guaranteed that your changes would be incorporated by anyone else, distributed by anyone else, or recognized as legitimate or valuable by anyone else.
We're just vastly more conscious of that in the 2020s, in our era of internet celebrities.
>There's a strange dynamic here where if you want to earn money from your project, then FOSS is not a business model, but if a megacorp wants to earn money from your project, then it's okay because everyone wins and you don't lose anyway. Or something like that.
You're perfectly permitted to charge for the software, just as the megacorp is. The fact that people are willing to pay the megacorp for it and not you is unrelated to FOSS philosophy.
>The Free Software movement rejected all ideas which could have contributed to a solution to the ecosystem thriving - non-commercial licences, usage-based licences, dual-licences (in the sense of different terms applying to different users) and countless other things in favour of strict textual interpretation of founding principles
There is actual philosophy behind these principles, not just literal dogma. For example, applying different terms to different users requires them to agree to a different license because that's kinda just how law works, and you can't withhold the option of the OSI-approved licenses from anyone because non-discrimination is part of the philosophy.
The logical step here is what's alluded to at the end of the article: self hosting and doing things with FLOSS is impossible unless you're sufficiently one-of-us to get into the guts of software. We can only make this situation change when we recognise that including a wider audience is necessary, and for that we need to run things and support them for other people. Cooperatives are the way that strikes me as being the right one: giving everyone a share and a vote instead of the code itself. Nobody is going to vote to have the contents of their emails mined or their attention bought and sold.
The legal infrastructure we would need for this is a Cooperative General License. The "four freedoms" in the GPL are born from academia and they show it. The "open" licenses are libertarianism and opportunistic IP laundering (from one failed startup to another) for the most part. A CGL could incorporate a restriction on use: personal or cooperative, but not corporate, not unless the form is the right kind.
This is an old debate in tech but as the author says, one worth revisiting. We shouldn't be working for bastards, and certainly not giving them the gift of our incredibly valuable work.
There's always going to be some tension between people who push free and open source software for ethical or political reasons, and those who push free software for technical and practical reasons. This article seems to give a very unfavorable view of the latter, suggesting that the primary purpose of the "open source" movement was to make free software more palatable to corporations. Well, I believe that it was largely for that purpose, but when you put it that way, it misses the point, I think. The average enthusiast of the "open" movement sees openness as a major advantage in and of itself, for a variety of practical reasons, and the potential benefits in terms of user freedoms are largely just a consequence of that, whereas free software enthusiasts generally seem to be mostly concerned about the movement as a solution to the problem of software that doesn't respect the user's freedoms.
I don't have any issues with the concept of free software, or having strong ethical convictions, or anything like that; however, I am not convinced by the arguments the FSF and Richard Stallman have made on a philosophical level, and frankly, as a programmer, I largely don't even really care to spend much more time worrying about the philosophy of it all. I am glad somebody does, but it's not for me, at least insofar as much as I can avoid having to do it.
And that's okay, because I don't really have to. The fact is that many of us have always preferred open source software primarily due to the side effects that having software with permissively licensed source code available has: the freedom to fork, the ability to audit code, fix bugs, re-use code for new purposes the author did not foresee, and of course, the fact that it is also typically free-as-in-beer. The funny thing about this ad-hoc list of benefits is that it is exactly the same for corporations as it is for users; the reason why so many startups release open source software is because it is a good way to gain trust and goodwill early on. Companies and their employees are more willing to adopt open source solutions because of these things, even if they don't necessarily intend to exercise all of the options it gives them.
The reason why open source is somewhat lopsided, e.g. it's obviously better for corporations than users in many cases, is because the arrangement in which there is no discrimination on who can use something for what yields the maximum utility for everyone. The moment you add even a single restriction, such as "People named Rick are not allowed to use this software", it opens up a world of complexity. Open source licenses can get a little complicated, but the broad strokes are so much simpler and easier to reason about. Think of the complexity that unfree packages place on Linux distributions.
And honestly, the discussion of the "for hackers, by hackers" mentality doesn't really seem fully baked. What does it even mean to be a "hacker" anymore? What identity are we asking people to assume?
I have another theory for why the Linux desktop and Firefox are not winning: it's because they kinda suck. Don't shoot just yet; I use Firefox and Linux all day, every day, and Firefox is in relatively good shape nowadays despite Mozilla seeming to always piss away any opportunity and goodwill they have. But, for-profit corporations have more money, more people and move faster. Competing with Google and Microsoft is hard. Open source can never beat Microsoft and Google at their own game.
However, what open source can do is beat them at games they're not playing. While corporations move fast, open source endures. In the longer term, I believe many prominent open source projects will rise to the top, as we've seen with a select few. Look, I don't expect a "year of the Linux desktop", but if it does happen, it won't be what most people were expecting in the 2000s. Microsoft's enshittification will most likely just lead to a situation where Linux wins by doing nothing, so to speak... I think we're going to see a lot more of that as the software industry slows down.
But honestly, the fact that these behemoth projects exist is a testament to it's success. I don't expect an offshoot to be able to replicate its success, less surpass it. Before people jump on board, I think there really needs to be some more convincing arguments for why we should abandon a movement this successful for an unproven alternative. It's not unfathomable... but personally, I remain unconvinced.
If he doesn't like companies making money when he helps commoditize things that complement what they provide, CC does have a non-commercial variant.
Nah. OSS is low quality because the feedback loops do not include its users unless they are also developers.
There are exceptions, but the ideology of OSS resists attempts to make good products. Or, rather, people who are associated with the OSS ideology fail to build products.
AI tools are removing coding barriers so everyone benefits except those who depend on restrictions, but even they benefit in the long run.
A couple minor issues: the AGPL and similar licenses exist and have been effective at stopping classes of appropriation, and the cake quote is the common misconception. It's referring to the nutritious leftovers usually discarded, and free software means the corporate code deemed nonessential enough to release under free licenses gets spread around, so let's eat that delicious free software cake.
I don't think this article proves it's thesis.
It claims FOSS isn't winning because it's stuck in the hackers for hackers mindset.
But the two big examples it gives for FOSS not winning are Firefox vs. Chrome and the Linux desktop.
But Firefox and Chrome are more or less the same product. The reason Chrome won is because it is faster. The difference is smaller now, but for a long time it was significantly faster, and that trained a lot of people to automatically install Chrome.
The Linux desktop is a different discussion. But even here, it's mostly a network effect. If you put a layman in front of a fresh Linux Mint install, and in front of a fresh Windows install, I don't think they'd have much of a problem with Mint.
But most laymen aren't laymen, they're Windows power users, to a certain extent, because that's what they've been using their whole life.
So I don't see how having a less hacker-for-hacker mindset could help either of these fights.