Paying maintainers isn’t a magic bullet | Hacker News Reader

Paying maintainers isn’t a magic bullet

blog.hansenpartnership.com

63 pointspabs342 comments

As someone trying to make a living out of my OSS work, this kind of clickbait titles really get on my nerves. Just yesterday, like most week a company did contact me to get support but when sent a payment link and instead of ignoring me took the path of questioning why would they pay for OSS. This kind of behaviour is so common, I have a canned answer for all those companies oss attracts trying to get things for free:

OSS doesn't mean free labour and I value my time. FYI I would have made better money flipping burgers than working on this and I can't realistically extend that principle to help everyone on the internet with a rate that's below minimum wage where I live.

The author did say that open source maintainers should be paid and was harsh about exploitive corporations. They were simply stating that we should not expect the maintenance problem to be solved by paying developers. Whether that's true or not is debatable. Historically, some have certainly made a living off of developing OSS since their work was valued enough for someone to hire them or because there have been people willing to pay for their services. On the other hand, many more have not been given that opportunity. Perhaps paying them will solve the maintenance problem, or perhaps it will only reveal other underlying hurdles with open source development.

> was harsh about exploitive corporations

taken from the article: "it’s not a problem to be solved, it’s a consequence to be embraced and exploited (if you’re clever enough)"

another interesting quote: "Microsoft and other companies railed about Open Source being a cancer on intellectual property: because it is"

none of the points made by the author are fundamental truth and most of his actual points are in direct opposition to what I do in my own OSS bubble everyday

> another interesting quote: "Microsoft and other companies railed about Open Source being a cancer on intellectual property: because it is"

well, that's wrong because Microsoft and other companies were worried about free (and therefore, open source) software.

IMO, open source (without any mention of the ideological aspect of liberty and freedom) is/was their remedy for this

Read the sentence in its entirety, dropping the parenthetical comment:

> If you follow the above, you’ll see the supposed “free rider” problem is simply a natural consequence of open source being free as in beer: it’s not a problem to be solved, it’s a consequence to be embraced and exploited.

"to be embraced and exploiting" is referring to the consequence, not to the open source developers.

What the author is trying to say is that a large segment of the users of open source software are using it because it is free as in beer. That sounds like a problem, and it is if you are narrow-mindedly thinking about how to monetize the open source software directly. But there are other ways for open source developers to take advantage of the opportunity.

That is how I read the bit about exploitation as well.

> That sounds like a problem, and it is if you are narrow-mindedly thinking about how to monetize the open source software directly.

That may be true, but there is another thing to consider: earning a living by running a business is challenging. Many people do not earn enough money for it to be worth while. Most will have to put up with other businesses trying to freeload, or at the very least cut an unreasonable deal. It is easy to claim the difficulties with open source exist because the software is free (as in beer). Indeed, it is probably more difficult to make money because of that. On the other hand, it is also a difficulty faced by all businesses.

OSS is business as any other, there is no need to take it personally. Be happy you discovered toxic customer before even signing contract.

> instead of ignoring me took the path of questioning why would they pay for OSS.

Sorry, I thought you wanted support, not a therapist... Here's a local listing of professionals nearby.

Corporations use OSS for avoidance of paying tax and labour. There is plenty of activist, idealist developers that can be tricked into doing work corporations need but without receiving remuneration. That's why OSS is being championed by big corporations as it saves them a ton of money that they otherwise would have to spend on R&D etc. With OSS they can cherry pick ripe OSS projects, appropriate them and use to make millions or even billions and original developers won't see a penny. You know you may get the so called "exposure", but try to pay your rent with it.

OSS really needs to be regulated, so that for profit businesses pay OSS contributors living wages at very least, if they use their projects.

The hard truth is that these "for profit businesses" wouldn't use many of the projects if they weren't free and open source in the first place. This is why the moral argument falls on deaf ears.

Relicensing is an increasingly popular choice for OSS devs to to try to monetize usage, and it works since it is oftentimes cheaper to just pay than to rewrite software.

> The hard truth is that these "for profit businesses" wouldn't use many of the projects if they weren't free and open source in the first place. This is why the moral argument falls on deaf ears.

There is no evidence for that. If something is good for business and cheaper to buy than develop in-house, business will buy.

> Relicensing is an increasingly popular choice for OSS devs to to try to monetize usage, and it works since it is oftentimes cheaper to just pay than to rewrite software.

This is great to see that more and more projects offer dual licensing, but this really should be regulated, so that developers won't be bullied into giving up fruits of their work for free. This would also create the level playing field where developers from disadvantaged background could participate in OSS knowing they'll be able to put food on their table if they commit time and their skill into it. Currently, OSS is really a domain of people with privilege, who have the luxury of being able to spend time doing unpaid work. They then often get advantage when being hired, by having OSS projects in their CV, whereas someone who is struggling financially can't afford to work on OSS. Many employers look for OSS contributions and that gives people with privilege an edge.

Relicensing is a big problem is the OSS world too. To work at all, it requires copyright attribution. For small projects this really hampers collaboration. But if you don't do it early it might become impossible to do later.

That would not level the field. Someone who is disadvantaged would not be able to dedicate as much time to writing software as someone who is not. So while someone who is disadvantaged may be able to dedicate, let's say, two hours a day and get paid for it, someone else may be able to get paid for two hours and then do it some more for two more hours for free.

The whole point of level playing field is that they must be remunerated, so there is no instance of someone doing something for free.

We had similar problem in my country with apprenticeships, that used to be unpaid. The good ones where snapped up white middle class children, while children from ethnic minorities had to work for money, not necessarily in their field of interest. That amplified the lack of diversity in the space of white collar jobs. Now unpaid apprenticeships are illegal and that has vastly improved the access to jobs for people from disadvantaged backgrounds.

It's exactly the same situation with OSS.

> If something is good for business and cheaper to buy than develop in-house, business will buy.

Sure, but "free" is frictionless for most developers at any business while "cheap" would still incur going through the purchasing process (at bigger corporations and/or governments) or asking your manager to pull out the company creditcard. From my experience with bigger orgs, it would be MUCH cheaper for the organization to write most dependencies in-house than to go through the process of buying it off the shelf. Even if the OSS dependency would only cost 1 USD for lifetime use or something.

This goes for most small-ish dependencies like test runners, API clients and even simple ORMs. Very small things like left-pad or faker.js would be much simpler to write in-house.

> Sure, but "free" is frictionless for most developers at any business while "cheap" would still incur going through the purchasing process (at bigger corporations and/or governments) or asking your manager to pull out the company creditcard.

This is really a non issue. Developers can evaluate whether the product meets needs and then get whoever is responsible to pay. In every company I worked, when I needed a commercial product, I put my case forward and most of the time got the company card to pay without issue.

I simply can't see how this is a valid excuse. No amount of friction can justify people not being paid for their work.

After all we can say why pay workers at all if there need to go through payroll, it's such a hassle...

I don't know how that would work. I don't think I would get a $1/mo VPS if the provider had to transfer his costs for all the OSS projects to the customer, me. And I like cheap cloud services.

I just thought of another possible explanation to give them. "All software has a marginal cost of $0, so all effort is in improvements and that still needs to be paid for."

The article contains a number of problematic statements. One is for instance the statement that Linux destroyed value. It may have for the proprietary UNIX vendors. But this should be balanced by the value Linux and OSS has created overall. Cloud Computing would never have scaled the way it has, if there would be operating system licenses and cloud computing providers would not be able to modify the OS according to their needs. I doubt that the profits of the proprietary UNIX firms over their lifetime are larger to what Amazon Web Services makes in a month.

OSS is basically exploiting a network effect that aggregates a lot of smaller contributions to a large one. Learning how to use a specific open-source software is the contribution the corporate free riders make.

The question is, how can the maintenance work be decentralized and distributed. Kubernetes changes the release manager for every version, so distributes the work over time. As long we rely on the single maintainer the results will not satisfy expectations. BTW a commercial company providing the support needs to solve the same problem.

Linux and open source removes the capacity for people to compete against it commercially. It takes air out of the room. Why would we pay you when we can get this for free?

free as in 0 cost is very difficult to compete against unless you are 100, 1000x times better.

There is the value that would have been provided (generated) in revenue to UNIX manufacturers had Linux not existed and the value that exists with Linux being free. These are two numbers.

One is a tangible financial number, the other is another tangible financial number from profits not needing to pay for a UNIX manufacturer.

It's difficult to say what would have happened, but you cannot deny that prices for corporate UNIX DID in fact exist, and they were not getting paid due to the availability of the free option.

Also, remember that the UNIX manufactuerers would have produced value too, so you need to include the value produced by UNIX versus the value produced by Linux minus licence costs.

If you're using the argument that if everything was free then there would be infinitely more value than things being commercial, then why isn't everything free, if that truly produces more value?

I would expect everything to be free if that was truly the better option.

>It's difficult to say what would have happened, but you cannot deny that prices for corporate UNIX DID in fact exist, and they were not getting paid due to the availability of the free option.

Yes, this is the problem of "the seen and the unseen" in economics. Prices for corporate Unix did exist, and they are not getting paid now. That is the seen. What is unseen is the value that was created by having a cheap (Linux isn't free --- you still need to pay people to set it up, patch it, provide support, etc) operating system that was available to anyone with an Internet connection. Would Google exist without Linux? Absolutely not. Would the iPod exist without Linux? No. What about all the value created by smartphones?

When balancing all of those against the profits of proprietary Unix vendors, I feel comfortable in saying that the existence of Linux is a giant net positive for the world, even if it did result in the proprietary Unix vendors going out of business.

Yes, I accept all the value that was generated by Linux is tangible and it's extreme.

I'm thinking from the principle of work: if you do some work for someone and it is not reimbursed, then it's obviously good for the people receiving your work but not necessarily the people working.

It's not the revenue from corporate UNIX that I'm thinking of, it's the being reimbursed for your work problem. The potential value produced by the non-free thing was destroyed.

Earning money is hard enough as it is, but today everyone expects everyone else to work for them for free, with nothing in return. If people stopped working on open source for free, would the value still be created?

To reiterate my point in my comment: if free things produce more value than commercial things, why isn't everything free?

> Would Google exist without Linux? Absolutely not.

What are you saying? That Google as it exists now would not exist? That's probably correct. That no one else would have created an equivalent search engine? That's much less obvious.

> Would the iPod exist without Linux? No.

Why do you need a FOSS OS to make a portable digital audio player with a hard drive?

> What about all the value created by smartphones?

Why do you need a FOSS OS to make cell phones with integrated computers?

Since when does iPod have anything to do with Linux?!?

The statement about Linux "destroying value" reminded me of Bastiat's famous satire, The Candlemaker's Petition [1]. Saying that Linux is a net negative value because it reduced demand for proprietary Unixes is like arguing that the sun is a net negative value, because it reduces demands for light bulbs.

[1] https://fee.org/articles/the-candlemakers-petition/

Sun's motto was the network is the computer for a reason.

Cloud computing is basically timesharing rebranded.

Even containers were already a thing in big iron UNIX, e.g HP-UX vaults introduced in 1998 thereabouts.

The way modern economists analyze this, any transaction creates value for both the buyer and the seller.

If owning a Thingie is worth $60 for me and $20 for you, we both gain $20 when I buy yours, and the world is $40 richer. You can think of all wealth as created by trade from this perspective.

In the Linux case, it destroyed value for proprietary UNIX vendors, and created value for millions of UNIX users.

Nobody claimed that. Paying maintainer is a start.

Ideally any "important" project would have organization behind it, with some people ready to react when security problem hits (even if usual maintainer/developer takes well-deserved vacation) but the whole problem is convincing corpo it's worth investing outside of disasters like OpenSSL which was essentially paid to make code worse.

I'm always interested in discussions about this topic on HN. I will toss this out there in hopes of fostering discussion of how to actually solve this problem:

I think we would do well to develop some systems for helping people monetize their thing if it gets popular. But we could also work on the cultural side of this and help people understand that you need to think about how much you are willing to give and that if your thing gets popular, making money with it is an additional job to do, not an entitlement.

https://news.ycombinator.com/item?id=17824166

I say that as someone who does a lot of stuff "for free" and hopes it will pay off because I'm handicapped and can't work on a schedule and can't do a normal job and blah blah blah. I keep hoping that if I deliver value, someday that will somehow lead to sufficient money for my needs.

So I am not attacking people who feel they add value and should get money for it. I'm on their side.

But the reality seems to be that extracting money from people is additional work on top of providing them with something of value, like it or not. I really wish we would get better about finding ways to uncouple certain things and help people who provide value to live comfortably. I recently left a comment on HN suggesting a lot of really aggravating products that cause problems are likely due to the fact "people gotta eat."

https://news.ycombinator.com/item?id=33827984

I would really love to live in a world where people doing genuinely valuable things and not great at selling get comfortable lives and people great at selling and happy to make the world worse by it would get actively encouraged to find a better way to live their lives. I just don't know how to get there from here.

Obviously I’m biased because I have skin in the game but articles like this really do a disservice for getting maintainers paid, something the author purports to support, because it gives people yet another excuse not to fund maintainers and maintain the status quo. I have yet to see illustrative examples in posts like these that show well funded projects doing worse because they got money.

I mean, it's the downfall of all facile answers, isn't it? "If you aren't the customer, you're the product" falls down when closed-source companies turn around and sell their smaller customers to their bigger customers, and the idea that you can't dictate terms to someone you're not paying falls down when someone debunks the contrapositive by saying no to a paying client. There's no silver bullet to make software always do what you want, so why should there be a silver bullet to always make humans always comply?

Using the argument that "X is not a magic bullet" is a pretty sure sign you're not worth listening to.

Nothing is a magic bullet, and we all know that.

> there’s really no proprietary UNIX maker left

Except for Apple, HP, Oracle and IBM.

https://en.wikipedia.org/wiki/Single_UNIX_Specification#Curr...

Oracle is not in the list because they just decided to stop paying for certification, other than that Solaris should still be compliant.

So much hate for this article in the comments. Are we really such a one sided community?

This article makes some great points which hurt to hear but should be addressed if we don't like them. Economically, it is true that paying and maintainer who is already trying to spend lots of his time on open source isn't going to make him spend more time on it. You might still be a jerk if you don't pay him, but we should address the fact that this is true to some extent.

It's also a great point that helping open source maintainers maintain their product can sometimes be better than payment to achieve your goals. What open source maintainer would turn down the help of some intern willing to write all the unit tests?

Whenever someone says they are for open source but then cry when someone makes money off of their open sourced code all I see is some greedy mf crying he didn’t figure out how to make money out of his project.

Almost always your open source project is nothing special when it is used by “everyone” you just happened to either be first or by pure luck were picked up by some other project that made your famous. That is the beauty of open source it doesn’t live or die based one one individual. If you don’t like the project anymore do something else if you want to make money then you need to start with that in mind. Trying to pivot to for profit after decade of free software is pure bait and switch.

It sure isn't, but how about we do it anyway, and then take it from there, because it's the right thing to do?

This is a good article, thank you for writing it.

I want good things to spring up and come up and exist, so I am happy paying for things. If I wasn't, those things shall eventually go away.

Free things take regular effort to maintain, if everybody that produced something that was free ceased work on it, it would too become less valuable with time and finally it would shut.

We have a societal cultural attitude problem that people expect others to do all their work for them and produce things for them with nothing in return.

This post sows the seeds of its own best counterargument: time is money and maintainer money means maintainer time.

There's no sharp either-or between just a hobby and full-time job. There are some inflection points, as where money becomes enough to take one fewer contract client, devote a day or afternoon per week, or leave a current full-time job. Overall, paying a maintainer who has to manage money enables them to spend more time. That addresses the fundamental unbalance between demand and capacity that risks burnout.

I think of burnout risk as having another key factor that isn't mentioned here: psychology. It isn't just demands versus wall-clock capacity. It's your perception of demands versus the focus you can bring in time allotted. Paying someone for their work is a very direct, accountable, and socially recognized way of confirming the value of their effort. Not feeling exploited changes how maintainers relate to bugs, patches, planning, and support.

As a maintainer of a slightly popular project I totally agree with the fact that paying maintainers would not fix much.

First of all, the project I maintain would likely lose 99% of users if we charged money. Being free (as in beer) is a large part of its appeal. And that's okay.

Secondly, the project is currently being maintained by a few people in their spare time. If you paid me, I wouldn't have any more free time, so to make it worthwhile, you would have to pay me enough so I would give up my job. And it just sounds unlikely that someone would pay so much for our project.

I really don't think Open Source needs fixing with regard to money. I like that Open Source is not about money. Not every thing has to be profitable.

> Open Source as a Destroyer of Market Value

New personal mission statement: destroy as much market value as possible.

What the author thinks is that paying him is not a magic bullet, and so must be true for everyone else.

Would you say "X is not a silver bullet" is a silver bullet for losing attention of clued-in audience ? :D

Reply on news.ycombinator.com