Article Preview
TopIntroduction
Highly publicized corporate involvement in open source software (OSS) development creates the illusion that volunteers are no longer that important to OSS development anymore. OSS projects such as Linux and Apache Web Server, which grew from seed planted by volunteers, are now commercialized and largely supported by business interests. These successes were followed by software donations by corporations to create new OSS projects. Sun Microsystems contributed OpenOffice, IBM open sourced Eclipse (toolkit for designing integrated development environments) and Lotus Symphony (document processing applications), and Netscape released its browser suite source code to start the Mozilla project, to name a few. As a result, many OSS developers are paid by information technology vendors, OSS foundations or companies that utilize OSS to work on OSS projects. Together, these observations support the image of a highly active, highly commercialized OSS development effort that no longer relies upon volunteers who are not paid to contribute, which we refer to as ‘volunteers’.
Although prominent projects such as Linux and Apache Web Server are largely sustained by commercial interests, they still benefit from volunteerism. For example, it is estimated that 18 - 25% of new Linux code still comes from volunteers (Smoker, 2010). Other less prominent OSS projects rely substantially on volunteers. Debian, a Linux distribution package, and Sahana, information management solutions for disaster response, are just two examples of the many OSS projects that rely heavily on volunteers. Drupal (content management platform) is an example of an OSS project that uses a balanced mix of volunteer developers and paid contributors.
OSS projects recruit volunteers not just for raw manpower, but more importantly, because of their remarkable skills and knowledge. Fang and Neufeld (2009) reviewed several qualitative studies on OSS project success and concluded that a lack of sustained volunteer participation is a major reason for OSS project failure. Their own qualitative study of OSS volunteers reveals that repeat volunteers make special contributions by taking on advising roles and making practical contributions such as code improvement. Consequently, there is considerable practical advantage in understanding what directs and sustains OSS developers’ voluntary participation for extended periods of time.
Motivation is what draws volunteers to OSS projects. Previous studies regarding OSS developers show a variety of evidence that different types of motivations are important, but there is much less understanding of the relative importance of motivational components in different contexts (e.g., work satisfaction, performance, effort, recruitment, and retention) (Krishnamurthy, 2006). A second limitation of past research is that it mainly examined simple correlations between motivators and behavior, but research on motivational structure has shown that developers’ motivations are not independent but rather are related in complex ways (Meyer et al., 2004; Roberts et al., 2006). Motivational structures need to be studied in order to better understand volunteer behavior. Thirdly, commitment, a distinguishable component of motivation, is neglected in research on OSS developers’ motivation. Commitment is a predictor of employee turnover and positively related to work performance (Meyer et al., 2004). Finally, prior studies of OSS developers’ motivations group volunteers and paid participants together for analysis. It is arguable that the motives of volunteers differ from those of paid participants as the involvement of volunteers is entirely discretionary.