Complex, Intelligent, and Software Intensive Systems

Abstract

Abstract— Open source software represents a new paradigm of software development based on a subjacent community. It is widely accepted in the literature the layered structure of open source communities, being the core group the most active contributors usually located at the center of the community. The tasks of this group include not only an intense activity in terms of contributions but also to promote participation among the rest of the community members. In this paper, the general role of this group is analyzed by modeling communities as Social Networks and applying Social Network Analysis techniques. Findings related their brokerage activity with open source software success. Keywords-wireless Open source communities; Virtual communities; Social Network Analysis; brokerage role. I. I NTRODUCTION Open Source Software (OSS) projects constitute a prominent example of software development based on communities [1]. They are supported by a community of developers and users who work in geographically distributed locations, rarely or never meet face to face, and coordinate their activities making an intense use of Internet. The wide success of several well known Open Source Software (OSS) projects has attracted much attention. Software engineering researchers and commercial companies alike have been trying to learn lessons from the success of OSS and apply some of them to the development of proprietary software [2]. However, there is a cruel reality that the vast majority of OSS projects fail to take off and soon become abandoned [3]. According to the popular open source portal, SourceForge (http://sourceforge.net/), most OSS projects have ended in failure: 58% do not move beyond the alpha developmental stage, 22% remain in the planning phase, 17% remain in the pre-alpha phase, and some become inactive [4]. Previous studies on OSS have been mainly focused on motivation of people participating in subjacent virtual communities. Motivations have been found to range from problem solving [5] [6], to reputation benefits and career opportunities [7] [8] [9]. But a more reduced amount of work has been focused on communities’ structure. It has been demonstrated that much of the OSS development is realized by a small percentage of individuals despite the fact that there are tens of thousands of available developers. Such concentration is called “participation inequality” [10], [11], and it can be explained by the different user profiles of open source communities. Consequently, the structure of OSS communities is not completely flat as it was claimed by the bazaar model of full participation [12]. II. OSS C