Software engineering has taken many different techniques for coordinating development work. Each technique has proven brittle, sometimes working well, other times not at all. The strategy of modular product structure uses the architecture to give form to the coordination problem that the development organization must solve. The strategy establishes design rules that fix module functionality and interfaces, and assigns development work for each module to a single team. The modules present relatively independent coordination problems that teams attempt to solve with all the traditional coordination mechanisms available to them. The applicability and effectiveness of this strategy is limited with increasing technical and organizational volatility. I present a theory of coordination, based on decision networks, that generalizes the modularity strategy. I review evidence testing several hypotheses derived from the theory, and explore how this theoretical view can drive coordination research and provide a theoretical basis for practical techniques to assist architects, developers, and managers.
This talk is a version of an invited keynote Jim presented at the FSE conference last year.
James Herbsleb is a Professor in the Institute for Software Research in the School of Computer Science at Carnegie Mellon University, where he serves as Director of the PhD program in Societal Computing. His research interests lie primarily in the intersection of software engineering, computer-supported cooperative work, and socio-technical systems, focusing on such areas as geographically distributed development teams and large-scale open source development. He holds a PhD in psychology, and an MS in computer science. He also holds a Juris Doctor degree and is a member of the Michigan Bar Association.
His research has won several awards at ICSE, including the Most Influential Paper award in 2010, Honorable Mention for Most Influential Paper award in 2011, and ACM Distinguished Paper Award in 2011. Other awards include an ACM Distinguished Paper award for ESEM 2008 and a Best Paper Award for CSCW 2006. He was recently awarded the SIGSOFT Outstanding Research Award in 2016, and previously the Alan Newell Award for Research Excellence in 2014.