A reason (beyond hardware limits) math cannot model weather?
Name:
Anonymous2016-09-20 3:54
Is the only reason that math cannot model weather far in advance (among other things such as boiling water patterns) because it would require too much computational power? Or is there more to it? Is weather, et al too "organic" and thus math is not applicable to it?
Name:
Anonymous2016-09-20 4:40
The weather is determined by the gods. Since gods are unpredictable and emotional, software cannot model their intentions.
Name:
Anonymous2016-09-20 6:14
Weather can't be organic because of Global Warming. I need my gluten-free hurricanes, shitlords.
Name:
Anonymous2016-09-20 6:51
>>1 computational power is one issue, the other is that weather is a chaotic process influenced by data we don't have access to or even (possibly) things we don't yet know about. to put it in perspective: I don't think there's any hard division between 'organic' and 'inorganic' stuff which would mean that it should be possible to algorithmically model a human brain. this does not mean we could easily make an algorithm to predict what your next /prog/ post be about.
Name:
Anonymous2016-09-20 13:43
This is the sort of thing that can only be explained by Jeff Goldblum.
Name:
Anonymous2016-09-20 14:23
According to chaos theory, first you must accurately model all the butterflies
Name:
Anonymous2016-09-20 18:27
This is much harder than predicting the stock market exactly, which is already proven to be impossible because even that's too stochastic. I think there are mathematical limits (chaos theory) that prevent predicting weather to ever be possible, regardless of technology.
Name:
Anonymous2016-09-20 19:13
Numerical instability in calculating differential equations.
Name:
Anonymous2016-09-20 20:44
>>7 But can any variable really contribute to stochastic processes? How can deterministic laws produce truly random processes? Doesn't this just indicate that our models of physical reality are incomplete, and we have simply not yet identified the underlying patterns that can be precisely measured and predicted? Perhaps what is needed is more powerful computers.
Name:
Anonymous2016-09-20 20:53
>>9 How can Schrödinger’s equation be completely deterministic and yet produce cats in boxes?
>>10 Schrödinger’s cat in box trick illustrates the absurdity of using probability to explain anything. The interpretations rely on pretzel-like linguistic tricks to try and make it seem plausible that the act of measurement creates something measurable out of a set of probabilities. Whether you subscribe to "many worlds" interpretation or Copenhagen, you are still left having to assume that human perception has a creative role to play in reality. But this is simply an interpretation of a set of equations, except it is more than that because it claims the interpretation can make an objective "judgement" on the equations. If the interpretation creates the reality underlying the mathematical model, then you are entering into the realm of philosophy and using math tricks to justify thought-based assumption. It gets all back to front. Suddenly math is creating concepts that cannot be tested or proven, yet one must accept that these concepts are inherent to reality or the entire framework collapses. That isn't right. If math leads to concepts indistinguishable from spiritual and philosophical thought, then math is not as valuable or reliable a tool for science as previously thought. It is this situation that has led me to abandon traditional math entirely. Now I use the Daoist tradition of The Number. All my math now only uses one symbol, the essential symbol: 9. Through removing all other symbols, I am able to develop a framework of reality so precise and accurate that a word-based interpretation is impossible. We are reaching the very essence of reality here, people.
>>14 Chaos "theory" (more like "placeholder") is itself the result of computers, a product of hardware limits rather than a law unto itself. Prior to computers, for example, people thought the mathematical monsters of fractal geometry were inconceivable and could never be viewed as shapes. Now people think that divergence patterns are inconceivable and can never be measured, but what if we really just needed a more powerful computer?
>>9 unpredictable does not imply truly random. functions which take multiple arguments and their outputs vary wildly and unpredictably with small chnages in those variables might be completely deterministic but they may appear random when you don't have full insight into their complete definitions and/or their arguments.
>>16 all theories in natural sciences are by definition placeholders and all theories in formal sciences are based on unverifiable axioms.
Name:
Anonymous2016-09-21 17:01
>>16 It's the initial conditions that can never be fully measured. The divergence patterns are just accumulated error because your model isn't entirely accurate.
Name:
Anonymous2016-09-23 20:30
>>1 To extrapolate weather, one needs a lot of sensors on Earth. The more sensors you have the more precise is prediction.
you are still left having to assume that human perception has a creative role to play in reality
It's called Magic. You wouldn't dare claim that all the druids of yore, Qabalists, voodoo people and Tibetan monks changing reality with their willpower were/are lying, would you? Reality is not real so of course human perception (as well as imagination) changes it, even if science dismisses magic as "coincidence".