“…Roughly speaking the choice of such a direction corresponds to a control mode of the fields at the boundary which is based on physical grounds (e.g. micro-canonical or grandcanonical ensembles) and it is related to the asymptotic behaviour of specific solutions and to the specific physical observable one is aimed to determine (see, e.g., [4,9,16,19,40,50,53]). We then believe that, inside a formalism which we want to be as general as possible, both physical and mathematical reasons suggest to us that only variations δQ should be reasonably considered.…”