Quote by Chalnoth
Yes, but if we're thinking a photon first passing by observer A, then that photon will be observed at an earlier time than the same photon observed by observer B. Observer B will see all CMB photons from all directions at this longer wavelength, because the universe has expanded while we waited for this one particular photon passing from observer A to reach B.

We picked the locations of A and B at random (or at least I did, I didn't say which was where, nor even specified where "where" was) so how can we possibly say that B receives all CMB photons from all directions at a longer wavelength than A does?
Once we take into account the simultaneity issues (since A and B are not at rest with respect to each other due to cosmological expansion), I would imagine that the randomly located A and B would have to receive the same sort of photons, at the same sort of colour, with the same sort of distribution.
Does this not make sense?
cheers,
neopolitan