Ask HN: Cooperative or Non-Profit Software Development vs. Open Source

2 points by kkoning ↗ HN
Like many of you, I suspect, my early professional development was heavily influenced by open source and free software. The freedom as in both free speech and free beer was definitely preferable to always being at the mercy of rigid and opaque proprietary software controlled by a profit-maximizing firm. It was even worse when there was no real competition (e.g., 90s Microsoft).

However, after I left tech and went to graduate school (law, economics, and antitrust, roughly...), I've found my perspective has changed quite a bit--to the point that RMS came to give a guest lecture and I'd cringe a little... OK, a lot. The thing is, his central critique still isn't wrong--proprietary software does constrain people to advance the interests of its owner. It's just that his views on free software are... sufficiently myopic that they're ultimately self-defeating.

The thing is, free software CAN work, economically, but it only works well where there's some cross-subsidy. There could be an adjacent market for professional services, huge corporate users that don't want to be beholden to a single external supplier, developers working cooperatively on their own tools, academics working towards publications, etc... Open source works great for Linux, Apache, Python, Postgres, etc..., and is preferable if and when it does.

But it doesn't seem to work as well outside of these situations--where there aren't strong cross-subsidies. For example, most people prefer to pay for Photoshop and QuickBooks, even though GIMP and GnuCash are free. The developers of those systems are probably great people sacrificing their own time and energy to contribute something to the world--it's just that there's only so much you can do when you have very limited resources.

The consequence, I think, it that free software can't really be successful outside of its cross-subsidized niches. And that's not a great situation, because the alternative in those cases seems to be fully proprietary software.

So... that got me thinking about some hybrid between the fully proprietary software model and the fully open source software model. Could there be an economic model for creating software that maximizes the interests of its users (as opposed to having a fiduciary duty to maximize shareholder value) but isn't starved for resources by the strict open-source requirement that its price must be zero?

The obvious answer would seem to be some form of non-profit organization, and an extremely permissive (but not completely open source) license, but I can't think of any examples. E.g., a license that says things like "here's the source, you can publish modifications for other paid users" or "this license reverts to MIT after 10 years", etc... A non-profit that sets the license price sufficient to pay developers, but not to line the pockets of management and shareholders? Sure, there are plenty of open-source projects with attached foundations, but being open source means their only economic support comes from either cross-subsidies (see above) or charity, and not the one that makes the most sense in market economies--license fees.

I can't be the first person to have thought along these lines, but I'm having a hard time finding anything to read on the subject. There's the "open source" vs. "free software" debate, of course, but neither camp seems to favor licenses requiring payment for use. What say you, HN? Is there a whole literature on this and I just missed it somehow? Is it just such a dumb idea that nobody talks about it? Is it somehow more novel than I assume?

3 comments

[ 3.6 ms ] story [ 21.6 ms ] thread
I think that we are about to see a new strain of F/LOSS motivations and advocacy emerge, centered on human rights and political freedom, anti-surveilance, privacy, resilience, and so on. On one level I expect it to continue the "redecentralize the web"[0][1][2] movement from a few years ago, but sideline the mostly cryptocurrency focused aspects of "web3"[3] in favor of the "fediverse"[4][5][6] and "local first software"[7].

[0] https://redecentralize.org/

[1] https://ruben.verborgh.org/articles/redecentralizing-the-web...

[2] https://github.com/redecentralize/alternative-internet

[3] https://medium.com/@shevski/how-decentralised-are-you-a6539e...

[4] https://en.m.wikipedia.org/wiki/Fediverse

[5] https://www.makeuseof.com/what-is-the-fediverse-and-can-it-d...

[6] https://thenewstack.io/why-developers-should-experiment-with...

[7] https://www.inkandswitch.com/local-first/

Thanks for taking the time to respond and collect all those links, I really appreciate it! Not just for the great reading material, but hopefully some potential community / contacts for people already working in this space (broadly).

I hope you're right, on... yeah, every single point here. It seems like we need something that's definitely not the mostly centralized world of the major tech giants, but also not completely atomized and economically unsustainable in a way that creates a power and technology vacuum... that leads inexorably back to centralized control and exploitation.

The essay on "local first software", in particular, really resonates. Going back to my point about our economic models for technology development (strongly) influencing the nature of the technologies that get developed, it seems like a lot of the existing open source technology is designed in way that facilitates commercial for-profit activities and structures. For example, most databases aren't designed for end-user access directly, nor do they enable users to bring different applications to the same data, in the way that traditional filesystems do. On the other hand, the traditional filesystem paradigm just doesn't scale like applications with big back-end databases. It's like... there's a whole type of infrastructure for which the technology stack just isn't really there yet. Or, maybe more accurately, hasn't yet come together in quite the right way.

I'm just starting to get more involved in this scene, but, just based on where I've come from studying the economics of ICT, this seems like just going to take considerable effort to get started. Probably a lot of work by activists/enthusiasts/etc... at the beginning. In the long run, I think it needs something more structured and economically SELF supporting than the traditional open source model as well. I'm hoping all of this is just... early in the process, so the organization that you'd see around larger projects just hasn't coalesced yet. I'll have to get in touch with some of these guys and have a few conversations...

Again, thanks so much for taking the time to reply.

You're very welcome. The Ink and Switch folks have more interesting posts (and a few released prototypes) that advance the conversation, which are dense and thoughtful:

https://www.inkandswitch.com/

You're right that the economic incentives aren't such as to produce highly polished frameworks, libraries, and apps yet, and the public discourse is fairly diffuse. An entrant that just popped onto my radar is Aphrodite:

https://news.ycombinator.com/item?id=31446780

Another org that may turn out to be a player is the Bluesky team that was spun out of Twitter:

https://blueskyweb.org/blog

Adhoc mesh networking is something that has been potentially available forever, but existing business models of network operators (who subsidize and resell — and therefore shape — hardware) are hostile to it. R&D-wise it is sidestepping mobile communication devices per-se and finding applications in niches like autonomous drone swarm coordination, which is obviously restricted to mostly military applications. In the absence of direct support for similar capabilities on smartphones (or the ability to add such support through Software Defined Radio networking) we see apps like Briar filling the need with occasionally-connected approaches:

https://briarproject.org/

Besides the various terms mentioned previously, you might also find the phrase "offline-first" of interest.