Blog

iSAQB Software Architecture Gathering

Creator of the Rockstar programming language
Published on October 9, 2026

Free Isn’t Free: Dylan Beattie on Open Source, Money, and Sustainable Software

1. Which association of free do you think is more often meant when used for software: free as in beer vs. free to adapt and re-use?

I think people overwhelmingly think “free software” means “free as in beer” - software that doesn’t cost any money. The best evidence I have for this is the number of times I’ve had to make the case for paying money for software which is published under an open-source license and been told “but we thought it was free” - in fact, the term “open source” was deliberately coined in the late 1990s, around the time that Netscape Navigator first published their source code. Navigator was still a paid commercial product at that time, so the folks associated with that endeavour wanted to find a term that captured the “free speech” sense - study, modify, adapt - without using that word “free”.

However, I note you specifically asked what is more often meant, and perhaps that’s a different question. Free software is as much about ideology as it is about practicality, and there is a long history of people aligned with that ideology for whom “free” has always referred to freedom, and who seem reluctant to engage with the idea that the word itself might be confusing people.

2. How high does "fame" rank among the reasons why people start open source projects? Is that a good reason and does it work?

Fame is perhaps the wrong word: I don’t think anybody starts an open source project thinking it’s going to make them famous in the sense that movie stars and pop stars are famous - but I do believe that there’s a desire for recognition and acknowledgement that drives a lot of people to share their code. Larry Wall, the creator of Perl, famously proposed that the three great virtues of a programmer were “laziness, impatience, and hubris” - laziness drives people to write code that makes their lives easier, impatience drives them to make that code fast and reliable, and hubris drives them to write code they want to show off to their peers.

As for “does it work” - that depends on your definition of success. Sure, there are open source developers who are very well known, but I think in a lot of cases that reputation is as much down to their “personal brand”, their presence on social media, posts to mailing lists, etc. as it is to their code.

Put it this way: if your only objective is to be famous, there are probably much easier ways to achieve it than starting an open source software project. 😉

3. Some corporations fund open source software by contributing or donating to projects. Some individuals donate, too. Isn't that a fair deal?

I think if you’re using open source software to make significant amounts of money, you should be prepared to contribute to the cost of maintaining that software. By “significant amounts” here I’m talking millions of dollars, and by “contribute” I’m talking hundreds, maybe thousands of dollars per year in ongoing contributions. I’ve spent some time over the last year or so putting together an initiative called “Use What Works” (https://usewhatworks.org), a collaboration with various folks connected with open source, intended to promote awareness and acceptance of these ideas and explore the various models and platforms that exist to help open source developers achieve the financial security they need to guarantee continuity of their projects.

The fact is that when your organisation takes a dependency on an open-source project, you’re taking a risk. You’re accepting the risk that there could be security vulnerabilities, bugs, performance problems, incompatibilities with future platforms and systems - and in too many cases, even fairly mature organisations have no model in place to capture and highlight that risk, and so we see instances of what’s become known as the “open source entitlement complex”, where large, well-funded organisations believe they’re entitled to priority support. The most notorious example of this perhaps being Microsoft’s posting on the ffmpeg issue tracker in 2024 “this is a high priority ticket and the FFMpeg version is currently used in a highly visible product in Microsoft” (https://trac.ffmpeg.org/ticket/10341#comment:4) - apparently after Microsoft had refused to pay for a support contract.

I’m sympathetic to the engineers in question - I know all too well that being tasked with fixing a particularly gnarly bug doesn’t give developers the discretion to spend money or negotiate support contracts - but at an institutional level, donations are the wrong mechanism.

Organisations want support contracts they can rely on; project maintainers want predictable revenue streams they can use to purchase tools, hire developers and allocate the resources they need to ensure the sustainability of those projects.

4. Why haven't we seen widespread fees / micropayment mechanisms for open source tools and libraries yet? Would they be circumvented all the time? Is it too complex? Is a "precise" compensation approach a dead end?

I honestly don’t know. There’s a very clear, logical model that could underpin this: source code is free, but you have to pay for binaries. Combine that with a package management ecosystem based on a Spotify-style model: consumers pay a subscription fee, revenue is allocated to the project maintainers based on downloads. Benefits to maintainers are a direct revenue stream based on the popularity of their packages, but without any implication of support, so they’re still free to negotiate support and maintenance agreements separately. Organisations get a reliable feed of binary packages, similar to how npm or NuGet operates today, potentially with benefits like audited package versions as a mitigation against supply chain attacks. David Whitney took this idea a step further by proposing that a companies’ package subscription just gets added as a line item on their GitHub or their Azure invoice - the kind of companies we’re talking about here wouldn’t even notice an extra $500/month on their Azure bill.

5. (Not 100% serious, but maybe a bit:) Does capitalism itself stand in the way of a system of supportive and fair cooperation(, again)?

Yes. I believe the particular strain of capitalism that’s endemic in the technology industry is optimised for short-term profit and growth above all other considerations, and that is often diametrically opposed to the decisions and investments that would promote long-term sustainability. The corporations that make significant contributions and investments in open source are generally only doing so because they think it’ll make them more profitable - to pick one specific example, I personally love the way Microsoft has embraced open source in the last two decades, but I also believe 100% that the main reason they’ve done it is that embracing open source makes it easier to promote their Azure cloud platform to potential customers who would otherwise be looking at AWS or Google Cloud.

6. Something else: Was there a certain (rock) song that you remember that influenced your perception and taste in music most profund. Is it still on your playlist?

The single song that made the biggest impression would have to be “Livin’ on a Prayer” by Bon Jovi, from the album “Slippery When Wet”. I remember it blasting out of the aging stereo system at a youth club I used to go to when I was about ten years old and being absolutely blown away; I’d never heard anything like it…

And not only is it still on my playlist, right now there is exactly one audio CD in my car, and you guessed it - it’s “Slippery When Wet” by Bon Jovi. One of the greatest albums ever recorded.

Many thanks to Dylan for the great interview!

Explore the full SAG 2026 program and secure your ticket for Berlin.