At first, it was a very practical thing. I always believe in using the best tools. And in the early 1970s (when I was an early teenager), I happened to get interested in physics. And I realized that the best way to figure out some things I wanted to figure out in physics was to use a computer. So I learned how to do that.

In the years that followed I became pretty good at using computers in practice to figure things out. And then at the beginning of the 1980s I decided to start building better tools for the kind of computing I wanted to do. And I realized that to do that I needed to understand more about the foundations of computation.

When I went on doing science, I then realized that I could apply what I’d learned about the foundations of computation. And the result was that I developed a new approach to science that’s very fundamentally based on computational ideas.