Number of points of function fields over finite fields

This is a revised and slightly expanded version. We point out that in the previous summary, "without cohomology" should really read "almost without cohomology" because of the proof of Lemma 2, that the idea to consider effective motives divisible by the Lefschetz motive was anticipated by Serre in letters to Gilles Lachaud and Marc Perret, and finally that the birational invariance of the number of points modulo q in fact follows from a 1983 Comptes Rendus note of Torsten Ekedahl (I am grateful to Antoine Chambert-Loir for making this observation and indicating this reference).