Coordinating Software Development through Predictive Conflict Detection

Speaker  Anita Sarma

Host  Andrew Begel

Affiliation  University of Nebraska

Duration  00:59:29

Date recorded  19 June 2013

Distributed software development poses many challenges. One of the main factors is the complexity of technical dependencies existing in the code base, which leads to complex social-technical dependencies among developers. This social-technical complexity inevitably leads to software conflicts because of coordination problems. Recently, we have been witnessing the development of new kinds of collaborative technology, and variations on existing technologies that support new collaborative development practices. New trends in organization of distributed, collaborative work and the development of new technologies supporting it result in an intriguing interplay of people and technology, which motivates my research in supporting coordination in software development. In this talk, I will discuss new coordination tools developed by my group. In particular I will focus on a novel conflict minimization technique that proactively identifies potential conflicts, encodes them as constraints, and solves the constraints space to recommend a set of conflict-minimal development paths.

©2013 Microsoft Corporation. All rights reserved.
> Coordinating Software Development through Predictive Conflict Detection