This article has a number of fundamental errors, and this leads to inaccurate, or ineffective conclusions. Just a few;
>> The current state of software is such that everything depends on everything else.
This is not true. There is a hierarchy of dependencies, yes. But the use of "everything" here is not helpful. There are clearly projects that are used a lot (OpenSSL) and there are plenty of projects that are used by nothing. And everything in between.
>> since the software is free, you can not sell or license it to make money.
again, very not true. RedHat being a prime example, but there are any number of companies selling and licensing open source software. That is explicitly allowed under the various licenses. Mixing the concept of "freedom" and "cannot be sold" is a common FOSS myth.
>> So why do people abandon their widely-used FOSS projects?
For exactly the same reasons they abandon commercial projects. The vast majority of all projects are abandoned, regardless of license. Because _most_ projects are rubbish. If we narrow the set of programs to "widely used", again there is no difference between FOSS and commercial - DOS was commercial, widely used, and abandoned. (For some definition of "abandoned".)
As long as the list posted in the article, there are also many other reasons. Developers lost interest / die / get reassigned / get married / have kids / learn to play golf.
My point - talking about FOSS as a single category is meaningless because it spans too large a ground. It's not like there's a single common cause to project failure, or indeed a single common reason they succeed. Do we even want projects to succeed forever?
>> Therefore, by donating money to developers, you prevent the project from being abandoned.
eh, maybe. For some small number of projects. For a while. Until one of the other items on the list raises it's head. But money is the easiest problem to solve, and in many ways the least consequential.
Took the words out of my mouth and then some, by a mile (In reality I only had ~ two of those arguments to make).
In particular, completely agree RE: the misplaced itch to ascribe some abandonware label to FOSS exclusively.
It is outright comical to place the locus of util loss on FOSS when deprecated proprietary software stacks are relinquished all the goddamn time (hi, bluetooth explorer in xcode tools!) — and usually without effort toward open sourcing the leftovers for continued community support. Just deadweight loss from that point and on.
Pretty clear there's a causal inference short circuit (charitably) in this blog post.
> thanklessly maintaining for decades with no pay. This is clearly an issue
Why is it an issue?
This is an assumption. Not everyone who creates and maintains FOSS expects to be paid in money. Many of us got to where we are off the backs of previous FOSS authors freely given efforts and feel that it's our way of giving back to the world.
If there is one critical dependency, relied upon by many, where the single author/maintainer is literally crying for help - in that they are unable or unwilling to continue giving their free time - then fine, maybe _that_ project needs outside help or outside funding... but that is entirely subjective.
I agree, it is a bad assumption. Not everything has to be about money.
In the last 15 years, I have averaged over 15 hours a week writing Creative Commons licensed eBooks and the open source code that goes with the books. Even though a few people choose to buy my books rather than take free copies, I do this with no expectation beyond hoping to meet interesting people and improve my networking.
I suspect that the majority of project maintainers who are hurting, do not “literally cry out for help”, especially if they regularly read news fora like this which underscore how meagre an amount of donations rolls in even when the maintainer cries out for help.
Perhaps we've got it the wrong way round, and rather than asking how Open Source contributors can be paid more, what about asking how everything can be cheaper? This is in fact what's happening as Open Source is deflationary. It's the reason that various central banks round the world can print so much money without the inflation rate getting as high as one would expect. I contribute to a few Open Source projects, and that's the approach I favour.
Abolish private property and patents. Let work be self-organized by the workers themselves in cooperatives, without control by a State or shareholders. Then every body can live "for free", work in the public interest (not for the profits of a vampire) and have an actual say in governance of society. That's in essence, the ideas of anarcho-syndicalism. The economic aspects of such have been widely studied by Proudhon who is a father of the cooperative movement (please don't read his view on women and jews). It's been practiced on a wide scale during the spanish revolution of 1936 with self-organized production across Catalonia.
It's still practiced today, though not so "profoundly" as a form of public policy in Chiapas (autonomous indegenous communities of Mexico) and Rojava (the autonomous, multicultural and secular revolutionary administration of north-east Syria).
Thanks, an interesting comment. The word "anarchy" has bad connotations, but anarchy sometimes works well. For example here in the USA, the relationships between states seems like a well functioning form of anarchy to me.
Obviously our system is rigged, and has been for centuries. I accept this and don't spend much time worrying about US Congress people being able to do insider training, accept more contributions than I am comfortable with, etc. That is the way the world is: corrupt.
That does not mean that we can't push back when we can by not accepting corporate news media (MSNBC and Fox News seem like the very worse, but they are all bad) and focus on our own productivity and on treating other people with kindness. I am very socially liberal but I would like to see a much smaller government - a weird combination that is probably impossible to achieve.
It depends who you ask. For us anarchists it's the most beautiful thing there is, and millions of us have died through the century for these ideas of unlimited freedom and equality.
> in the USA, the relationships between states seems like a well functioning form of anarchy to me
not really? I mean there's a federal government ruling from above, no? You couldn't just have a State abolishing wage slavery and private property and declaring food and housing for all, could you? I'm betting other States would send the federal army to crush this rebellion, but admittedly my understanding of US politics is limited. I'm basing it on how the federal government has dealt with indigenous nations (crushing and betraying them at every turn) and the war of secession, during which both sides were happy to exploit black people but many soldiers killed about how to do it precisely.
> I am very socially liberal but I would like to see a much smaller government - a weird combination that is probably impossible to achieve.
You may be interested to read some anarchist literature, because it's not a weird combination at all. Maybe Emma Goldman or Murray Bookchin could be of interest to you to get started?
Anarchy does not ensure freedom, equal justice or equal political rights. Only minimal government can provide for that, but most people understand anarchy to mean the absence of even a minimal government.
Sorry when i said "a form of public policy" i did not mean from the State. The mexican government has very little intervention in the region of Chiapas, especially outside of big cities in the indigenous communities. I was talking about the roles, cultures and politics of those organized indegenous communities as part of the zapatist movement.
You will be thrilled to learn that these folks built their own hospitals and schools because despite theoretical "federal aid" the government does nothing for the people, except sending the military to fend off any popular uprising and/or arming and financing fascist paramilitaries.
It's more visible over there than over here (France), but the same logic applies. No ruler will ever do something useful for the people. We have to do it by ourselves and for ourselves, such is the logic of democracy, and any parliament ruling from above is fundamentally incompatible with the basic principles of democracy (or anarchy if you'd rather call it that).
The economics of FOSS is not a solved problem and the author deserves kudos here for asking questions that I don't see asked anywhere else. The examples chosen are maybe not the most current, and there are counterexamples especially in B2B contexts but I'm applauding here and I'll save my more critical review for the book when it comes out.
>The economics of FOSS is not a solved problem and the author deserves kudos here for asking questions that I don't see asked anywhere else.
Are you sure? FOSS has been around for decades now and FOSS underpins all commercial products. It feels like whatever the economics of FOSS are, they are working.
I'm assuming they're rewarded fairly at market rates for continuing to ensure FOSS developers are spending years of their lives debugging and solving hard problems, and that no one else is capable or willing to do an equivalent job for less.
This article is somewhat interesting but kind of disappointing. I was expecting the author to suggest some policy recommendations based on the observations after the problem and various options were analyzed, but it ended abruptly.
In spite of how much we all benefit from the likes of Google, Apple, Amazon, etc., it's surprising and very disappointing how some (greedy, lazy) open source developers eventually tire of doing large amounts of work for free in order to help these excellent and public-spirited companies (or specifically their executives and shareholders) enjoy massive profits.
It's a shame that so few open source developers seem willing to beg for tips on github, and can't even be bothered to sign up wealthy patrons and donors for their projects, as recommended by this fine article.
So was the .pn TLD for the Pitcairn Islands. There ended up being a dispute between that person (who was actually an islander) and the government of said islands which ICANN resolved in favor of the government.
There are a number of flat-out incorrect statements in that article.
>FOSS is typically created by individuals, not by corporations (although there is a shift toward FOSS companies), and since the software is free, you can not sell or license it to make money.
Most major FOSS projects have significant corporate backing--and have for a long time. And those companies sell open source or otherwise use it in ways that make them money.
Or is Red Hat the exception that proves the rule? Red Hat has been around for a long time and is still one of the very few OSS organizations of that magnitude. The other ones I can think of like SuSE all also date back to the same time period. We're not seeing new Red Hats or SuSEs being born.
There's also Canonical and Debian in the distro world. Elastic is still around, seemingly panicking. Docker still exists for now, but probably should have sold when they had the chance. Grafana seems to be doing fine. Rancher got by without even trying to sell enterprise licensed versions of the software, just by selling support, and was recently acquired by SuSE.
I think there are plenty of examples out there, but they're all enterprisey, not consumer-targeting. Consumers tend not to have sufficient requirements to make it worth purchasing support and licensed versions of anything bigger than Sublime Text. Businesses love this stuff, though, since open source allows them to fix things in-house and send patches upstream rather than having to wait for the vendor to get to it.
>> The current state of software is such that everything depends on everything else.
This is not true. There is a hierarchy of dependencies, yes. But the use of "everything" here is not helpful. There are clearly projects that are used a lot (OpenSSL) and there are plenty of projects that are used by nothing. And everything in between.
>> since the software is free, you can not sell or license it to make money.
again, very not true. RedHat being a prime example, but there are any number of companies selling and licensing open source software. That is explicitly allowed under the various licenses. Mixing the concept of "freedom" and "cannot be sold" is a common FOSS myth.
>> So why do people abandon their widely-used FOSS projects?
For exactly the same reasons they abandon commercial projects. The vast majority of all projects are abandoned, regardless of license. Because _most_ projects are rubbish. If we narrow the set of programs to "widely used", again there is no difference between FOSS and commercial - DOS was commercial, widely used, and abandoned. (For some definition of "abandoned".)
As long as the list posted in the article, there are also many other reasons. Developers lost interest / die / get reassigned / get married / have kids / learn to play golf.
My point - talking about FOSS as a single category is meaningless because it spans too large a ground. It's not like there's a single common cause to project failure, or indeed a single common reason they succeed. Do we even want projects to succeed forever?
>> Therefore, by donating money to developers, you prevent the project from being abandoned.
eh, maybe. For some small number of projects. For a while. Until one of the other items on the list raises it's head. But money is the easiest problem to solve, and in many ways the least consequential.