Computation Is Beautiful

Showing artists that computation is not the enemy of mystery — it is one of its deepest sources.

The idea

Rigor doesn't drain the magic out of art. Done right, it shows you where the magic actually lives — and proves that some of it can never be pinned down at all.

A lot of artists meet computation as a threat: a force that flattens, automates, and explains away. I want to offer the opposite encounter. Mathematics can clarify why a musical system feels inexhaustible. A formal language can reveal the hidden architecture of a constraint you've been using by instinct. And theoretical computer science can tell you, precisely, which artistic problems will never reduce to a simple answer. That last result isn't a defeat — it's a guarantee that the mystery is real.

Ableton is a computer

One of my favorite findings: the digital audio workstation Ableton is Turing-complete — in a deep sense, the same kind of object as our most powerful computers. I don't treat that as a party trick. I treat it as an aesthetic. It means a live set is not just a sequence of clips but a computational system with its own unpredictability, its own emergent behavior, its own capacity to surprise the person at the controls. The instrument is stranger and more alive than it looks.

Where formalization breaks down — on purpose

My research into undecidability in modern music-making asks where the felt logic of music provably escapes any algorithm: constraints on polymeter, satisfiability in just-intonation harmony, the limits of new ordering systems. These are exactly the places where "what makes a song work" stops being a formula and starts being a question only listening can answer. Computation, here, is the tool that maps the edge of its own usefulness — and hands the rest back to the artist.

Bring me your theory of your own art

If you carry a private theory about your work — a rule you always follow, a feeling you chase, a structure you suspect is there — I'd love to help you formalize it in a language mathematicians and computer scientists would recognize, and then see what the formalization reveals. Not to replace your intuition with an equation, but to give your intuition a mirror precise enough to argue with. AI and computation are instruments in that conversation, never the author of it.

Selected work

Undecidability Results and Their Relevance in Modern Music-Making → Improving Structural Diversity of Blackbox LLMs via Chain-of-Specification Prompting → Axiomatic Hermeneutics → All work on Google Scholar →