SciBrowser: Exploration and Analysis of the Complexity, Structure, and Activity Dynamics of Open Source Science Communities

Abstract

Open Biomedical Ontology (OBO) is a socio-technical community that is comprised of individuals dispersed geographically, but function as a coherent unit through the use of cyber-infrastructure. This study explores dynamics of open source science in such virtual socio-technical networks. Innovation within a socio-technical network can be de ned as the approach to work that leads to the generation of novel and useful ideas and processes. Among the factors that influence innovation are structural properties such as centrality, den- sity, clustering coe cient, and average path length of socio-technical networks, as well as effectiveness in collaboration. Hence, we explore virtual scienti c communities from three main perspectives: network, collaboration, and activity. Structural network metrics measure the resilience of socio-technical networks. Collaboration analysis aims to discover interaction patterns among participants and between knowledge domains. Activity analysis facilitate discerning artifact submission and community growth patterns over time. To expedite analysis, a computational ethnography tool, called SciBrowser, is introduced. Using SciBrowser, we observe power law degree distributions, which indicate presence of scale-free network con figurations. Such con gurations provide an explanation for the resilience of research communities in cyberspace. A new metric, called activity strength, suggests that major contributors of the project are weak collaborators. As a result, their strong contribution factor is nulli ed by their weak collaboration intensity. Activity patterns of the observed projects suggest the presence of an adaptive renewal cycle, which is the epitome of behavior in innovation ecosystems.