We have a secure if still schematic standard model for cosmology, and the prospect for considerable enlargement from the application of the cosmological tests. The theoretical basis for the tests was discovered seven decades ago. A significant application likely will take a lot less than seven more decades: the constraints in Table 1 already are serious, if debatable, and people know how to do better.
Application of the tests could yield a set of tightly constrained values of the cosmological parameters and a clear characterization of the primeval departure from homogeneity. If so cosmology could divide at a fixed point, the situation at z = 1015, say, when the universe is well described by a slightly perturbed Friedmann-Lemaître model. One branch of research would analyze evolution from these initial conditions to the present complex structure of the universe. The other would search for the physics of the very early universe that produced these initial conditions. But before making any long-term plans based on this scenario I would wait to see whether the evidence really is that the early universe is simple enough to allow such a division of labor.
I am grateful to the organizers for the invitation to this stimulating meeting. The work in this paper was supported in part by the USA National Science Foundation.