palmer eldtrich said:
So to stick with the chess board analogy, you can still record the chess game as a definite sequence of moves even if you can't specify how long each event was? Should we think of conformal time as something like that?
Yes.
palmer eldtrich said:
So if there are only massless particles, there is still conformal time ( the sequence of moves , but not poper time , how much time has elapsed between the moves).
Not quite. In order to even define a sequence of moves with only massless particles, you need massless particles that are moving in different directions, so that they intersect; the intersections are the "moves". (You can see that in the example I gave.) But it turns out that, if you have a network of massless particles moving in different directions, you can construct timelike intervals out of them, which means you can construct a notion of proper time out of them.
For example, in the scenario I described in my previous post, if we add a second light ray, D, moving in the positive ##x## direction, then we have four intersections ("moves" or events): AB, AC, DB, DC. We assume that the orderings are "AB then AC", and "DB then DC". Then the pair of events "AB, DC" defines a timelike interval, and the pair of events "DB, AC" defines a spacelike interval.
I haven't read enough about Penrose's model, the one discussed in the linked post, to know how all this affects it, if at all.