Conceptual content management for software engineering processes

  • Authors:
  • Sebastian Bossung;Hans-Werner Sehring;Michael Skusa;Joachim W. Schmidt

  • Affiliations:
  • Software Technology and Systems Institute (STS), Hamburg University of Science and Technology (TUHH);Software Technology and Systems Institute (STS), Hamburg University of Science and Technology (TUHH);Software Technology and Systems Institute (STS), Hamburg University of Science and Technology (TUHH);Software Technology and Systems Institute (STS), Hamburg University of Science and Technology (TUHH)

  • Venue:
  • ADBIS'05 Proceedings of the 9th East European conference on Advances in Databases and Information Systems
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

A major application area of information systems technology and multimedia content management is that of support systems for engineering processes. This includes the particularly important area of software engineering. Effective support of software engineering processes requires large amounts of content (texts, diagrams, code, data, executables etc.) from different conceptual domains. The term “software crisis” disappeared gradually when content modelling and management addressed domains from application analysis and system design in addition to the sheer computational code domain. In this paper we introduce an innovative conceptual content model and apply it in support of software engineering processes and their artefacts. We base our approach on the core model of the computational domain which abstracts computational content (bodies of function code) by the computational concept of signatures (lists of typed function parameters). We generalise this functional abstraction model beyond the computational domain by introducing the notion of asset abstraction which models entities domain-independently by general content-concept pairs. We introduce an asset language and discuss the essentials of an asset system implementation. In the application part of the paper we argue that software engineering can be substantially simplified by modelling SE entities from all the domains involved in an SE process homogeneously in an asset-oriented approach—entities ranging from application domains over intermediate architectural and design domains down to the computational domain. Furthermore, we discuss how the mappings between such domains can be substantially supported by services based on asset-oriented information systems.