Why RSB

I am a software developer because of open source.

That is not a figure of speech. More than thirty years ago I learned to program, and I learned it from the work other people gave away — not only their source code, but the articles and books they wrote to explain it. Knowledge handed over by people I never met, for no reason other than that sharing it was worth doing. Across a long career since, every job has run on open source in one way or another. The community handed me a roadmap for how to learn and how to build, and somewhere along that road I picked up the idea that has shaped how I work ever since: the value of a feedback loop — that you make something, watch how it actually behaves, and let what you learn correct the next pass.

For all of that, I have always been on the using side of open source, never the contributing side — not by choice so much as circumstance. A working career has a way of taking all the hours you have, and the time to take part was always the time I did not have. That is what I want to change. RSB is my contribution, offered in the same spirit it was given to me — not to settle a debt, because that is not how this community works; you do not pay open source back so much as keep it going. After a long time on the outside of it, this is me joining in.

But the community I want to join is not in the shape it was when it taught me. Two things are closing in on it, and together they are why this project takes the form it does.

The first is happening to the knowledge itself. The web of articles, answers, and shared code that taught me to program — and that has taught everyone who came after — is being quietly drained. Someone with a question who would once have landed on a project's documentation, its forum, its repository, now gets the answer from an AI and never arrives. For many projects that traffic was not vanity; it was how they were found, how they drew contributors, sometimes how they kept the lights on.

What worries me more is what fills the gap. A growing share of what gets written is now generated rather than authored — produced fast, published without anyone checking whether it is true, and then read back in by the next model as if it were settled. Knowledge no one validated becomes training data, becomes more knowledge no one validated. It is a feedback loop — the same idea that shaped how I work — but running the wrong way, each pass leaving the web a little less trustworthy than the last. I am not against AI — I use it where appropriate, but I validate everything it produces and never treat it as a source of truth. It is a double-edged sword, one that has to be used responsibly. What I am against is not the technology but the carelessness around it: knowledge entering circulation that no person ever stood behind.

The second enclosure is older and has nothing to do with AI. The tools serious creative work depends on — the full-featured editors for photographs, for documents, for video — have been fenced off. A handful of large companies own that ground, and the cost of admission, in price or subscription or lock-in, quietly decides who gets to make serious work and who does not. I came to this through photography, which I love, and where the pattern is plain: the software a professional relies on is rented, not owned, and a beginner's ambition meets a paywall long before it meets the limits of their own skill.

Two things the community gives freely, then, both being closed off: the knowledge that teaches people to build, and the tools they build with. That is what RSB is a response to — and it is why the response has two halves.

The two halves are built by the same hand, to the same standard, and that standard is the part I care about most, so let me start there. Everything in RSB is built in the open and held to a small set of pillars — the test every decision is measured against — and nothing, no claim and no line of code, is meant to enter circulation without a person standing behind it. The pillars are written down, in public, so you can hold me to them.

The first half is the tool. I am building Lab: a free, professional-grade photo editor for the desktop — photography being the work I know well. Not a stripped-down alternative to the expensive thing, but the serious tool itself, given away. Underneath, it is built as small crates with enforced boundaries, ordered from the most general to the most specific, so each piece can be understood and trusted on its own rather than tangled into a single mass — the full shape lives in the architecture reference. And it is free in both senses: it costs nothing, and the source is open under the GPL, so no one can fence it off again.

The second half is the knowledge. Building the tool this way produces, almost as a byproduct, the very thing disappearing from the web: a body of architecture and engineering knowledge that can be trusted because it was validated. Every significant decision is reasoned through and recorded; the rules those decisions imply are written down as standards; and all of it answers to the same pillars. It is centred on Rust, because that is what Lab is built in, but the principles are not — they are how to build serious software well. A developer starting out should be able to read it and learn not only what was decided but why, and trust that a person stood behind every word.

These two halves hold each other up. The knowledge is not theory, because there is a real, running system that had to obey it — the tool is the proof. And the tool can be trusted, because every decision inside it is open, reasoned, and validated — the knowledge is the receipt. Most projects offer one or the other: software with opaque internals, or writing with nothing running behind it. RSB insists on both, because that is what open source gave me — code I could read and the writing that explained it — and it is the standard I want to give back.

Those two halves are the project. Beyond it, there is something larger I am hoping for — something no single project can be on its own. RSB is one island of trusted, validated knowledge, and it would mean little as the only one. So part of the work is to find the others — the projects and writers and sites that hold their own work to a standard like this one — and point to them, openly, so a person learning to build has somewhere to stand. A single trustworthy source is a refuge; a network of them is a place to grow up in. I would like RSB to be a good neighbour in that network, not an island pretending to be a continent.

I should be honest about where this is. It is early. There is more written down than there is built, and what is built is the foundation rather than the house. I am one person, and for now the community is whoever is reading this. That is not something I am hiding — it is the invitation. The people I hope this reaches are the ones who would rather build one thing correctly than ten things quickly, who read a decision and want to know why, who would tell me when I am wrong and expect me to listen. If that is you, you are early, and early is the best time to arrive.

There are as many ways in as there are pieces of the work. Use the tools and tell me where they fail you. Read the decisions and the standards and hold them to the pillars — they are public precisely so you can. Write something to the same standard and let me point to it. Fix what is wrong. Or just build alongside, in your own corner, with the same care, and let that be the contribution. None of it has to be code.

Open source did not survive because anyone was paid to keep it alive. It survived because people who had been given something useful chose to add to it rather than only take. That is the whole mechanism, and it is the only one. I was given a great deal, and for a long time I only took. This is me adding to it — and an invitation, if any of this is yours too, to add to it with me.