It would probably be more sustainable if the companies that depend on 100s of FOSS OS'/libraries/applications/etc to generate billions of dollars in profit would contribute more significantly.
Individuals have agency over discretionary spending, in other words you can wake up and decide to donate money, hire a OSS dev, and so on.
Because you personally have that agency, and because companies have lots of money, you project that agency onto companies.
But that's not how companies work.
People at companies have very little personal agency when it comes to spending money. Spending has to be approved, justified, and some value has to be received.
FOSS basically is (mostly) incompatible with this model. Some companies do pay staff to work on OSS but it's rare (and exclusively tech companies with a motivation.)
So while your statement is completely true, it's also not possible.
It's important to recognize that donations and corruption are indistinguishable, and company finances tend to be set up to avoid corruption.
Other than the humor in this sentence, I'm not sure why it would be limited to donations. They can hire devs to work on FOSS.
> People at companies have very little personal agency when it comes to spending money. Spending has to be approved, justified, and some value has to be received.
Approved by... people.
These FOSS we personally use. And FOSS we don't. Most of the FOSS we use already has plenty of paid devs. Think Linux, Firefox etc.
But what about small projects? How does that hiring conversation go? How do I explain my hiring request to my supervisor? Why am I spending 100k a year on a text editor? Why have I mandated we use my cousin's text editor and we're paying him to write it? When there hasn't been a significant update in 2 years? When we had to layoff staff to make mandated cost cutting?
Wow. What absurdly unfair examples.
In other words how do I choose which projects to support? How do I detect when said support is above board or when there are other factors?
You may take issue with the specific example but "hiring OSS devs", or worse making OSS donations, look exactly like this (if not as blatant.)
As a business owner How do I know these decisions are made in good faith? As a shareholder how much of this am I prepared to tolerate?
Those businesses making "billions in profit" have very high levels of accountability. Lots of people care deeply about that money.
Same argument can be applied to closed-source software. For example, how do you decide which vendor is your company going to select for their internal or external business. Why did you choose Cisco security solution over Fortinet or whatever other alternative there is? Is it corruption in terms of "I know this guy in Cisco" or is it "their solution is the most powerful"? Essentially, as you say, it is indistinguishable. I don't see how this is any different than to select some FOSS project to support if it generates your company/product a value.
Reason why companies aren't paying for FOSS is simply because they don't have to. And also there's no business on the other side to buy something back from you which is many times the case with B2B deals.
Selecting software (commercial or OSS) is no problem. And OSS bring free is a plus in that column.
Paying for commercial software is easy. They give us an invoice, and we pay it. OSS has donated button on a web site. There is absolutely no motivation to click it, and indeed clicking it leads to all kinds of extra hassle and work.
OSS has no business model. That makes it hard for companies that are used to business.
Yup. All this other stuff is just noise: this is the real reason.
If I ran an open source project that businesses used, I could make something that looked exactly like an invoice for a piece of commercial software, but say it's optional to pay, and companies would -- completely correctly -- not pay it.
> As a business owner How do I know these decisions are made in good faith? As a shareholder how much of this am I prepared to tolerate?
Hypotheticals that I'm not going to bother trying to answer, because that's going to depend on the business, the business owners, the projects they choose to support, and a myriad of other factors. Good luck getting some generic answer to such questions.
> In other words how do I choose which projects to support?
You're right. Impossible to decide, so let's not even try.
Or, like, do some research and have a discussion to determine which of the various FOSS your company uses that could do with some funding. Evaluate them individually. Decide on a monetary limit. Re-evaluate regularly. You know, normal business accountability things, right?
No, not at all. A normal business accountability thing would involve not paying for something when you don't have to. Or, rather, not paying when paying doesn't grant additional value beyond what not paying gives you.
For the most part, a company that uses an open source project will not see any upside if they pay. Collectively, yes, it would be better to financially support the projects that companies depend on, but on an individual basis, a company logically sees no reason to pay.
And even if the worst happens, say the sole maintainer gets hit by a bus and dies... most companies will be content to wait and see, and deal with that problem if and when it happens, not before then.
And they do. Most big OSS projects have at least some of the maintainers and contributors be employees at a company where that OSS is used.
All of those reasons you listed are choices that people make — not laws of nature.
People can of course choose to spend their political capital, and discretionary budget on random OSS projects. Or they can choose to spend it on their project, their goals, the outcomes that make their walk in the company easier, the actions that will get them promoted and not fired.
These are choices people make, and frankly OSS funding delivers very very little bang for the buck. (On a buck by buck basis.)
But neither goal is aided by hyperbole, pretending that not paying OSS is gravity.
which is why companies generally don't pay anything for OSS. If the cost is zero, but the benefit is not zero, then the bang for buck is infinite!
seems to all hinge on the justification part, for which ppl that do it for the lulz don't really care
No idea what the future will look like in general in 5, 10, or 20 years but I am reasonably confident that donations to OSS won’t be drastically more than they are now.
Hard to see why those things will stop being true.
Refer to any manager or executive at a tech company who uses open source to generate profits but doesn’t contribute as a “deadbeat” — so their choice becomes a source of social embarrassment.
That’s why I said to shame individuals, not faceless entities. And I think it’s fascinating that you didn’t reply to what I actually said.
Even as you tried to shame me (ie “if you actually believe that, you’re so dumb you’d buy something ridiculous!”) because you recognize that shaming is an effective tactic.
To quote someone else who's worked with Big Corp:
>>Do not fall into the trap of anthropomorphizing management. Think of management the way you think of a lawn mower.You don't anthropomorphize your lawnmower, the lawnmower just mows the lawn, you stick your hand in there and it'll chop it off, the end. You don't think 'oh, the lawnmower hates me' -- lawnmower doesn't give a shit about you, lawnmower can't hate you. Don't anthropomorphize the lawnmower. Don't fall into that trap about management.
>Even as you tried to shame me (ie “if you actually believe that, you’re so dumb you’d buy something ridiculous!”) because you recognize that shaming is an effective tactic.
I don't give a fuck what you think. I want to convince other people that you're wrong and we need better solutions for writing open source software because I enjoy doing it and I'd love to get paid for it. As far as I'm concerned you're a badly put together memetic lawnmower whose a danger to everyone around you - the end.
Bryan Cantrill (quote@) https://youtu.be/-zRN7XLCRhc?t=2303
specifically about Larry Ellison (and Oracle), not about management in general.
Worth watching from: https://www.youtube.com/watch?v=-zRN7XLCRhc&t=33m as he starts by praising Sun management.
And I'm not holding my breath that even that would sink in. People are amazingly talented at hearing only what they want to hear to justify doing it like they've always done it.
unless they're using some sort of hosted service for free, this cannot be critical. After all, software doens't rot, and they could continue to use the existing release until a (new) solution is found.
Look at how crowdstrike triggered outage didn't cause bankruptcy - that is more critical than most OSS would be.
Though I will note I'm agreeing that it's highly unlikely you can put a gun to the heads of corporations and get them to cough up, so I'm not sure what the point is here.
which doesn't happen instantly. For example, the end of life of the old java versions (1.5, 7 and 8 etc) - plenty of companies simply just paid a support fee and get support, while others paid to upgrade (or even change stack).
Most open source software, even with lack of updates, does not immediately start failing. The huge amount of time and leeway, even with security issues, is what prevents it from being critical, and prevents OSS from causing a bankruptcy.
Well, there's plenty of mission-critical FOSS used by plenty of companies. But you are right in that it doesn't just fail one day, and companies have plenty of time and options for dealing with abandoned FOSS.
(Which is one of the major benefits of FOSS. It's more likely with proprietary software that it can just disappear one day, with little recourse for users.)
And plenty others simply keep using the old 1.8 version because there's no budget to upgrade and there's no budget to 'pay a support fee'. And there's no budget to 'change stack'. Because... there's no budget.
Convincing people you need to upgrade or switch to keep current is often a hard problem, and sometimes has to be done with "you'll get all these new features!". But often "hey, we need some money to upgrade system X" is met with "hrm... it's software! It doesn't rot!".
> paid a support fee and get support
Me: "I think you cannot get corporations to cough up without some ridiculous extreme event like a behemoth dying. And I'm not holding my breath that would really do it."
You: "Your extreme ridiculous scenario is extremely ridiculous and here's why..."
Rinse and repeat.
Sure, but that won't happen immediately when the maintainer abandons it. It might not happen at all. There's usually going to be plenty of time for a company to switch to an alternative, or even take on maintainership themselves.
Which is absolutely correct!
As an open-source author and maintainer, I have no desire or motivation to call any of my users "deadbeats", especially when I license my software under terms that specifically do not require any kind of payment. That would be pretty hypocritical, as I've used lots of open source software (both personally and professionally) without paying for it.
[1] apart from Meta and Apple, they seem ridiculously low.
MPLv2/EUPL come to mind: they are compatible with proprietary products, but they make it mandatory to distribute changes/extensions of the library, not the whole product.
FOSS authors have a responsibility when they choose a permissive license.
No they won't. They'll only contribute if they're required to, or if doing so will be beneficial to them, and they'll do that regardless of whether the license says they have to or not.
When I've worked at companies that use FOSS, and have needed to modify those sources, I'll contribute back (regardless of license) if I think that change is likely to be accepted upstream, because I'd rather not have to maintain a fork. This would fall under "contribute if doing so will be beneficial to them".
At any rate, no FOSS license (that I'm aware of, or is in wide use) requires users to contribute. At most, they require that changes be made available. There's nothing that says the changes need to be submitted (or accepted) upstream. Often getting a change into a state where it would be accepted upstream is a significant amount of work beyond what the company has already done for their own purposes, so they don't bother.
Did you actually read my comment before you answered? Because I said that copyleft licenses "make it mandatory to distribute changes/extensions", which means that companies are required to contribute if the license is reciprocal.
> At any rate, no FOSS license (that I'm aware of, or is in wide use) requires users to contribute. At most, they require that changes be made available.
Making changes available is a form of contribution. If you work on a proof of concept for a month and at the end your company decides not to use it in a product (thanks to the learning from your work), do you say that you did not contribute, so you should not be paid?
Feels like you're being pedantic just for the sake of the argument.
With copyleft licenses, nothing say that you have to get your changes upstream: you just have to distribute them to the users. It's not a whole lot more complicated than attribution: set up a repo and put your fork there publicly.