There are many open questions in this area. For example, one should extend the result on probabilistic (2,3)-generation in reference [20] to probabilistic (a,b)-generation for a wide range of values of a,b. There are also many probabilistic questions in permutation group theory and applications: for example, a well known conjecture of L. Pyber states that if G is a primitive permutation group of degree n, then G should have a base (that is, a collection of points of the underlying set, such that only the identity fixes every point in the collection) of size at most c.log |G| / log n. Various cases of this have been proved, using arguments with a strong probabilistic flavour.