A reusable, academic-strength, metrics-based software engineering process for capstone courses and projects

  • Authors:
  • Richard Conn

  • Affiliations:
  • Kennesaw State University and Microsoft Corporation, Redmond, WA

  • Venue:
  • Proceedings of the 35th SIGCSE technical symposium on Computer science education
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper describes a mature Software Engineering Process that may be applied to capstone courses, student projects, and research projects in a university environment. This process, based in part on the Team Software Process of the Software Engineering Institute, features mature software engineering best practices, including extensive use of metrics to gain insight into process effectiveness and product quality. It is designed to be executed in a single 16-week semester, and it can easily be modified for a longer time frame. It is also designed for the academic culture: (1) it executes as a communicating sequential process running in parallel with and passing information to a series of in-class course presentations and (2) it takes into account that the students cannot devote full time to this one project. This process is independent of the computer language used and any project-specific technology (so it may be applied to software-only projects, robotics projects, smart device projects, embedded programming projects, and any other kind of software-intensive system project). This process is reusable without modification, or it may be tailored. To support tailoring, the process is described using meta data in a format that may be analyzed by automated tools to generate a hypertext document. This process has been tested by using it in a capstone course.