> points of reference such that it does or doesn't It's too bad (but not surprising) that you couldn't have gotten a bit deeper into general relativity where e

> points of reference such that it does or doesn't It's too bad (but not surprising) that you couldn't have gotten a bit deeper into general relativity where even the idea of "points of reference" gets a lot tricker. Even though there's a lot of pop-science talked about "space time" I find most people don't really grasp how really challenging of an idea it truly is. Newtonian physics and special relatively are quite happy to imagine one observer looking at another from a particular frame of reference. The classic example is of course watching someone driving while you are in a train. Special relativity seems really crazy at first because you realize the length of the car can change if either of them are going close to the speed of light. But GR does away with this idea that you can understand the universe at all while sitting on a train. In GR frames of reference only exist locally in infinitesimally small units of space and time and are defined by being simple such that all particles being observed move in straight lines. The big insight of GR is that the question "which 'revolves' around the other" is the wrong question because the only way to understand the universe is in slices of space and time where everything moves straight (well you can't ever get 'everything' straight, so even then you just worry about getting it right for a few particles at time). In GR when you jump forward and land on the ground in front of you, you moved in a straight line the entire time. This sounds impossible because we can't imagine a single frame of reference where we could observe this. And that's what makes GR so hard to understand because the jump forward can only be understood by looking at the way those infinitesimal local reference frames change, but there is no general frame of reference that we can find where we can clearly observe this. In the exact same way that calculus understands a curved line as an infinite sequence of perfectly straight lines, GR understands curved space time as an infinite sequence of perfectly straight frames of reference. It is, in fact, much more complex than the simple case of trying to understand who is right from the two reports of two observers seeing each other on passing trains.