Fluorescence time traces are used to report on dynamical properties of biomolecules. The basic unit of information drawn from these traces is the individual photon. Single photons carry instantaneous information from the biomolecule, from which they are emitted, to the detector on timescales as fast as microseconds. Thus from confocal microscope it is theoretically possible to monitor biomolecular dynamics at such timescales. In practice, however, signals are stochastic and in order to deduce dynamical information through traditional means-such as fluorescence correlation spectroscopy (FCS) and related techniques-fluorescence signals are collected and temporally auto-correlated over several minutes. So far, it has been impossible to analyze dynamical properties of biomolecules on timescales approaching data acquisition as this demands that we estimate the instantaneous number of biomolecules emitting photons and their positions within the confocal volume. The mathematical structure of this problem demands that we abandon the normal ("parametric") Bayesian paradigm. Here, we exploit novel mathematical tools, namely Bayesian nonparametrics, that allow us to deduce in a principled fashion the same information normally deduced from FCS but from the direct analysis of significantly smaller datasets starting from individual photon arrivals. We discuss the implications of this method in helping dramatically reduce phototoxic damage on the sample and the ability to monitor out-of-equilibrium processes.