James C. McKim

Class interface design: designing for correctness

Abstract We present heuristics for the design of correct and understandable class interfaces. Following these heuristics, we design a number of rigorously specified classes. Each example class is first designed around a minimal set of fully defined features. This minimal set may be extended, possibly through inheritance, to a more complete and usable collection for a particular application. We use the Eiffel language to present the examples as it provides the most extensive support for our ideas; however, our heuristics are language independent. Our goal is to foster discussion of standards for class interface design.

Evaluating and extending NPath as a software complexity measure

Abstract The purpose of this article is to use criteria developed by Weyuker to evaluate a complexity metric, NPath, developed by Nejmeh. Weyuker has applied these criteria to several widely known complexity metrics. We apply Weyukers criteria to NPath and compare the results to those metrics already evaluated by Weyuker. Neither NPath nor any of the metrics evaluated by Weyuker satisfy all of the criteria. We note a possible weakness in NPath which Weyukers criteria do not uncover. Finally we present an untested complexity metric which satisfies all of Weyukers criteria and therefore demonstrates the consistency of said criteria.

An alternative culminating experience for master's students in computer science

Most schools that offer Master’s Degrees in Computer Science require some sort of culminating experience. Typically these experiences consist of a full fledged master’s thesis or project, an examination based on certain core courses, or a “capstone” course. The Hartford Graduate Center offers an unusual, perhaps unique, culminating experien ce. Students in the final year of their master’s program are required to take a one credit course called Computer Science Seminar. Over the course of the year the students attend six presentations by leaders in their re spective areas of computer science. Concurren tly, they work on modest sized (20 doublespaced pages) original or expository research papers. These papers are thoroughly reviewed by the faculty, and may be revised multiple times. At the end of the academic year, a conference is held at which each student presents his/her paper. Proceedings of the conference are published and a copy given to each st udent. Permission to copy without fee all or part of this material is granted provided that the copies are not made or distributed for direct commercial advantage, the ACM copyright notice and the title of the publication and its date appear, and notice is given that copying is by permission of the Association for Computing Machinery. To copy otherwise, or to republish, requires a fee and/or specific permission. 01991 ACM 0-89791-377-9191 10002-0210 . ..S1 .50

Teaching Object Oriented Programming and Design with Eiffel

The Object Oriented Paradigm is touted as a software development methodology that promotes reuse, models the problem space, facilitates maintenance, incorporates changes easily, shortens the development lifecycle, encourages good software engineering techniques, and cures the common cold. A course in Object Oriented Programming and Design should address these claims. One way (perhaps the only way) for students of the paradigm to test such claims is to build a small but high quality product as part of the course. The Eiffel language provides an ideal platform for such a course due to its simplicity, straightforward support for 00 concepts, and its assertion mechanism.