I have a theory about fast variable G, (a few approaches) but almost no comments on this. https://www.physicsforums.com/showthread.php?t=138968 One person give a comment: "The author doesn’t define the vacuum state, which is different for different observators and therefore he can’t built in a properly way the Hilbert space for the quantum gravity. Even more childish is to set simple relation between the average value of G tensor between Phi states, when we don’t have the transition rules between vacuum state and multiparticle states. This thing is very complicated in “flat” space in QED and what about QG? We can set the Cauchy problem in a proper way in classical gravity, but this problem is much more complicated. The similar problem is in the QCD, where doesn’t exist any Hilbert space (isn’t known the way how to build it), but there are very important results got from it, so the theory (or just the way of doing the calculations) is valid (or important) anyway. Thus, if the basic relation is not valid (we can’t simply just do it, as we do it in QM,we don’t whether QM formalism is valid in QG or not, unless we prove that it is valid), so the whole relation is invalid." But he commented only one approach and he missed that this approach (of Mark Hadley) is only symmbolic so it is not in contraction with comment above. But I have still other arguments against this. I hope for better comments about variable G.