It is argued here that the discipline of information systems does not have a clear and substantive conceptualization of its most fundamental category, namely, information itself. As a first stage in addressing the problem, this paper evaluates a wide range of theories or concepts of information in order to assess their suitability as a basis for information systems. Particular importance is placed on the extent to which they deal with the semantic and pragmatic dimensions of information and its relation to meaning. It is concluded that Dretske's analysis of knowledge and information provides the most suitable basis for further development.