The dopamine of complexity

I finally caught up on Tobi's latest conversation with Shane Parrish. I like listening to him. His thinking is unusual. He builds simple mental models for complex problems, and he tends to be ahead of the room on what is coming. I did not fully appreciate that when I was at Shopify. Now, on the other side of the desk building a company, a lot of it clicks.

Two things stuck with me.

The first is that humans love complexity because our dopamine wiring rewards it. Tobi puts it plainly: "we love complexity" and "that is part of our dopamine discrimination system," and "people peddle complex answers to simple problems all the time." Building something intricate feels great. I have been allergic to that reflex for years. It is the reason we started Tuist. The tooling around us looked complex then, and much of it still is. Coding agents help by doing on-the-fly conceptual compression, but compression does not make hard things simple. It hides them.

An example. How many teams today build a heavy client that pulls state from the server and rehydrates the whole page in the browser, when a server rendering the pages directly would have done the job? We reach for the tools the giants built. Query languages designed for federated schemas, binary protocols invented for microservice fleets, pick your poison. Using them makes us feel like we are playing at their level. We are not Meta, but at least we build like Meta. Even when what we are building is small.

I look for the opposite. Tools that maximize value and simplicity at the same time. That is how we ended up on Elixir and Erlang. It feels unusual to a lot of people. It is also why we move as fast as we do.

The second thing that stuck with me is his suspicion of orthodoxy. His words: "if there is an orthodox way to solve a problem I am incredibly suspicious." Sometimes the orthodox way is right, especially in regulated fields. His first move is still to question it. Mine too. As Tuist expands into new ecosystems, I try to walk into each one with fresh eyes. I look around. I watch the people. I watch the problems. I watch the social dynamics. Slowly I build a picture of why things are the way they are, and whether they still need to be.

I do not try to change those rooms from the inside. Most cannot be changed that way. They have grown too dogmatic. And dogma usually shows up wrapped around complexity.

That is why I love what Jeff has done with Mise. The Nix tribe has its story about purity and determinism. Jeff looked at it and said: I want my environment provisioned fast, intuitively, with the best developer experience I can imagine. He embraced a stateful reality and built something people actually want to use.

David Heinemeier Hansson is doing the same with Omarchy, and getting flak for it. Interestingly, Tobi runs Omarchy too, and in the same interview he calls it a computer that "fulfills wishes" and points at it as a preview of where software is going. Sure, it is config files. But it starts from the experience and works down through layers that had gone dormant for decades. That is inspiring.

This is Tuist's motto, and it is not going to change. I believe this is how you build technology people remember. Not because they climbed a peak of complexity and got to tell their manager they are now indispensable, while ranting that everything else does not scale. Fuck that. You build technology people remember because of the emotion its simplicity evokes.