Right, and that's also true of physical inventions that you might think are a straightforward application of scientific knowledge. The steam engine is a good example -- it was produced by tinkering, and we devised a theory of how it works much later.

Probably for the vast majority of human inventions, the thing exists long before the theory of the thing. I think that feels counterintuitive in part because examples to the contrary are very conspicuous -- e.g., the A-bomb. But inventions like that, where a theory is meticulously worked out then applied, probably only happen when you have to follow that path for whatever reason -- for the A-bomb, because of enormous capital expenditures. Yet there are countless inventions that came to be only because someone noticed an interesting effect and built something around it (off the top of my head, the microwave is another example), without creating a theory beforehand.