Non-classical logics, and in particular many-valued logics, are increasingly used in the study of formal aspects of computing. or example, a recent paper by P.F. Gibbins in this Journal presents a 3-valued propositional logic for VDM. In the use of such logics one naturally relies on earlier work done by logicians, a case in point being Gibbin's use of the concept of autodescriptivity, introduced by N. Rescher. The purpose of this Note is to sound a warning that Rescher's exposition of autodescriptivity is seriously flawed, and to clarify the autodescriptivity of the logic of VDM.