The browser is still the most interesting application platform

I still think the browser is the most interesting application platform we have. It is everywhere, weirdly capable, constrained in productive ways, and close enough to users that a small idea can become usable quickly.

That is why I like building browser-native experiments: a tiny computer that runs slowly enough to understand, a guitar that computes sound in real time, a simulation where motion explains the system better than a paragraph could. These projects are not separate from my product work. They train the same instincts.

A good interactive system has to make state visible. It has to respond quickly enough that users trust it. It has to explain itself without making the user read a manual. Those are the same problems inside a SaaS workflow, just with different costumes.

Side projects also keep me honest. When you build a small system end to end, you cannot hide behind process. The audio either sounds right. The simulation either moves. The CPU either steps through the instruction. The interface either helps or gets in the way.

I like that directness. It reminds me that software is ultimately experienced, not described. Architecture, design systems, and product strategy all have to land in something a person can use.

The browser is a good place for that kind of practice because it rewards clarity. You can start with a blank page and end with a working instrument, machine, or model that someone can touch.

That still feels worth being excited about.