Peter Haas’s paper “Temporally-Biased Sampling for Online Model Management” has been selected as the best paper of EDBT 2018.