tag:blogger.com,1999:blog-5337555368793819627.post4840875777460847979..comments2023-06-07T09:04:36.390-04:00Comments on More Grumbine Science: Models and ModellingRobert Grumbinehttp://www.blogger.com/profile/10783453972811796911noreply@blogger.comBlogger1125tag:blogger.com,1999:blog-5337555368793819627.post-36423983530018708532009-09-02T22:56:08.754-04:002009-09-02T22:56:08.754-04:00I think it's kind of worth noting that all sci...I think it's kind of worth noting that all science is, and all it can ever be, is a collection of models. All hypotheses and all theories ultimately are just representations of reality, which (apparently universally) get proved wrong in certain situations - i.e. Newtonian mechanics is a model that is extremely accurate at representing real world mechanics in most situations, but starts failing when you approach the speed of light.<br /><br />A model's goodness is proportional to the amount of reality it can describe. A matchbox car is not a great model of a real car, since it doesn't go anywhere. A remote control car is better, because it does, and can even go around corners, approximating the physics of a real car. A Crash test is even better, since it actually uses a real car, and the only models are the bodies, the control, and the wall. But even that isn't excellent, and it's actually impossible to create a real model car, because in reality, cars are driven by variable people with variable loads, and variable driving styles. And no model can account for all of them.<br /><br />I would even contend that climate models these days are far BETTER than Newtonian mechanics. After all, climate is innately chaotic, and climate models still manage to be able to approximate it. Newtonian mechanics completely ignores nearly all non-linearities, and there are a lot of them, even at low speeds.naught101http://eco101.wordpress.comnoreply@blogger.com