Over the decades, the increasing use of software has enabled construction and deployment of ever more capable computer-based systems. However, managing, controlling, developing, and assessing the development of software-intensive systems are complex and error-prone tasks that are made even more difficult by the lack of meaningful data on key products and processes. Such data should provide visibility into the health and status of the evolving software system, assist in early identification of current and potential problem areas, and aid in predicting such software system characteristics as reliability, maintainability, cost, and schedule. Additionally, such data are essential to process improvement activities, since they provide a means for quantifying process characteristics and assessing changes *both improvements and degradations* in these characteristics.

To enable software system metrics to provide real data for the purposes mentioned above, comprehensive data collection and analysis methods which treat software as an integral part of a larger system are needed. This presentation introduces a software system metrics approach that has been developed for this purpose. The presentation discusses the utility of the approach, introduces a set of recommended metrics that covers both the system and the software throughout the life cycle, and provides some examples of these in the area of metrics for requirements. Next, key concepts and planning guidelines for metrics programs are highlighted and the relationship of metrics to process improvement is addressed. The presentation concludes by identifying some current topics in metrics and some sources of additional information.

Biographies

Sharon K. Hoting is the Director of the Software Acquisition and Analysis Department at the Aerospace Corporation. She has over 25 years of demonstrated success in software engineering organizations in the development/management/ GSE&I of government software-intensive systems, both as a development contractor and as a government advisor. She is knowledgeable in software process, software products and their assessment, and the relationship of software to the systems engineering process and products. As a contractor, she was responsible for the development of proposals, functional management of software engineering personnel, project management of a variety of software development efforts, and technical development of many software systems and tools. She has a Bachelors degree in Mathematics and a Masters degree in Computer Science from the University of California, Los Angeles.