We study the computational complexity of the valued constraint satisfaction problem (VCSP) for all valued structures over the domain Q preserved by all order-preserving bijections. Such VCSPs will be called temporal, in analogy to temporal constraint satisfaction problems. Many optimization problems that have been studied intensively in the literature can be phrased as a temporal VCSP, for example, the min correlation clustering problem or the minimum feedback arc set problem. We present a P vs. NP-complete dichotomy for temporal VCSPs, based on the concepts of fractional polymorphisms and expressibility in valued structures. This is the first dichotomy result for VCSPs over infinite domains which is complete in the sense that it treats all valued structures that contain a given automorphism group. This is joint work with Manuel Bodirsky and Édouard Bonnet.