I'm not convinced that it's solely our scale that makes euclidean geometry the preferred one. I think it also has to do with the fact the the ancient Greeks di

I'm not convinced that it's solely our scale that makes euclidean geometry the preferred one. I think it also has to do with the fact the the ancient Greeks did their math with pointy sticks in beds of sand--beds that took on a certain shape due to environmental factors beyond just scale. This caused us to associate the concept of orthogonality with a 90 degree angle, which is (in most cases) a property of the representations we use--not a property of the underlying mathematical objects. It's not a bias of the coordinate systems, it's a bias of how we conceive of the space, uh, coordinated by those systems. And it's a convenient one--it evolved in our environment for a reason--it works well here. But so did the idea that the Earth was in the center of the universe. I realize that it's a kooky theory, but it takes time to turn a kooky theory into a serious one, and so far I think it has some merits so I'm going to keep working on it. Particularly, it would explain the seemingly arbitrary values of certain mathematical constants. If our concept of "straight line" or "flat surface" were to be not special at all, but just some weird quirk of how humans do things, then constants like pi would tell us less about circles and more about our own habits of perception, and I think that would be a much more satisfying way of framing things. Or think about how the derivative of e^x is e^x. That's a cool property, but if you look at how the difference quotient is constructed you'll see that it implies axes on a flat surface set at 90 degrees to each other. So maybe what's special about e says more about the spaces implied by the coordinate system. I realize that e is not _defined_ this way. That's where the theory needs work. Somehow our concept of the natural numbers is bound up with our perspective on flatness, and of how independent things ought to be represented and compared. I've yet to be able to construct something number-like that I can map to alternative values of e and pi for different geometries, but I think it can be done.