ITR: Collaborative Research Proposal for a National Center for Empirical Software Engineering Research

点击翻译按钮获取中文摘要

A necessary step towards the goal of building more reliable software systems, on time and within budget, is to establish an institutionalized empirical discipline for understanding causal relationships among the processes, components, and technologies that affect the building of systems. As in the physical and natural sciences, experimentation in software engineering requires a community with support for collaboration, experimental replication and refinement, and sharing of experimental data and results.For these reasons the Center for Empirical Software Engineering Research (CESER) undertakes original empirical research and is developing a prototype system for sharing and evolving the results of such research with a community of affiliated researchers and practitioners. CESER develops and refines techniques to increase the descriptive and predictive power of empirical models, and studies specific software development technologies to enable industrial organizations to understand the benefits and drawbacks of those technologies in their specific context. The Center provides courses and symposia on empirical methodologies and results, and assists the use of empirical knowledge in software engineering education. The Center's initial focus is on empirical studies of software COTS integration and software quality improvement phenomenology.The center is initially organized as a collaborative effort among the University of Maryland, the Fraunhofer Center - Maryland, the University of Southern California, the University of Nebraska at Lincoln, and Mississippi State University.