Flow Diagrams, Assertions, and Formal Methods

摘要

This paper examines the early history of the flow diagram notation developed by Herman Goldstine and John von Neumann in the mid-1940s. It emphasizes the motivation for the notation's mathematical aspects and the provision made for formally checking the consistency of diagrams. Goldstine and von Neumann's introduction of assertion boxes is considered in detail. The practical use of flow diagrams is considered briefly, and the paper then reads Turing's 1949 essay on "Checking a large routine" in the light of his likely knowledge of the Goldstine/von Neumann notation. In particular, his different use of the term "assertion" is considered, and related to the earlier work.