The Lack of Compensation in Open Source Software Is Unsustainable
trstringer.com
trstringer.com
If someone releases code then that's it. You have it. Do what you want with it (within it's licence). Be thankful.
Unless that guy specifically says he's going to maintain it for free you are entitled to exactly absolutely nothing more you ungrateful little git.
If you add this code to your project, you should fix and share any problems you encounter as a token of gratitude.
That works both ways though, a maintainer cannot expect users to not complain (so they have to develop management strategies where they ignore the noise, rather than try to engage/capitulate). A maintainer can also not lament when users do not feel the desire to contribute monetarily.
I believe this is called “please use the template when submitting an issue” and we all know how that goes.
I’ve honestly found myself laughing like a madman at GitHub issues where the maintainer calmly and repeatedly tries to explain to the increasingly disgruntled reporter to UTFT (use the fine template)
Also, "Issues" should really, really have a different name.
Taking and cropping a screenshot takes more time than just uploading the plaintext log file conveniently and lovingly placed right next to the executable. Please, just upload the file. Please.
The project was a Big Deal in the circles I was in. Hundreds of thousands of people knew of it, we had hundreds to thousands of active users each with dozens of clients. It was so stressful and I caught so much abuse that I exited software altogether. We did get some money, but my cut worked out to $60 a month.
I still write and publish code, but I don't work on anything that anyone wants to use. I don't get involved in communities and I don't interact with users.
And you know what? I am much happier now.
Github could improve the experience 10 fold.
Eh… Isn’t that fairly close to the definition of entitlement (i.e. you complain when you have no reason to)?
Instead, it is more productive to develop strategies to ignore the noise from entitled voices. You do not need to respond to them. You do not need to convince them. You do not have to keep them happy.
There is: it discourages people from acting entitled.
The solution is quite simple really.
1) have your repo private and release only tarball files or have a read only repo with no pull/merge request functionnality 2) do not use an issue tracker
Basically, do not use a forge such as github/gitlab, at least not publicly. Problem solved.
More maintainers need to take the "no guarantee of fitness for purpose" part of their license more seriously. Don't fall for the temptation to be everything for everyone. Don't cave to social pressure.
I get this, as a very bottom line. But a lot of great projects are great because the developers consciously want people to use and depend on their work.
Their is a living changing informal social bargain, unique to each project, along with licensing, economic and other concerns.
Is it unsustainable like the title of the article says? I suppose, but it's not some state of affairs that is unavoidable or that we are stuck with through no fault of our own. A guy wants people to depend on his work that he does for free, a company sees a core component of their business that they can get for free, a few years later the guy is upset that he is maintaining this thing everyone depends on and the company is scared their business will fall apart without him. They each got themselves there, it's a predictable state of affairs, the solutions for each party are very clear. As a developer though, the solution is what I've outlined: take the "no warranty of mechantibility or fitness for purpose" part of your license seriously and tell entitled people to solve their own problems.
I use tons of free software. I've never either demanded that anyone work on it for free, nor have I expressed any sense of entitlement or expectation.
I've never maintained a popular open source project so maybe there's something about the situation I just don't understand. But it seems like:
> Thank you for your feature request, we will add it to the backlog. The core team doesn't work on unfunded feature requests because they use up a lot of time and resources. We are happy to review high quality PRs from anyone interested in implementing the feature. We also have a variety of sponsorship options, and a list of past contributors and maintainers available for contract work.
would be reasonable and polite?
One reason is that you're being told that that's an awful thing to do by basically every resource on "proper open-source" you can find.
Another is that these people are pretty good at starting shit storms trying to ruin your reputation if you don't comply with their unreasonable demands.
It's also worth nothing that some requests/issues/questions might be reasonable when viewed in isolation but not if there are hundreds of them.
Think for example stuff unrelated to the project but where you as a hacker could nonetheless help because you do know the answers/possess the skill. For me at least, I find it hard to deal with that, because I know that I could in theory help that person. I just can't in practice because time and energy are both finite.
> We are happy to review high quality PRs
Are we though? There's a lot of work attached to reviewing even high-quality PRs. Also, even if the PR is high quality, the maintainers will still be the ones maintaining that new feature so it's still significantly more work.
It can be ignored, yes, but the worst offenders go beyond a polite discourse and will send emails to whatever email they can find, DM you on Twitter, Mastodon, etc., and drop weird and annoying comments anywhere they think you have a chance of seeing them.
Some people have thicker skins than others, but it's just a bit tiresome no matter how much you can deflect.
- for American developers, many of them have provisions in their employment contracts that allow them to do unpaid open source work, but ban any kind of commercial (i.e. in exchange for money) activity. The fact that this reach into off-time by employers is possible is nuts anyway, but doesn't make the problem go away.
- as soon as any kind of money is involved, a lot of jurisdictions have provisions regarding warranties and liabilities - and these can be pretty enormous, see the log4j fallout. Some of these can be put aside by contracts, but nevertheless it's a legal minefield.
- some jurisdictions don't allow you to just take money in exchange for a project, it exposes developers to tax and social security liabilities. Even labeling such stuff as pure "donations" isn't safe if your tax auditor is particularly focused on nailing you.
- what to do if someone from a sanctioned country donates you money? What to do if you're European and get money from someone in Cuba (which is not sanctioned by the EU), but are employed by / work for American companies or intend to travel to the US?
- what to do if some arms manufacturer donates you / funds money for a project that could be used in weapons? Virtually all countries have some sort of equivalent to ITAR regulations that you really don't want to run afoul of.
There very likely isn't just one reason that applies to all maintainers.
But there are some reasons one can hypothesize that probably apply to some or many maintainers.
This happens less where the FOSS choice is a drop in a sea of established proprietary packages (FreeCAD, KiCad, Godot) but way way more when they have already established themselves as the popular pick (OBS Studio, Blender) so they get flooded by less tech-savy, more casual users that don't really see the value of open source other than they don't pay for it.
"Normal" people have always had stuff given to them for "free" (either "you are the product" or built-in licenses like Windows) so they don't realize the goodwill and sacrifices that FOSS goes through.
this was solved 30 years ago by an important socio-technical invention called the FAQ, used together with a social convention of not elevating or rewarding vexatious messages.
However, there are entire industries that leverage open-source / free software, and put unreasonable, uncompensated demands on it.
At the end of the day though, I don't see the problem. As a maintainer of open-source, gratis, software, just don't do the work. It isn't like it is a job. If you don't do the work, they can't fire you.
Is that good for the community? Surely not, but who is asking whether the status quo is working for the developers? Nobody but them. So look out for yourself, and scratch your own itch, but don't treat open source as a job.
Weee. Exposure.
Free-rider problem [1].
Imagine visiting a coffee shop and telling the barista: "hey, by the way, I can make my own coffee at home for free, you know?"
People seem somehow OK with thinking this way when it comes to software.
People will spend $10 on a coffee drink at Starbucks without blinking. Suggest that they spend $10 on a piece of software and they’ll throw a fit and claim you are taking away their rights.
They seem to only be present in the big metro areas that attract a lot of tourists, travelers and immigrants who are familiar with the Starbucks brand and go for that out of habbit and know quality, similar to how McDonald's is so popular.
As you say, people go to Starbucks and McDonalds for familiar known quality and you should be happy for it. That way the tourists and immigrants stay out of your local cafe.
Edit: to be clear I'm not defending supranational billion dollar corporations here.
This is so ridiculously optimistic and misinformed. In Poland’s cities, for example, the cafes mainly belong to chains like Costa Coffee and Green Coffee Nero, which act identically to Starbucks in terms of prices, quality, and range of drinks. There aren’t many independent cafes left, let alone ones with lower prices than a Starbucks. Similarly, in Helsinki the choice largely comes down to chains like Espresso House and Robert’s Coffee that are no different than Starbucks.
Most of those people won’t spend $100 on open source software.
If you make a purchasing decision, your ass is on the line when people ask why certain product/service isn't fulfilling some arbitrary business need. In theory, assuming the functionality is part of the advertised purchase, the liability of the thing you're paying for lies on that third party. You did your due diligence. If you chose some open source combination with in-house build, they're going to question why you didn't outsource some envisioned cheaper third party option (sometimes, this is a legitimate strategy, often from my experience it's not). So you default to big vendor big solution to protect yourself.
Apologies are made and blah blah, discussions about "what alternatives do/did we have" and you often end up landing on implementing some mixture of leveraging public domain software and in-house customization atop/leveraging it to solve the problem you were paying for. In the end, you end up doing what you probably knew was the correct path anyways: this vendor solution is questionable, it doesnt completely align with our business needs, and it's not going to get the actual need done. Conveying that to business leaders is often impossible though. So, to pass liability/responsibility and cover your ass with incompetent business leadership, you throw often thousands, tens, or hundreds of thousands away.
I've had this discussion so many times and sat in these meetings so many times it grows tiring. The fact is, sometimes a generic solution works for your business (Office for example is a pretty generic need and often aligns), often it really doesn't (some arbitrary more niche/custom thing you do? Maybe, good luck).
It's like telling people not to be Christian/Buddhist/muslim because religion is just a bunch of fantasy stories.
The whole point of the GPL is to change "should" into "must" for these fixes. That makes it a two-way street.
The entitlement and ranting is a direct consequence of the recent GPL-hating campaign.
Any project which is not a going concern should:
A: Not, IMHO, be called a ”project”
B: Be clearly labelled, in its public-facing information, as being offered as-is, without any implied updates or future development.
So if I had a text file in the root of my repo that said:
THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
I genuinely don't understand what's so difficult to grasp here.
Pretty sure every open source license includes this in the warranty line...
If you read the license (for most licenses anyway), that info is clearly right there.
The problem is that the barrier to use any software is so low that it attracts people who have no clue and demand support. I am not talking about you, even the larger companies always mention a wish to force smaller developers to patch security issues for free, this is an issue in supply chain security at the moment.
The gist is that you can fix it yourself.
(This is a bit like how a stable economy depends on there being a crucial threshold number of long-term, high-trust relationships. You cannot have a functioning economy when everybody is always backstabbing everybody else. Similarly, you arguably cannot have a functioning Open Source ecosystem if everybody is just throwing code over the wall all the time.)
> The problem is that the barrier to use any software is so low that it attracts people who have no clue and demand support.
I think this phenomenon is caused by:
1. Some users being a bit whiny and entitled, just like some people are rude to waiters. Some have been taught and brought up to behave this way, and others have just gotten into bad habits.
2. Many developers being overly defensive when presented with legitimate complaints from users. This is just human nature, harmful as it may be.
3. Due to 2., users exaggerate and act rudely when reporting complaints, because they expect pushback from developers. This then exacerbates 2. again, leading to a vicious cycle.
Some developers who are burnt out by 1., and are not realizing what is going on, are, as a way of psychological self-defense, adopting an attitude of “I don’t care about you users, you’ll get nothing and you’ll ******* like it.” This then necessitates the same developers to argue that all users who expect anything are merely “entitled”, because if any user’s expectations would be reasonable, then the developer’s attitude would be unwarranted, and the developers feel that they need that attitude for their own well-being.
None of this is new; the old jargon word “lusers” was frequently used in ages past with contempt and disdain for users.
Do you have standards? It doesn't sound like you do.
People with healthy boundaries set standards for themselves as far as what they give to others and what treatment they accept in return.
I think the fallacy in your argument is that you're blaming contributors for noticing that they aren't being compensated for the work they've done, rather than blaming others for using that work without giving anything in return. I see your sentiment reflected in society in the way we treat low-wage workers with disdain for not doing more lucrative work. You're applying the principle of rugged individualism to a systems-level problem.
A healthier way to approach this would be to list a number of possible solutions and debate them in an open forum like this. When we find solutions but fail to adopt them, then that's a criticism of our agency. We are all failing ourselves by failing open source contributors. Then we can look beyond that to find the reasons why. Which are obvious because they are the same as with any other power imbalance. The fault lies with the wealthy and powerful people and corporations who profit from free and low-wage labor. The solution is to organize labor into a unified front so that exploitation can no longer happen.
Our failure to solve open source compensation is analogous to failing to stop suffering in developing nations which provide labor and resources for wealthy ones. Your argument places guilt and shame on workers instead of identifying exploitation by the wealthy, which might be better spent on something like an open source endowment or UBI more generally.
> I think the fallacy in your argument is that you're blaming contributors for noticing that they aren't being compensated for the work they've done, rather than blaming others for using that work without giving anything in return.
Is anyone being forced to work on open source software? Unlike low wage jobs, where you could be forced in order to pay bills, eat, nobody is forcing anyone to work on open source ventures.
Just because you do something useful does not mean you are inherently entitled to compensation in the form you want.
If you are being forced to do something against your will that is bad. If it is some hobby you happen to like doing that is totally ok.
> The solution is to organize labor into a unified front so that exploitation can no longer happen.
Lol. What type of leverage do you think open source devs have to form a union? Open source in many ways is designed to remove all economic leverage from source code. Its not a bug its a feature.
--
If I follow your logic, then you're saying that there's no economic incentive to work on open source software, since it's not compensated financially. Which seems to create a paradox:
A) Capitalism doesn't apply to open source software because there's no exchange of capital for labor
B) Capitalism applies to open source software because it generates billions of dollars of revenue for people and corporations
It sounds like the only rational act under capitalism is to not work on open source software, since the work is not compensated.
Meaning that any solution we come up with will act outside of capitalism.
Can you present a solution that works within capitalism to fund open source software?
Edit: I forgot to mention that the primary power of organized labor is to withhold labor until compensated. For open source, that might look like deciding as programmers to withhold all of our contributions until we solve this. Since we won't do that, we're all scabs supporting the status quo.
That's the exact opposite of what they are doing. They are blaming the users for expecting more than they should. They should expect nothing more than literally just the current version of the code, as is.
Don’t do it! Who’s asking you to work on it?
I say this as a minor OSS developer who receives some donations. I love the money, but it would be pretty dumb for me to say they oughta pay me, because the only reason I work on it is addiction/compulsion/procrastination of real work.
If it bothered me that people use it for free, then I could just stop working on it and let it rot, just like lodash’ maintainer did for a few years. You always have the option.
I keep reading from project leaders like ESLint’s things like “if only they paid $1 each…” as if that made any economic sense at all. If they paid $1 each, you would not hand thousands let alone millions of users.
From another angle, if I gave a dollar a month to every Free Software program or library I used that month, it would cost me thousands of dollars a month. Obviously that isn't realistic. I have donated to projects before but those products need to individually make the case to me that my donation is needed. Expecting users to donate to every project they use by default just isn't realistic. Most projects won't get donations from most people most of the time; that's just the way it is. Developers should be at peace with this before they decide to personally commit themself to a project.
I have worked hard to escape closed source, commercial software, but I still purchase several box products (FamilyHistorian and SuperMemo come to mind immediately) from small companies. I acknowledge e.g., Microsoft or Adobe can make a killing by making their products subscription-based, but is that the way small companies (who can accept the lack of continuous income and charge a significant portion of full price for major upgrades) have to go?
I think the issue here is that we don't have a system where being an OSS contributor is a sustainable career path.
The value we get from OSS compared to the monetary compensation for that work is disproportionately small.
OSS generally doesn't depend on an army of people paid to make you buy shit you don't need or want.
We do. You work for a company that needs software X, and you contribute patches and fixes to X from your paid time. Software X is essentially a collab between different companies then.
The Linux base (kernel, libc and compilers) basically works that way.
As is true for many (most?) crafts.
If I am an artist (e.g. painting, digital art, etc). and give my work away for free, my fellow artists are not going to sympathize with my inability to make a living.
Ditto photography ("look at all these companies taking my free stock photos and not giving me any money!") Ditto music. Ditto writing. Ditto anything that involves a significant amount of creativity.
It's silly to expect SW to be any different.
True, I can't disagree.
However, it's important to note that there is something different about the software world, in that very large subsets of applications and infrastructure of large for-profit companies are built on top of the free labor of open source.
Not sure if that's true in any other craft.
If unpaid open source were to magically disappear overnight, all of the Internet and all tech-using companies would collapse immediately.
Let people do creative work without needing to work for exploitative corporate ghouls, and let the world-killing planet-burning exploiters cry themselves to sleep on their mega-yachts about it.
Why isn't it a sustainable career path to be an artist or photographer or musician or writer or open source contributor? (Outside of the lucky 1% or so at the top of those crafts, of course.)
Why are creative pursuits so much harder to make a sustainable wage on in our society? Why do we expect most of them to be unpaid hobbies? Why do we expect the arts and crafts that are the fruits of their labor to start at "cheap as free" unless they work to be insanely talented and are lucky enough to win corporate sponsorship/patronage?
I don't have good answers either, but there are a lot of questions of what actually are we valuing about our use of our labor here, as a society, in general, across the board.
Why shouldn't it?
They knew that before they started.
Why should it be? Most OSS authors get paid to work on their stuff by some company who wants it. If there's no one willing to hire you, then your software is not worth it. I don't mean this negatively. I have lots of open source projects that are not worth it. I have one that I've been hired to work on before, but am no longer working on it anymore
This is additionally supported by the point that only few maintainers will hand out commit rights to other devs. They rather complain about all the workload and demands than give up some of all of their control, so others with more passion or who are even paid can continue working on the project.
If you want to be paid for your code, then pick a license model that matches it, but it will mean that you won't rise to the top, as only a limited amount of people will use it.
I wrote some more thoughts on this, during the "Moq incident" earlier this year: https://duerrenberger.dev/blog/2023/09/23/foss-funding/
Why should that be the exception? Why shouldn't people's passions, especially when they are widely useful to others, not be encouraged and turned into paying jobs? Everyone wins. The passionate can keep doing what they're passionate about without having to split their attention to a job to pay the bills, and in exchange everyone else gets better quality output.
Because there is no fair way of estimating what someone's work is worth without a free market. Sometimes it turns up weird outcomes like maintaining a critical driver being worth $0.
If someone is willing to do something for free and the marginal cost of copying the work is 0 then by simple economics they will not get paid for doing the thing. Same logic applies to having children, advocating good ideas in politics and a lot of creative work.
Besides, why should someone doing what they are passionate about entitle them to a leg up? What about someone doing plumbing and hating it? They're making more of a sacrifice for the benefit of others, they deserve more money. And if someone is adding enormous value then let them who recognise it pay for it.
Nothing wrong with people working on their passions and making money of course, but words like 'should' are suspicious. Once you get to software development, people are in a world where market forces are fair and reasonable.
I think people get pissed off because they're working out of a spirit of generosity, and the users who they interact with most are definitely not.
Suppose my neighborhood regularly throws a block party, and everyone makes and shares some food, because they enjoy making and sharing food. Great! But suppose one neighbor grabs portions of everyone's freely shared dishes, packages them up, and begins selling your freely-shared food as plate lunches to others, and pocketing the proceeds. They come back and ask that you use more spice, and by the way, do you have any napkins and plastic cutlery? Is the right response here "well, if you don't love making and sharing food for the love of it, you should stop?" Or is it reasonable to want to share with people who are willing to engage in the same spirit of mutual benefit?
I build stuff and keep my projects to myself. I would happily share with other people who are building hobby projects for the love of it. I would happily let almost any non-profit use my work for free (perhaps excluding some political or aggressively religious organizations). I have zero desire to gift anything to anyone's for-profit company. But for some reason, there's a strong stigma against sharing source code but not allowing a total free-for-all of what it's published for. "That's not open source," I'm told. So I don't publish at all, but that has nothing to do with not loving what I build.
Meanwhile, a hobbyist makes music, and publishes some recordings with a CC non-commercial license, people get it. No one says, "oh if you object to companies using your recording as background in their ads, it must be because you don't love making music."
The dirty secret of the GitHub era of open source is that many developers and open source codebases would be better off if such a thing happened.
A lot of stuff associated with the development processes that are fashionable with GitHub and programmer Twitter provide net negative value (e.g. issue close bots) or neutral value at best (because they solve a problem that really exists but not at the scale that most downstream projects that opt in are actually at). If it suddenly became necessary or even just strongly encouraged for programmers to pay a dollar a month for things that are only as pervasive as they are because they are free and give the false sense of productivity, then we'd see a huge dropoff in the adoption and use of lots of things that are of dubious value to begin with.
But I agree with you. Open source projects shouldn't have any obligation to listen to the people if they change the license. Even if they intentionally bait and switch, the users should only expect the current version to be open source and shouldn't expect free updates for life.
Mostly this is the point where people start thinking about how to monetize it -- and they should! Your time is valuable, you should get paid for it. If you don't figure out how to get in the middle of the cash flows that are happening _because of your work_, someone else will do it.
I suppose projects can get so big that developers literally don't have enough free time to fulfill people's expectations even if they wanted to do it. At that point they should get sponsored so that it become possible for them to devote more time to their projects. Sponsorships seem like a perfectly ethical way to make money to me. With enough sponsors people can afford to quit their jobs and work on their own thing full time. I've seen news of developers who did just that when they got enough support. I hope they're doing well.
(I maintained a small OSS project for a few years. I loved the folks that would get all mad when it didn't do x even though the code should theoretically be able to do it and "it's not that hard!". If only they submitted pull requests with their issues...)
All of the people who demand things from the maintainers are. Very few of them are willing to sponsor or pay consulting fees. If you give up they complain loudly, and then often pay a contractor to do the work they wanted.
If they paid $1 each, you would not hand thousands let alone millions of users.
You could though, if the mindset of open source software consumers were to shift to 'pay for things you get value from'. Even if it was $1000 and only companies that paid, the landscape would shift dramatically.
I think OP's point is that the maintainer is under no obligation to deliver on these demands.
What if the mindset of producers were to shift? What if people only worked for free on things that they want to work on? Isn't that both more realistic and better for everyone?
Any solution to a problem that relies on humans becoming more ethical is not, in fact, a solution.
It isnt the engineering manager's fault at any large company. The policies make no sense, but there may be a larger wisdom that I do not understand.
Economic injustices and inefficiencies can and do exist in free-ish markets. But this isn't one of them. It's more like "old man shouts at supply and demand".
We could have paid for 50 months of that service with the money we spent trying to figure out who would be responsible for the bill.
If the reason you're doing something is for money, and you're not getting money, then stop doing it. It's that simple.
Ignore them. I mean, it's not like they're paying you. Scratch your own itch instead.
I'm probably being dense, but I'm still not seeing a problem. I think if we dig, we'll find non-monetary incentives that open source authors are a bit shy about admitting to. Otherwise why hold on to the project at all? If you're not being paid, and you think you should be, and that's the only reason you're building the thing, just stop building it.
Open source isn't unsustainable. We've got decades of evidence of that.
That is crazy. /s
Because the number of times I’ve needed a left pad lib is literally 0, but other, more useful libraries always have it as a dependency.
Just stop. It's a hobby. As in life in general, don't have hobbies you don't enjoy. That's weird. Give your project to someone else - if this is you, skip to the bottom for my advice on how to do this.
I maintain a couple popular open source libraries, and a ton of unpopular libraries. At least two of the popular ones are way more popular than this guy says his is. For the most part they maintain themselves via pull requests. A good test suite, static analyzer and code standards validator go a long way towards making this possible with far less intervention. If you don't have those, that is where you should put your time.
When someone does open an issue, you look at it for 2 seconds, make a judgment call. Either label it wontfix, or ask them directly if they think they can handle fixing it.
If they can't fix it, you again make a judgment call about if it's worth your time or if you want to slap "Help Wanted" on it. And that's it, go eat dinner with your family. That's all you needed to do. Takes 30 seconds.
Once in a blue moon there's a hotfix that demands immediate attention, but it should be rare if you're building good software. If it's not rare, I'm sorry, I have bad news. Every once in a while a platform change requires some refactoring, but even that will often get PR'd by someone in the community who enjoys living on the edge.
At the end of the day, if you don't enjoy it, don't do it. If the library is truly popular just hand the reigns off to someone else. Forcefully if need be.
And don't just post a vague "maintainers wanted" somewhere. No one will see it. Scan your contributions for quality PRs and actually reach out to those people who are clearly invested in your code enough to understand how it works.
One of the projects I maintain, the original author literally just threw maintainer rights at me after having opened a handful of quality PRs myself. Didn't even ask. It worked :shrug: but YMMV.
But why should they, if they can turn their hobby into a job -- there are many very valuable companies that have been built on top of open source projects.
I think donations is a bad way to handle it though. If you don't want to do the full SaaS model, another tried and true way is consulting and training.
I live in a country completely unsuited to making money from a hobby project on the side. To take donations I'd need to register as self-employed, and pay monthly for social security as long as I'm registered, even if I make no revenue whatsoever in that month. That's an absolutely awful idea for a project just getting started. I'd be losing a very appreciable amount of money, regularly. It'd take lots of effort to have enough support that I'm back to zero, and still not making anything. This is because this is a system made for plumbers, not for people doing rare jobs on the side which might some day grow into something serious.
Apparently the government sort of looks the other way until you start making minimum wage, but that's not a bet I'm comfortable making.
But finally, we (https://overte.org/overte_ev.html) managed. It took us a long time and a lot of effort to get a non-profit registered, and it absolutely required the participation of multiple knowledgeable people. This couldn't have been done as a lone wolf effort.
And after that hurdle of course you have to somehow get people to notice you exist, and convince them to donate. This is unfortunately a tough job for people whose main passion is software development and who threw all their effort into that. Going out there and figuring out how to advertise yourself and how to ask for money is a whole new skill to learn, and a big time and effort investment.
I highly suspect that the reason why libcurl shows up here every week or so is because Steinberg spends about as much time marketing it as writing code and it's clearly working. That project appears to be quite successful in getting donations.
GitHub Sponsors directs individuals to Open Collective https://opencollective.com/ , which can serve as a "fiscal host." The advantage here is that the default rule at law for how a group of developers working together will be treated is partnership, which means joint and several liability. Working with a fiscal host partitions individual liability from group liability.
But there are still open questions. I don't know all the details of how Open Collective works from a corporate law perspective. How do they partition the liability of different collectives that are hosted by the same fiscal host? That is important to understand because otherwise, the collectives that share a fiscal host are partners. This is better than all of the individuals who contribute to each and all of the collectives being partners (because there's still a partition between individual and group liability), but worse than if each collective's liability was partitioned from the other collectives' liability (and the fiscal host itself's liability).
This seems like an active area for legal innovation. No jurisdiction I know of is optimizing to maximize speed and minimize cost of setting up corporations. Yet without the ability to shield the contributors of capital and labor from individual liability, there isn't much a group of individuals can do in our modern economy. The problems that many DAOs have had is a case study in this.
People who care about open source should also care about scaling up the speed and minimizing the cost of incorporating. Stripe Atlas and similar services are underrated sources of economic growth for this reason, IMHO.
We need to keep in mind systems can be changed. You've set a very clear example that can be understood and communicated. Legislators should get to know this. Receiving donations for community work should be possible without hurdle (at least until it reaches a very high level) everywhere.
I call the idea that we spontaneously support what is right for all of us a Donation Economy. If most people are ethical, this would work well.
---
Furthermore, I think we should also fight, in the long term, for organizations (collectively supported) that provide this function (supporting community work). Like pollution is a negative externality, where the act of someone is (an unpriced) bad for everyone, contributing to OSS is a positive externality, where the work of someone is (again unpriced) good for everyone. I propose creating distributed institutions for identifying and pricing those externalities (positive and negative), evaluating and rewarding (or pricing) them accordingly. What is the metric for externalities? Collective meaning and wellbeing of everyone.[1]
There are foundations like NLNet[2] that do this for OSS. I think we should donate to them in the meantime.
[1] More about this here: https://news.ycombinator.com/item?id=29043752
Crypto just reduces the hassle.
And what about when you want to spend that money? At some point you have to declare that income and be taxed and the government doesn't care if you were paid in euros, bitcoin or seashells.
As many problems as there are with GitHub (social lock-in, Copilot laundering Open Source license violations), GitHub Sponsors is incredibly valuable and substantially reduces the friction for getting support for an Open Source project.
He did not say which country, but my guess would be Germany. You have to declare every penny here.
That is, maintainer burnout rate can be high, but a continuous influx of new maintainers could compensate.
In fact, a lot of industries work like that, and numerous examples regularly appear on HN. Video game companies are notoriously full of super bright young developers, which are squeezed until burnout, just to be replaced by a new ones.
As for the compensation bit, although I agree with the principle, I tend to disagree that it would solve the problem as a whole. Sure, some maintainers would be more willing to spend time on their OSS projects if they were paid, some might even quit their job and work full time on their OSS. But my personal experience is that money is but a short term replacement for passion. Once you reach the point where you're not willingly working on a project, additional money can give you a boost of motivation, but ultimately you will tend to procrastinate to work on it, until you just won't be able to force yourself to touch it.
Wow this is such a great description of the arts, fashion, Hollywood, game development, startups (?), …
It's easy to forget, due to the vast wealth of genuinely great volunteer-driven projects out there, that an overwhelming share of open source contributions are actually funded by capital. It's hard to imagine the Linux kernel, or llvm, or even projects like React ever fizzling out due to maintainer burnout or disinterest.
This is fascinating, in my opinion, because our systems of accounting and economical analysis are by design terrible at keeping track of externalities. They are not good at analyzing the potentially holistic value of mutualist projects. Largely speaking whatever cannot fit on a balance sheet becomes the province of philosophy and culture.
Despite all that, in the modern culture of tech companies, the value of FLOSS seems to be understood and the companies that recognize it genuinely seem to out-compete those that don't. They recognize the value it creates for them, even if it's very difficult to estimate the costs that are avoided by using a FLOSS system that would otherwise need to be licensed or built. Even if the completely optional act of using one's own resources to contribute to open source software does not easily map to an equivalent or greater return in accounts receivable or company valuation.
What I mean by this is that the sustainability of "open source" broadly speaking is already demonstrated and I don't see that changing any time soon. What's not demonstrated however is the sustainability of OP's project. I think this is a challenge that they and their users should tackle without implicating the entirety of open source software as a culture, or attempting to impose any responsibilities upon users of open source software that don't already exist.
I find it especially funny when people complain "that they can't use the software 'at work'" when you release something as GPL.
How is that exploitation if you write some software for yourself, think it's useful for others, release it and then someone else uses it? If the software wasn't released, then others may have the same need and thus similar software may be written multiple times in a huge waste of time and effort.
To my mind, open source software is returning to the basics of human society - helping each other where we can. If you only ever do things with the intention of getting paid to do them, then you live a sad life.
> To my mind, open source software is returning to the basics of human society - helping each other where we can. If you only ever do things with the intention of getting paid to do them, then you live a sad life.
This honestly is horseshit because we wouldn't have this conversation if the profit of off open source was being distributed in any way fairly (or to put it patronisingly: used to help the contributors). By releasing your software under a license that permits this type of exploatation you are putting yourself in a precarious position of being at the mercy of whoever exploits it for commercial gain.
So as I am not misunderstood, I am not against open source in general or say GPL, but against using licenses like MIT/BSD by default.
But with non-commercial usage, I don't see it as exploiting the author as they wanted to write it for their own reasons and had no interest in commercialising it or believed it to be non-viable. If someone releases under MIT/BSD, then they're pretty much saying "here's the code, do whatever you want with it", so I don't see a problem with companies using it.
OS authors aren't broke. In fact the guys from Microsoft are some of the richest dudes in the world. Maybe OS authors that give away their stuff for free are broke?
A lot of people think open source is about not paying for software, even some open source devs think this.
You are allowed to sell your software even if its GPL'd or whatever. You are not obligated to give the source away for free, you can charge people for the source and for binaries, you just have to provide the source when someone does buy it. I would like to see more people try this out.
The most common counterargument here is that someone can pay for your source code, and then upload it to github, or even sell it themselves, allowing people to circumvent paying (you) for it. This is true, but most software needs to be maintained in order to be useful, and you are not obligated to provide the updates to people who haven't paid you, and I think you could charge for access to the updates, even for people who have purchased your software before.
You can also use trademarks, anyone who would attempt to resell it would not be able to use your "marks", which could be the project name or logo for example.
Maybe there are reasons nobody is doing this, maybe its not viable, but I haven't seen any cases where someone has at least tried this.
Some contributors will quit, but that doesn't imply that the system is unsustainable, as long as some more people will temporarily make some contributions, and we're not seeing an imminent collapse where people will stop scratching their itches and publishing the resulting code as open source.
https://kodare.net/2023/11/17/open_source_could_be_better.ht... (https://news.ycombinator.com/item?id=38302122)
I stole that thing from how medium does it. I think it's a good middle ground between absolutely no feedback and comments.I strongly believe that open-source contributors should be compensated for their work. So, in my opinion, there's a real need to establish an open-source economy where contributors can be fairly compensated for their work without compromising open-source principles.
I'm currently working on finding a compensation solution for OSS contributors. If you're interested in learning more, please don't hesitate to get in touch with me (see my profile).
Don't get me wrong though I love me a good open source project. I think the idea of being compensated might increase the quality, but let's point fingers at github. It would be as simple as adding a "support contributors" button on repos.
I don't know how divide the winnings fairly. But it should be possible.
(edit: actually IIRC it's fine to sometimes promote what you're doing on HN, so long as it's not Every Single Post) . Can you tell us a bit more?
Cost may be a component. But not the only component, and cost doesn't have to be binary. You can charge and still be less than the alternatives.
Companies don't make decisions on individual OSS dependencies. Individual engineers and engineering teams do. That's if anyone actually spares a thought at all (eek)!
It's not my money. Why would I particularly care if we have to pay for some software?
But it is my time and effort. If I have to go through a lot of red tape and politics just to use some library then ... screw that. Doesn't matter how much it costs.
That's one of the reasons it generally makes no sense to offer enterprise software for a low cost. Even if it only costs $1, your users still have the red tape to deal with. You may as well make it cost $1k, so you actually make some money from the few users who fight the tape.
Those that care, CI/CD only fetches from internal repos, and stuff is only uploaded into them after an audit.
This does not matter in the slightest; your only obligations to other entities are self-imposed. Using a developer's piece of code does not confer an additional right to that developer's time.
> The bad parts are that there’s a dozen issues that I haven’t even reviewed much less triaged, investigated, and fixed
If those 3/4 million downloaders really found it that useful, aren't there code contributions?
> time = passion + money
I didn't _start_ writing code because of money, and I'd still be writing code on my own even if I weren't being paid to do it. I know this because I code on my own without being paid to do it, only work on things that are interesting to me, and I don't have a problem dropping a project if I'm no longer interested in it.
> Those companies in that list are contributing to the success of those projects
Go work for one of those companies then? Seems like a win-win.
You have a new son and a partner that need a lot of you right now. Those PRs and bug reports can wait. The time you have all to yourself right now is extraordinarily rare. Use it for you.
That said, I do agree with the premise of the post.
You can use MIT/BSD/Apache/GPL3 dependencies without issue; never need to worry about derivative works or anything like that, just set it and forget it, and because corporations avoid you like the plague you can code away without a care in the world for backwards compatibility.
The BSD/MIT license is free as in free tech support, always has been.
“It’s too hard for dev teams to fund open source”: yes, it is; but we’re not going to fix that by giving them the code for free. If businesses really want something, they’ll pay. What’s missing are better ways for commercial entities-and dev teams therein- to monetarily support devs when they need to use a licence for commercial purposes because clearly the current parasitic economic model isn’t working.
Like, what, open-source police are going to arrest you for not addressing support questions?
MIT/BSD is "free as in the maintainer is free to ignore any and all tech support requests."
As a similar sort of person, now on my second child, with the first now 4yo, my advice is this;
Focus on yourself and your family. Get that rest, when the baby is sleeping, get some sleep, it gets harder before it gets easier.
Presumably paternity has finished now, and your partner (again I'm going to make an assumption) may be able to do much of the heavy lifted on maternity, but that'll end; at that point you'll both be working, you'll both have a massive sleep deficit, and in all likelihood (in my experience) you'll have a ~1yo who doesn't sleep through the night yet.
There were times where it got really difficult for me, and the passion projects became a source of depression because family life just didn't allow them to happen. Different people work differently, but for me, it took a while for me to realise that my priorities have changed, looking after and enjoying my life and my family became passion projects of a significantly higher priority than anything I build in software or hardware. The time will come (I assume, too early yet) when my children can take part in those with me.
The projects people are using to keep their companies running, on the back of your work, without paying for it, fuck them, let them be someone else's problem for a while.
This is because the core premise is simply wrong. Sites such as opensource.guide are not written for you. They're written for people profiting from you thinking they're written for you. If you follow their guidance, you'll end up unhappy and writing posts like that.
The only way to escape this is to ignore that con and instead pick a sustainable business model. That can mean not open-sourcing at all, but I'd wager that most of the time, it just means saying no, setting boundaries and knowing your worth + the value you provide.
Seems like a ridiculous statement (and definitely doesn't address the real issue of cyberbullying)...
_but_ when it comes to open-source, that's my exact attitude. Notifications off, users opening issues and pull requests? May as well not exist. I'll occasionally check but that's it.
The easiest answer is that deep down you need to accept that you can only be "tortured" by the incessant, needy, unfair demands of your users if you choose to. If you can't handle it, disable it, archive your project, walk away.
Open source entered a bubble because companies started requiring active open source contributions to staff+ engineering requirements. The model was never sustainable. A select few contribute to open source out of sheer passion, but the majority are/were using it as a career advancement mechanism or network effect.
The result is that major companies were able to advance their technology and profits rapidly off of people's free work. The result is better products, services, and tooling for everyone, but let's not kid ourselves about who benefits most from open source: large companies.
If you know that someone is going to take your code and make money off of it anyways (which is almost guaranteed whether you pick an FSF or OSI license), then you might as well make it so that society can benefit from this too. GPL/AGPL do this, MIT/BSD/etc do not.
And as an aside, I feel like it should be easier to profit from open source if you pick a license like AGPL. Companies that want to use it can pay you (the sole rights holder) for a commercial friendly license, while everyone else can use the free license. This is the same model for Qt/KDE.
You don't even need to sell support, and could probably even throw up a self service checkout page for commercial licenses. Thats minimal effort for maximizing profit lol.
Maybe in absolute terms but because of FOSS software I have been made free from the shackles of the windows ecosystem and liberated from the prison of Apple lock in to be free forever in the fields of the Linux and that alone has been invaluable to me.
Compensation according to value created is fair. It's a goal we should strive for, but it requires either not going through free markets, or it requires empowering market participants to generate fairer outcomes.
This is pure fantasy, for many reasons: But if there was a union of OSS developers that would negotiate corporate open source users contribution to open source, and that would have the power to withhold license rights, compensation would probably be fairer overall.
Essentially a code collective rather than decentral anything goes open source ala GPL. I wonder if that would be attractive to some open source developers?
I think there sometimes is a strong aversion to get organized. "I just want to code!". But the idea that we can do without actively organizing society is naive, and if the outcomes don't suite us it's because those that actively organize (e.g. corporations) hold the power to shape the world in their interests.
If you burn yourself out on serving your entitled users for free, that's on you, not on them. You've created that sense of entitlement. You're an adult capable of saying "no" or "pay me."
money = contributions * usage
A more accurate and sane formular would be
money = time_spent_on_project * avg_salaryOnce I've finished scratching my itch, I will only spend T hours per week on the project:
T = patron_weekly_income / 20
1 <= T <= 40
I have written and maintained dozens of open-source project [1], but I have only managed to make one sustainable open-source project [2]. It's also the only project for which I did all the things necessary to build a business (documentation, payment, invoicing, marketing, content, support, design, product management, tax fillings, legal, finance, etc).
I encourage everyone to open-source their code. But if you want to make a living out of it, you must also be an entrepreneur.
[1]: https://github.com/fzaninotto [2]: https://marmelab.com/blog/2023/11/13/open-source-profit-2.ht...
Thought this was a good quote to pull.
- 12800+ stars
- 1M downloads last month, 37k per day
- Paper about the project: 1500+ citations
- Used in 18k other repositories and 317 packages.
=> People use it.
But!
- 365 open issues
- 25 pull request
that hang for years
Only one sponsor.
And this is fine. People use the result of our work, but we do not feel that we are entitled / deserve / [some other vomit words] of more support.
In the beginning, we decided that we would do it:
- only for fan
- when and how we want it
- if some user is unhappy with our commitment or decisions
=> feel free to fork.
But!
We do enjoy when people thank us, create pull requests (we do review and merge them, although it could take time), or create feature requests or bug reports.
I can see open source as a great pet project that you do for fun and to improve your skills, but unless it is an OpenCore business or another setup where maintainers are financially compensated, all whiners and complainers can go and fuck themselves.
I would not recommend maintainers of the open source software even notice them.
If working on OSS is not fun - do not do it. Life is too short for unnecessary stress.
Radio stations in Sweden pay a fee to an organisation which distributes the money to artists in proportion to the amount of playtime.
Imagine a new type of open source license that mandated paying a membership fee to a global foundation to use the code commercially. Non-commersial use would still be free.
Companies would have to pay royalty to this organisation in proportion to their size or some other metric. The organisation would distribute the money to projects according to some usage criteria such a download count or similar.
For it to work there would have to be one or very few such organizations to that it is easy for the companies to handle. It should also not be very expensive for the companies. But even if it gets every company to contribute just a few thousands to open source it would still inject a lot more money into the system.
Even Stallman has suggested such a system [1], with the amount a given copyright owner gets for a work being proportional to the cube root of how much it is copied.
A common suggestion for the tax is a tax on internet access.
For entertainment such as movies and music and games that could probably work well. Probably also for closed source software. For open source software it might be too difficult to figure out how to allocate the money.
[1] https://www.gnu.org/philosophy/copyright-versus-community.en...
I am sure many companies realize that they gain value from open source. Thus they are willing to pay something. However they do not want to handle transactions with every transitive dependency they use. Just like a radio station doesnt want a contract with every artist. This is why isolated commersial licenses wouldnt work for anything but the very largest projecys.
That would by definition not be open source.
My ideal scenario is that code is open, can be improved and reused among commercial and non-commercial endeavors. I also would like that some of the value created by users of the open code flows back to the creators.
Individual commercial licenses do not create this ideal as they are monolithic and does not reflect that open source is a network of many dependencies.
Companies would be willing to pay for open source but they do not want to manage each node in their dependency graph individually. Thus the need for some centralized tax and redistribution system.
Publishing opensource to a package repository should have higher hurdles than picking a LICENSE file and pushing code to Github. Java/Maven repos typically have higher quality because to get code there, you have to have a domain, register it and get through the packaging requirements/bureaucracy. The use of packages from Github with merely a repo/git-hash is a bad trend. On the other side, if everything had a visible price/cost, then we might not end up with simple functions being overused and compromise a library ecosystem.
I choose an unusual hobby, that other people seems to find useful. Hence people should give me money.
I don't get it, I really don't.
If you enjoy working for free, by all means continue doing it. If you don't then stop. Nobody is forcing you. If you think your time and effort is worth more then you get now, go closed source, paid-support or whatever and see how much value you really bring to the market.
(I'm taking a -4 karma hit for this post, I know. I got it last time, do your worst)
Maybe it's a clout thing? That some people really want to make a living _selling_ software, but at the same time do not want to miss out on the cachet of "I'm doing open source"?
Whatever the reason, there is some amount of cognitive dissonance involved.
Now the "pressure from users" is an actual contractual obligation! I can't walk away. What a nightmare.
Or share the maintainer role with some other users. Just select one, two, or three people who seem to be responsible and give them all the necessary write access.
Or maybe configure the repo in a way that all PRs need at least two (or so) reviewers from a selected group of reviewers.
Then this would greatly reduce the effort on your side, and development should continue on its own.
Here in America I cannot work on OSS because if I tried to make that my main form of contribution, I would not survive due to healthcare costs, living costs etc. Companies being made to pay more or subsidize OSS I believe would have a detrimental effect on the system, because then it warps OSS to serve the desires of big companies in order for its contributors to survive.
The actual solution to this problem would be things like socialized healthcare detached from work. This isn't to say I think companies shouldn't pay for OSS (they absolutely should be funding and contributing to work that underpins their software) but rather that we need to reduce the reliance on corporate funding for people to be full time contributors.
I don't think this is true. You need an extra condition 'that few people want to produce'.
There is lots of good free art. Why? Because lots of people want to be artists and make art. There is tons of good free writing. Why? Because lots of people want to write. There is masses of good free music. Why? Because many, many people enjoy making music.
There aren't people who collect garbage, clean toilets, dig holes in the ground, or work in oil refineries for free. But there are people publishing science, doing research, writing philosophy, producing erotic material, designing things, putting on theatre, producing textbooks and teaching people things, making clothes, thinking of jokes, answering questions, providing peer support to addicts, playing music, making games, making animations, all without monetary compensation. This is because the people doing these things want to do them.
This isn't a failure of our economic system. It's a great thing - it makes the products better, the producers happier (provided they have the economic freedom to spend time on these projects) and the consumers better off.
First of all, it's obvious that in the vast majority of cases, writing free software falls into the 'amateur art' category not the 'dirty, boring and necessary job' category. Many, many people enjoy the time spent on writing and maintaining software, are motivated to solve their and other people's problems, and take pride in doing so well. You might expect that only games, intellectual toys or fanciful projects would motivate people to work on them in their free time. The reality is that software projects which could be seen as dry and boring to non-technical people (OS kernel design, file transfer protocols, laptop power management support, database and webserver stability, document rendering) attract many very talented people to work on them.
Secondly, if we think that there's some deep inequality or instability in our society because (for example) critical Internet infrastructure depends on hobbyists and volunteers, doesn't it make more sense to try and improve the conditions for hobbyists and volunteers, and make it possible for there to be more of them? The alternative put forward seems to be to turn them into more of the people who both don't enjoy the time spent on what they do, nor produce the best product that they can.
The existence of the path to monetization is entirely outside their control though. Millions of people make viral videos, very few have benefited from it. The financial system disincentivizes or outright bans open-product monetization.
Yet, since many users request features well outside my use cases, and I have only one sample of the many alarm models that use the same protocol, I thought I'd be successful in getting some help from the manufacturer, from lent hardware (at pessimist side) to some small monetary contribution (optimist side). I got no response at all...
People are also fast to point things that are "wrong" but don't want to submit a PR at all.
Software patents don't actually do this to a sufficient granularity (at least imao) , but get paid for; while floss does this to sufficient granularity that a compiler actually can implement it.
Basically we seem to have a collection of loose bits that variously ensures we have a public foundation for software development, and other loose bits that ensure people get paid for disclosing things.
The thing we don't have is one single cohesive system where people get paid to disclose the state of the art as a public foundation for software development.
You don't have to try turn every project into a business... There's also a pretty big catch-22 here in that most of these projects are only widely useful precisely because they are free. With that came the users, the plugins, bug fixes, and features. Take that away and many projects will have never became popular and then you'll have solved the issue of not being paid for your work (because no users will exist to buy your software.)
0. Don't give away something and expect tips, donations, or anything else because it will only lead to resentment. If you voluntarily give something away, you basically lose all control of how it's used in the real world. If someone intends to make money from code, then OSS is probably the wrong approach.
The challenges from the user side are:
1. Donate time, money, and effort based on utility and necessity to support what's vital.
2. Be a proactive self-starter whenever possible rather than depend on others for support.
3. Some maintainers and communities turn out to have an un{cool,professional} culture of hostility, uncooperativeness, or unhelpfulness. Don't waste time with that.
They're entitled to do that, and you're entitled to fork it.
> This all boils down to a situation where you have many profit-generating companies using software that some programmer volunteered to write. That software contributes to that company making even more money. But the developer sees none
Which developer? Because what I see going on matches what Stringer actually says earlier in the post: "There are lots of users, many in a corporate sense using my software to further progress their organization."
So you have at least two persons here; there is no "the" developer.
First, you have a company trying to make money. And then you have a developer trying to get money from their company (and not just that, but getting it, and generally trying to get more). Let's call this a type-1 developer. On top of that, you have a developer upstream writing the software in question that "contributes to that company making even more money". Let's call this a type-2 developer.
What we're neglecting to acknowledge here, and what most conversations like these fail to acknowledge , is that it's not merely "the company" that is benefiting from the work of the upstream developer. It's the type-1 developer, too.
If type-1 developers are effective at converting the labor from type-2 developers into personal enrichment, internal accolades from their employer, general career trajectory, &c, then type-1 developers really ought to acknowledge their culpability in the system that leaves type-2 developers undercompensated. This doesn't really happen, though. Most developers with a type-1 role wrt some money-earning scheme (i.e. the ones employed at a company) seem to treat their TCP, which is on average includes a salary alone that is well above the combined income of a typical household, as a sacrosanct natural right that should not be examined at this level.
My idea about what's currently in open-source domain is that it can be split into two categories: infrastructure projects that enable users to use their computers at the basic level, enable networking communication and maintain state-provided services land in "category A", while "category B" can be roughly described as interesting hobbies, leisure, entertainment etc.
What needs to happen, in my mind is that we need (international) organizations like eg. WHO to take care of the "category A" projects, where governments would have to allocate resources to finance and oversee these projects. Much in the same way how governments spend money on postal service or regulate / oversee banking. So, projects like Linux or OpenStack need to be under such international umbrella. This would require a bureaucratic process of examining such projects, estimating their usefulness, budgeting them etc.
On the other hand, non-essential projects, or projects that explicitly don't want government oversight / intervention could still work on the currently employed scheme: donations, sponsorship, volunteering.
In other words, I believe that some people working on open-source projects today ought to be paid. We just need a framework which establishes how much and how many such people can be employed for how long etc. It's prudent to make them (international) government employees to avoid playing into interests of sponsors who might not act in the interest of anyone but themselves.
For me, this would also solve a situation whereas employees of private companies are told / paid to work with open-source technology, but are powerless against the maintainers of such technology, while having no plausible alternatives. A lot of such projects succeed based solely on the good will of the open-source maintainers, but some fail due to the lack of, or deliver lower quality products. Having essential services covered by a government entity, and, by extension, being open to citizen complaints and wishes would make it possible to fight back against maintainers lacking the said good will.
But nobody ever thought to make a similar system to pay people for the value they add to the chain. On the internet it's possible, and many blockchain systems promise something similar. But big tech wants all the profits, from everyone's creations, all for itself. And governments are silently guilty standing in the background
of course, you are joking! look in the USENET news archives.. from the early 1990s.. you will see many proposals like "micropayments" .. later, even standards proposed. Also see the importance of "ad free Internet" and how certain govt-company groups try to add their own "ID" to people and machines.
I know the math above can be challenged from multiple aspects. But we need to start from somewhere.
The proliferation of tiny NPM packages is bad enough already. It will only be made worse by the cobra effect.
It notices the upstream dependencies and bills you accordingly.
There is no malice. FOSS is free and open. There are many softwares that chose to be paid ones and they are successful.
We’ve been appealing to our end users for funding to help keep the development of the mobile apps going, which you can contribute to, as well: https://opentransitsoftwarefoundation.org/donate/
Companies that depend on your hobby code can send you some maintainers if they aren’t happy with your hobby output progress rate.
There's user entitlement, but there's also lack of dev entitlement toward the license terms they themselves chose.
This should be a template response in GitHub issues by default:
> Remember that this project is:
> * PROVIDED “AS IS”
> * FREE OF CHARGE
> * WITHOUT WARRANTY OF ANY KIND
> * WITHOUT PROMISE OF FITNESS FOR YOUR USE CASE
> * WITHOUT LIMITS FOR YOU TO USE OR EDITLong story short: I recently put the marketplace down, because I lack business skills
If any of you is able to help with business side of the project, please reach out - mail in my profile
- Your company's perspective: Pay for the software to receive reliable products from trusted vendors
Business receives reliable open source software that is maintained on a daily basis, features are constantly developed and they receive a support from open source creators
- Why?
It's cheaper to pay a few (dozens / hundreds / thousands) bucks each year for a library, tool, component, etc instead of build it on your own, maintain, test, take care of security updates, do customer support
- Yours (open source developer's) perspective: Get paid to build, maintain and support your software
You as a developer make money by selling licenses to your open source software
- Why not free open source?
Because you still need a day job and do your open source after hours. Thousands (millions?) of companies make millions (billions?) dollars each year by incorporating your software in their products or using your software in their ecosystem, but you are left with $0 profit from it
- But I want my open source to be available to regular people for free
With poss market license it remains free for non-profit/personal/scientific users, if you choose to. It's just MIT License slightly modified
Yes there's edge cases, but at the moment there isn't really a good way to spread my money around.
Python 3.11.6 (main, Oct 3 2023, 00:00:00) [GCC 13.2.1 20230728 (Red Hat 13.2.1-1)] on linux
Type "help", "copyright", "credits" or "license()" for more information.
>>> population_base = [100_000_000, 1_000_000_000, 8_000_000_000]
>>> for population in population_base:
... print('15s/day for',population, 'people equals:', round((15*population)/3600), 'man-hours/day')
...
...
15s/day for 100000000 people equals: 416667 man-hours/day
15s/day for 1000000000 people equals: 4166667 man-hours/day
15s/day for 8000000000 people equals: 33333333 man-hours/dayPeople pay for software, notably Office, IntelliJ, OS, Adobe, video games, enterprise s/w and many more. For normal consumer products things are simple you pay some money for product and can have simple expectations that s/w will work. Every once a decade the direction a s/w takes major change, and you have to decide if you want to stay with it or move to something else.
Now with OSS both sides expectations have become implicit and ambiguous. In letter they come with no warranty but in spirit it does. No sane person would depend on database/compiler which can stop working or won't fix bugs without recourse. It may be OSS but I don't have time/skills to fix myself, so implicit understanding is that community at large will fix it.
Next is the issue of payment to the community. how much should be the payment, who should be paying and should be paid? No payment was discussed up-front, but it is expected implicitly. Should I review all s/w I use to understand the payment I should have been making. Maybe I should stick to only commercial s/w as expectation of payment is clear or maybe s/w being free but created by large corporates are ok as no payment is expected. Is making s/w in NodeJS / Python risky, as I don't know if community is paying for all the libraries I am using and how much I should be paying?
That would absolutely get people off the fence and to start agreeing.
It's a bit annoying as from an accounting point of view I'd rather they just invoiced me for a service.
Focus on your passion.
The success of OSS projects is a function of how many people are passionate about it.
> It is quite literally a free lunch at the expense of hard-working individuals.
Yes! This. That's the point.
I'm a Free software developer and that's why I give my code away: to solve someone's problems with software. I can do that. I'm not particularly good at most things, but when it comes to writing software I'm very capable. In other words, I am a "free lunch" generator.
To me the entire point of computers and software (where these intersect with economics) is to change the structure of the economic system itself to a more humane system (that works in harmony with the global ecology, but that's a tangent to my main point today.) "Let the robots do the work and we'll take their pay."
Science and technology have won the day. We have the knowledge and resources to take care of everybody on Earth without "disadvantaging anybody" (as Bucky Fuller liked to say.) You don't have to "earn a living" anymore. We won history.
Software is just the clearest, cleanest example of the general phenomenon of science and technology obviating the physical bounds that held us back until now.
Getting back to software development: technology should be deployed so that folks only have to work a few hours a week to pay for their living expenses (e.g. $300 per month should cover food, housing, clothing, health care, etc., and over time that number should decrease!)
In that milieu people can develop software and give it away without the extra constraints of having to make it profitable in and of itself.
This is the whole point of technology in general and computers and software in particular: change the structure of the economy so that we can all "live happily ever after".
(And get on with space exploration and mitigating climate change.)
Maybe you can find a second job, and focus on creating software as a hobby.
You can argue that Linux can't reach that kind of successful if it is nonfree, but there were a lot of poorly written OS that sold millions.
Each culture is different, but many people now see 'free' and don't think of how that product is getting created.
For Open Source the problem is even worse. As a project becomes more popular there are more demands on a maintainer's time, which is in short supply for most good senior devs. They have a demanding job, they have a wife, they have children, they have other family, they have friends, they have their own self-learning. This doesn't count the lucky ones whose employers pay them to work on Open Source software. Once a FOSS project gets to a certain adoption level it becomes a supply & demand problem where the demand quickly outstrips the supply.
How do we fix this? Well, we can either reduce the demand (PRs, issues, etc on a single project) or increase the supply (hours of dev time).
One way of reducing the demand is to have many projects that do similar small things, splitting users across the ecosystem. The NodeJS NPM system is like that. I think that leads to problems with reliability, technology selection choices, and vulnerability finding/fixing. Then again, NodeJS is still going strong.
Increasing the supply from one person is hard. For most the supply spent on personal relationships (family, friends) should not change, except to go up. That leaves the supply spent on work. This is a hard problem too. Most devs work a minimum of a 40 hour week. The only possibilities are going freelance, so you can choose your own hours if you can financially justify it (many freelancers work more than a 40 hr week though), or going part-time (which is not an option for many). Let's say a senior engineer makes $160k salary (not looking at options, etc in this math). For him to go freelance he would need to cover family health insurance and life insurance, say $2000 per month, $24k total [1]. Taxes will be around $60k [2]. These are both much more than he'd pay working for a company. To get the same spending power freelance as he did at a company, he's going to need to make about 40% more (16% for FICA taxes, 24% for insurance), or around $200k. That's not doable from donations, for any other than a select few maybe.
What about around half? He can make $100k from freelancing, $104k from Open Source donations. For Patreon the average backer pledges $6 per month [3], or $1.50 per week. You'd also need to figure out what pledger rewards were. You would need around 1800 backers to do that. As a project rises in popularity you will also have an opportunity to get a co-maintainer or two who helps oversee PRs, etc. The pledges should be split among the co-maintainers, which means more pledges needed. The co-maintainers can also start to have some of their freelancing be consulting on using the project. JBoss is a great example of that, so are a number of other projects. When you get to enough consulting on it, with 2-4 other maintainers, you might save each other costs by incorporating as an S-Corp.
Ok, so if that works, why isn't everyone doing it?
There are a number of reasons.
First, security & stability. Some people need the stability, structure, and security that comes from working in a company. Freelancing is a lot of work, across a diverse set of skills, and there are no guarantees of a next paycheck. You might make $10k one month and $1k the next (an outlier, but it happens). So you have to be very budgeted, with a nice safety net in your account for dry spells.
Next, skill set & capabilities. Some people just are not good at marketing, or working with people. I think you cannot be successful in freelancing without having some skill at marketing and at working with people. I also think you can't really be successful as a software engineer without those skills either,but that's a different topic.
Finally, risk. I touched on this in the first point above, but it's worth touching on again. Freelancing is high risk. Your work could hit a dry spell. You could get sued (you did get liability insurance right?). Some people, or their family, find it unhealthy to have to deal with that level of risk.
Are there other solutions? Probably. Better ones? Possibly. Dev collectives, bounties, FOSS organization grants, FOSS organization patronage (Google Summer of Code is an example) are all other ideas that come to mind.
As always, I am a dev - I am neither a lawyer, nor an accountant, so do not take the above as legal or financial advice.
1. https://www.ramseysolutions.com/insurance/how-much-does-heal....
2. https://www.quora.com/If-a-professional-freelancer-were-to-m...
3. https://www.crowdcrux.com/patreon-statistics-and-demographic....
I was very inspired by "Walkaway" by Cory Doctorow. It involves a world where people detach themselves from "Default" (global capitalist society) by living in abandoned towns, building the tooling they need to re-establish a modern quality of life. He pulled a great deal on the open source movement in his speculations of how this might look. What I didn't realize at the time is it is essentially an anarchist proposition of community self-reliance. In the novel, there's no point in seeking compensation for your work, because your basic needs are already met by a share-and-share-alike society, and therefore everything you do fulfills either a very clear personal or community need (building a tractor, a house, software to manage a farm, a public spa, or repair schedules), or, is purely for pleasure.
My friend that's sticking it out in the USA is doing the "correct" path for an engineer: First gig in our hometown, transitioned to NYC, did a 4 year tour there saving a couple hundred K, house upstate, still working and saving for retirement but also farming ants and doing his other odd projects for his pleasure. In a recent conversation he mentioned frustration at the poor retirement opportunities for most of our generation. You gamble your life saving's on the stock market, or, do something that doesn't really benefit the world like flipping properties, if you can afford it. If your interest is self fulfillment, community fulfillment, and financial fulfillment as we grow older, that doesn't really exist, at least not in any combination that we've been able to figure out. He's thinking about some kind of ethical business venture, or maybe just a fun thing he can kick off that he can hire his non-engineering friends into like a cute little sandwich shop or something, but as he enters that world he's realizing all his competition is hyper-capitalized businesses or people that he can't possibly compete with if he doesn't do the same shit they're doing, such as filling their kitchens with undocumented immigrants. Basically, if he wants to do good and get paid for it, the opportunities just don't seem to be there.
So therefore, long term what I want to strive for as I build out open source software is actions that "break us out of the box." I and my like minded friends don't really think capitalism is going to cut it in terms of actually rewarding with money our efforts to do good in the world; after all, an investment banker makes more than a teacher and firefighter combined, and does functionally nothing useful. Therefore I'm interested in building things that free people from a financial burden. Every little financial burden I can free people from is a success. I love when I read stories about people building out little GPT programs that can automatically negotiate parking bills or whatever. Or scan your email to automatically apply for rebates and coupons. Or, outside of actual coding, helping people find out that their library has a streaming app they can use for free, so they don't have to pay netflix anymore. Or helping people turn their lawns into gardens, to reduce their food bills.
From a software standpoint, the "Awesome Selfhosted" project is very inspiring in this vein: https://github.com/awesome-selfhosted/awesome-selfhosted Lots of tools that many people pay for that you can instead deploy on your own and use for free in a way you control.
That's the kind of open source software I want to build. Trying to get paid to build FOSS is a distraction, instead I want to build things that will help people not have to pay for other stuff. I make plenty of money in my day job, I'm good on that front. Pipe dream, we do enough of it, and the question of "getting paid" becomes moot.
Even in this thread we can see the toxicity around trying to get paid to do good in the world. "What, you did something good for free and expected to get paid for it, what are you, an entitled moron?" vs "FOSS software is the backbone of your organization, if you don't pay for it, you are evil." I think it's just so frustrating how twisted up things are. Normally you get paid to do engineering, but if you build something genuinely useful for people, now you don't get paid, but somehow you're an asshole for asking to get paid, and also you're naive for thinking any organization would ever pay for something they don't have to pay for... what a mess. What a rats nest of competing values and accusations. It's not worth trying to fit what imo often inspires FOSS, namely "fixing a problem in our society," into the capitalist mode of thinking, "and getting rewarded for it." If you want to fix things in society, build FOSS, if you want to get rich, take contracts, charge your clients 200/hr for engineering time, put Indian engineers on it, and pay them 10/hr. There, now you're rich. So what?
Edit: on reflection I think basically our efforts to do good for people by writing foss will simply be exploited for profit by corporations, so better to walk into this understanding that, and for those that aren't happy with that state of affairs, try writing foss that can help other people escape these kinds of exploitation.
That's economics 101 and not the problem of a particular profession
I'm sure the OP can competently explain all of these things. What they can't do is justify them, or reconcile them with the principles that their society has instilled in them.
If you want to get paid for your work, why are you choosing to give it away for free? I'm not judging, I'm genuinely curious. Are there any open source boosters left, and if so how would they make the case for open source at this point?
Maybe what's needed is a "for-profit open source" license, where the code is free for anyone to read, but it's illegal to use for commercial purposes without paying. I'm guessing this approach is already being explored?
It’s pure ego, but I hope it benefits the world too :)
One solution is to announce that you're abandoning the project and suggest that its users make a plan to fork it / take over maintenance. Instead of adding a social norm that open source devs should get paid, we could drop the norm that open source devs should feel obligated to maintain projects for free. Maybe every README could have info about the primary maintainers and how enthusiastic they think they're going to be about the project going forward, so people can make informed technology choices. That way no one complains about a bait and switch.
Another idea is for the dev to respond to issues on Github by saying things like "I can fix this if you pay $X"
Benefits:
- recruit high quality developers
- up-skill existing developers
- devs will sometimes fix things that the company is using
Nonfree and proprietary licenses have always been an option. The problem is developers want the convenience that FOSS culture provides without actually making the sacrifice of putting end-user rights before their own profit. Coding (specifically web dev) has turned into a money train and FOSS devs want on board.
FOSS licensing is an ethical stance. That ownership of software, (nonfree) copyright and intellectual property is fundamentally immoral, and authorship doesn't grant you privilege over the code you write. And that stance has a cost. If developers want to get paid, get a job writing code and a paycheck, or use a nonfree license. Otherwise, stop being hypocritical. You chose this path.
I consider the way I publish MIT projects to be mostly in good faith. I put it out there because I like having my name associated with useful implementations of things. If my code, worked on with my own hands, is spread via unattributed copy-pasting? Good. Even if I don't know about it, somewhere out there my code is powering something.
I started this mindset when I was a new grad, and to achieve a task I would first look at a bunch of different open-source projects to get a sense of how they were approaching the problem domain.