# Epicycles In /The Structure of Scientific Revolutions/<> notes how any time a theory starts to add awkward contortions in order to match observed data it suggests that it’s time for a paradigm shift. The idea that many wildly successful products and technologies are ‘solving problems people didn’t know they had’ is one of these contortions. Instead of thinking about solving problems, we should think about removing constraints and enabling superpowers. This perceptual shift isn’t just semantic - it suggests precise ways of exploring new ideas. Ancient and medieval astronomers modeled the solar system as planets embedded in a set of rotating, concentric spheres centered on Earth. The astronomers were actually quite good at getting this model to fit the data of how planets travelled through the sky. Most of the time planets move forward through the sky, but occasionally they go backwards for a bit before resuming their path. This ‘retrograde motion’ wouldn’t happen if they were embedded in a constantly rotating sphere. To explain this little loop-dee-loop, the Greeks created the concept of ‘epicycles’ wherein the planets were actually on little spheres embedded in the bigger spheres. Every time a planet did something inconsistent with the large sphere they added an epicycle. Epicycles were the glaring exception to the otherwise neat theory that eventually led Copernicus, Galileo, and Brahe to the Sun-centric model of the solar system. This new model both unlocked new knowledge (the discovery of the outer planets) and abilities (sending probes to Mars.)  - - - ### Related * [[kuhnStructureScientificRevolutions1962]] <!-- #mental-model --> [Web URL for this note](http://notes.benjaminreinhardt.com/Epicycles) [Comment on this note](http://via.hypothes.is/http://notes.benjaminreinhardt.com/Epicycles)