> I assumed that reproducing Newtonian gravity in the classical limit GR itself is a classical (i.e., non-quantum) theory. The Newtonian limit of GR is not usu

> I assumed that reproducing Newtonian gravity in the classical limit GR itself is a classical (i.e., non-quantum) theory. The Newtonian limit of GR is not usually considered a "classical" limit, just a weak field, slow motion limit. > would be considered theoretical Ok, then we just have different definitions of "theoretical". To me, having to account for a mass of experimental data is an experimental constraint. It might be easier for a theory to meet it if the theory can be shown to reduce in an appropriate limit to a previous theory that accounts for all that data. By contrast, for researchers investigating possible quantum theories of gravity, any quantum constraints on their theories can only be theoretical, since we have no experimental data at all on quantum aspects of gravity. But to me, the fact that a quantum theory of gravity has to account for all of the classical experimental data on gravity is an experimental constraint (which any such theory would presumably meet by reducing to GR in the classical limit). > Such a theory is not "useful", for many definitions of useful. No theory is useful for all definitions of useful. > Before launching the GPS constellation, we launched the experimental Timation satellites to act as a proof of concept. As I understand it, the Timation satellites did not keep independent time using clocks aboard the satellites. They just re-broadcast time signals from the ground. One of the major difficulties with getting GPS operational was the additional complexity involved in having independent clocks on the satellites that would keep their own time, with only periodic updates from the ground. And one of the major issues with that was that many non-scientists involved in the GPS project did not understand relativity and did not believe that the "natural" clock rates of the clocks on the satellites would be different from those of ground clocks, so some kind of adjustment would have to be made. And there was no data from previous projects to convince them since the relative clock rates of previous satellites vs. ground clocks had not been measured. If the non-scientists in the GPS project had not given in to the scientists who wanted to include the capability for adjusting the clock frequency in the initial set of GPS satellites, those satellites would indeed have been useless.