CAREER: A Study of Open Source Software Project Success

University Of Maryland, College Park, College Park MD

Investigators

Abstract

The PI will develop and test a model of the influences of team, project, and user factors on success in open source software (OSS) development. A multi-dimensional conceptualization of success will be explored, including both a development and a usage perspective. The following general propositions are developed: Team factors that enhance identification and commitment attract and retain developer input and facilitate the successful conversion of that input into output; Project factors that are aligned with developer motivations attract developer input; Project factors that facilitate the coordination and combination of developer contributions increase the successful conversion of development input into output; Project factors act as cues to usefulness and ease of use of the software, thereby influencing user interest and adoption; The alignment of project factors with user factors influences user interest and impact; Positive impact of prior OSS adoption will enhance user interest in other OSS products. These propositions are further developed to provide the basis for testable hypotheses. Data will be collected on a large sample of open source projects using several sources: community websites (SourceForge.net and Freshmeat.net), project homepages, surveys of project participants, and case studies and surveys of users. Projects in the sample will be monitored over the proposed research period, so that both cross-sectional between-projects effects and longitudinal within-project effects can be analyzed. To start the process, data collected in three pilot studies previously conducted by the PI to refine her theoretical model and data collection strategies will be integrated and significantly expanded. The PI expects this work will advance our knowledge and understanding of OSS as a unique and important phenomenon by providing a large sample investigation of influences on development success, and by introducing a usage perspective on success that has been lacking in prior work. Using a large sample of newly introduced projects and following them over time, this study will complement prior work by addressing limitations that have resulted from studying mainly large, successful OSS projects and/or limiting study to cross-sectional analysis of publicly available data (i.e., SourceForge data). This study combines both publicly available data from OSS websites as well as survey data and case studies, to provide a richer picture of success. Examining team, project, and user factors in a single study will allow for interaction effects and comparison of the relative importance of different kinds of success antecedents. The research model is grounded in emerging research on OSS and relevant organizational research literatures, investigating both factors unique in the OSS setting, as well as factors drawn from other studies of virtual teams. The research is expected to contribute not only to our understanding of OSS, but also to the wider research literatures from which the model is derived, including the information systems (IS) literatures on virtual teams and IS adoption and use. Broader Impacts: Open source software represents a new means of innovation that may have great societal benefits. The PI expects the results of the research will enable organizations to effectively use open source software and development approaches, and will further help policy makers to determine appropriate means for fostering innovation. The PI will widely disseminate the results of this research both in academia and industry. The project will directly promote learning through the participation of undergraduate and graduate students in the study activities; in particular, as three of the four students who worked with the PI on her pilot studies (and who she expect to continue working with her) are African Americans, the research may directly increase the participation of under-represented groups.