Applying our knowledge to the future is an essential part of science. With the help of science we constantly seek to make significant predictions and rational decisions on how to act. This also applies to the theory of science. By seeking a general understanding of how scientific knowledge develops, the theory or methodology of science helps formulate guidelines for future scientific efforts. Methodology can hardly avoid giving advice or influencing decisions about which research projects to pursue and which to drop, both on the level of the individual research worker and on the level of national or global research policy. This influence need not be direct through the application of explicit methodological rules, but can be indirect through contributions to the conceptual schemes in which scientists, planners, politicians, and everybody else think about science. Nevertheless, analytical philosophy of science has in recent decades tended to narrow the field of rational methodology to the evaluation of fully formulated theories in the light of available evidence. This philosophy of science has concentrated on explaining the justification of finished theories rather than the rationale of their discovery or creation. There are important exceptions to the trend, like N.R. Hanson, who explicitly sought a theory of science that covers the "context of discovery" as well as the "context of justification. ''1 But Karl Popper, for example, in his classic work The Logic o f Scientific Discovery, distinguishes sharply between "the process of conceiving a new idea, and the methods and results of examining it logically." He claims that only the latter belongs to "the logic of knowledge", while the former is merely a matter for "psychology of knowledge. ''2 On this view the heuristic rules on how to make interesting dis-