A Guide to Natural Naming

Abstract

The naming scheme described in this paper is a set of hints and guidelines on how to select names in a procedural programming language. The analysis of the structure of identifiers shows that type identifiers are the core of most names. Therefore the starting point for a meaningful naming is the set of names for the types within a program. Short class names should be chosen for these. The other identifiers can now be derived from the type names. Additional hints are given to complete the naming scheme. This naming convention is perceived as a real aid in finding good names, not as a restricting set of rules. It has been taught and used in projects with success in the past two years. The resulting programs are highly readable. © 1990, ACM. All rights reserved.