Preface: Why Recurrence, Why Odisena, Why Now

I did not set out to write a book about the Fibonacci sequence. I set out to solve a problem I kept having, which was that my own systems kept forgetting how they had arrived at their current state. A deployment would succeed and I could not say precisely what had changed. A document would reach its final form and I could not reconstruct the reasoning that produced a particular paragraph. A device recovery would half-work, and the partial evidence — the one clue that would have told me what to try next — would be erased by my own next attempt. The failures were never dramatic. They were quiet losses of provenance, and they compounded.

The pattern behind all of these failures was the same. In each case, I had treated growth as a single forward motion: do the next thing, overwrite the last thing, keep moving. Speed felt like progress. But a system that only moves forward cannot go back, and a system that cannot go back cannot truly validate, because validation means comparing a new state against a preserved old one. Without preservation, every check is a guess.

The Fibonacci sequence entered the story because it is the cleanest small example I know of a system that grows by remembering. Each term is literally the sum of the two terms before it. You cannot compute the next number without holding on to the previous ones. The sequence is not a metaphor for memory; it is a mechanism that requires memory to function at all. That property — growth that structurally depends on preserved prior state — is exactly the property I wanted in my own work. So I began using the sequence as a teaching instrument: a place to practice the discipline of building forward without losing the past.

Out of that practice came a vocabulary, which over time hardened into a method. The method has four framing moves I call RFPA — Recover, Field, Principal, Assumption — that you make before you allow a system to grow. It has four execution beats I call AVPT — Attempt, Validate, Preserve, Track — that you cycle through as the system grows. And it has a governing idea I call the Infinity Engine: the notion that the value of a system comes not from running once, but from running many governed cycles, each of which preserves enough evidence to be validated, reversed, and continued. These are working names for working practices. They are not claims about the universe.

This is why the book is so insistent about labels. The mathematics in these pages is real mathematics, and I have taken care to state it correctly and cite it to sources you can check. The method is real method, and I have used it on real projects. But the bridge between them — the mapping that says "computing the next Fibonacci term is like attempting the next state of your system" — is an analogy, and analogies are powerful precisely because they are seductive. A good analogy teaches; a bad analogy smuggles in false certainty. The whole apparatus of CANON, FACT, METHOD, INTERPRETATION, and CASE exists so that you always know which one you are reading.

Why now? Because the cost of forgetting has never been higher. We are building systems — infrastructure, organizations, and increasingly systems that write and decide with the help of machine intelligence — that change faster than any individual can track by memory alone. A generated answer, an automated deployment, a migration run at three in the morning: each of these produces a new state, and each of them, done carelessly, destroys the evidence of how the previous state was reached. The organizations and individuals that will do well are not the fastest ones. They are the ones that can grow and still say, at any moment, exactly how they got here and how to get back.

That is what this book is about. It is a book for builders who want their systems to be recoverable, for founders and leaders who want their organizations to remember, and for serious general readers who suspect, correctly, that the difference between real progress and mere motion is whether you preserved the evidence. It uses a two-thousand-year-old sequence to teach a very contemporary discipline. It asks you to slow down exactly enough to never lose your place.

— Ian Dansereau, 2026