Evaluating the Impact of Peer Impression on Project Outcomes in Free/Libre, Open Source Software (FLOSS) Projects

University Of Alabama Tuscaloosa, Tuscaloosa AL

Investigators

Abstract

Software development organizations that employ a distributed development model (i.e. participants are not co-located) are an important class of virtual organizations, operating as sociotechnical systems that incorporate computer-mediated communication and collaboration to enable their success. Despite the availability of various sociotechnical interaction techniques, the literature suggests that distributed teams require more time and resources than co-located teams to do the same amount of work. The lack of physical interaction among members of distributed teams hinders the process of impression formation, because individuals have difficulty obtaining an accurate view of the expertise and ability of teammates. Inaccurate perceptions may lead distributed software developers to improperly disregard teammates' input, ultimately reducing the productivity of the team as a whole. In this project we will model the process of peer impression formation in Free/Libre and Open Source Software (FLOSS) projects, one of the most common types of virtual software development organizations. The objectives of this project are 1) to build and evaluate an empirical model describing the impression-formation process, 2) to measure the accuracy of peer impressions, and 3) to observe how peer impressions affect project outcomes. This project seeks to identify and understand how various types of sociotechnical interactions (e.g. via code review, bug repositories, and code repositories) influence peer impression formation in distributed FLOSS projects. We will evaluate how peer impressions affect collaboration and project outcomes. The results of this work will improve the software development process for FLOSS projects. Therefore, this research will have broad impacts on society because of the large number of FLOSS projects that provide key software for many important tasks. This work will also contribute to building the first theoretical model of impression formation processes in FLOSS projects. In addition to their direct applicability in FLOSS project, these results will be also applicable to other types of distributed software development, which is largely similar to FLOSS development and has been adopted worldwide by many commercial organizations.