In theory, cars are fairly simple. If they don't start, it's either the fuel system or the electrical system. Teach yourself about the path of each in your engine and tracing it is fairly straightforward. But at the beginning, mastering each new system seems like an unreachable shore. The car is effectively a black box.

My advice is keep your lips away from the spinning things.

In the summer of 2002, we had spent six weeks shooting the three pilots of Mythbusters, and Jamie[Hyneman] called me up afterward - well, first he called me up to tell me to clear my crap back out of his shop - and he said, "Well, that was kind of fun, wasn't it? I mean, I don't see where this could go, because we pretty much did everything. But it was fun."

In the spirit of science, there really is no such thing as a 'failed experiment.' Any test that yields valid data is a valid test.

Growing up in New York, I was sort of shocked when I realized that my children are Californians. They are 14 years old, and I explain to them frequently that they will never realize the glory of a snow day. You wake up and the world says, 'Oops, it's too much fun to go to school, you've got to stay home and deal with the snow!'

We don't necessarily stand by our faults every time, but we will always stand by our methodologies and ethos.

Back when I was a professional model-maker at Industrial Light & Magic, my specialty was hard-edged construction - spaceships, miniature sets, and architectural stuff. These objects were sometimes just 12 inches across yet needed enough detail to fill a movie screen.