Two Fields at Once

The problems that interest me most always seem to require looking sideways. Not the core competency — the adjacent thing that turns out to be load-bearing.

Solar positioning needs orbital mechanics. Interpretable machine learning needs cognitive psychology. Good process automation needs organizational theory. You can't get there by going deeper into the main field; you have to go wide enough to see what the main field borrowed from.

I don't think this is special insight. It's just what happens when you stay curious about how things actually work, rather than just the patterns for handling them.

The skill isn't knowing two fields. It's noticing when a problem has a second field hiding in it.

There's a version of this that becomes a trap — the person who's always going wide, never going deep, treating everything as adjacent to everything else. I've been that person. It's less useful than it feels.

The actual move is: go deep in one thing long enough to understand its limits, then notice what's on the other side of those limits. That's usually where the interesting work is.