You sometimes have to make a mess before you can get it to go right

Jocelyn Bell Burnell talks about the logic of Science in this episode of Beautiful Minds. The following is my transcription:

Way back in the Middle Ages, they thought that planets went round the sun in circles. Perfect circles. They had to be perfect. They were heavenly bodies. And then they got better telescopes, better data, and they recognised that the planets weren't where they expected them to be. They weren't in the right place. But they were reluctant to relinquish the circles. So they invented epicycles -- little circles on the rim of the big cirlce. Like a roundabout roundabouts. They could explain what they observed like that. Then they got better telescopes, better data, and found they had to add more epicycles and it got messier and messier and messier. And then one of the key astronomers of the time, Johannes Kepler, said maybe it's not circles, or circles plus epicycles, or circles plus epicycles with epicyles on them. Maybe it's slightly squashed circles. … And that cleared the air wonderfully. And suddenly... it was clear and simple again.

She also said: "Science always doesn't go forward. It's a bit like doing a Rubix's Cube. You sometimes have to make more of a mess with the Rubix's Cube before you can get it to go right. "