scientific article; zbMATH DE number 3984558

From MaRDI portal
Publication:3747711





The term 'software engineering' was first coined in the 1960s in response to the realisation that the activity of software construction could be viewed as an engineering discipline. This view enabled a clearer insight into the shortcomings of software construction methods by encouraging comparisons with other branches of engineering. These shortcomings can to a large extent be traced to inadequacies in programming language support for common engineering techniques. Despite some progress in recent years several major areas of difficulty remain. Those particularly focused on this thesis are: the lack of a powerful, general model for sub-components in complex systems, the poor understanding of collections of objects and their internal structures, and confusion concerning the relationship between design and implementation. The thesis examines the requirements for a general module construct and it is argued that the module construct of Leibniz fulfills these requirements to a greater degree than the module constructs of other languages. In particular more uniformity is achieved in the definition of different kinds of software units. Files (including standard input and output) are seen simply as instances of abstract data types. Records are also seen simply as module type, allowing the definition of operations on these types.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3747711)