Was it fundamentally wrong? Or did we just find another model that: 1. Accurately predicts phenomena the old model predicted accurately and predicts things we

Was it fundamentally wrong? Or did we just find another model that: 1. Accurately predicts phenomena the old model predicted accurately and predicts things we hadn't expected from the previous model or from our data that we later measure to be correct. 2. Effectively describes phenomena previously thought to be unrelated to the problem of interest. 3. We appreciated the simplicity of. (i.e. We could come up with more ad-hoc silliness that makes the old model accommodate our new data, but why?) What's interesting about Ptolemy is that in a sense there is nothing inaccurate about his model. In my reading, it could hardly be called a model so much as a technique for creating models. Extending it to account for any deviations from our empirical measurements is trivial: just add another epicycle. The modelling technique Ptolemy demonstrates in his Almagest, seems to me to just be the Fourier transformation of a periodic function.