Michael Klug and James P. Bagrow: “Massive datasets describing the activity patterns of large human populations now provide researchers with rich
opportunities to quantitatively study human dynamics including the activities of groups or teams. With the rise in prominence of network science [5, 6], much effort has gone into discovering meaningful groups within social networks and quantifying their evolution. Teams are increasingly important in research and industrial efforts and small, coordinated groups are a significant component of modern human conflict. Recently, there has been much debate on the “group size hypothesis,” that larger groupsare more robust or perform better than smaller ones. In scientific research, scholars have noted for decades that collaborative teams have been growing in size and importance. Case studies show that leadership and experience are key components of successful team outcomes, while specialization and multitasking are important but potentially error-prone mechanisms for dealing with complexity and cognitive overload. New tools, including electronic sensor systems, can quantify team activity and performance. In all of these areas, large-scale, quantitative data can push the study of teams forward. Users of the Github web platform can form teams to work on real-world projects, primarily software development but also music, literature, design work, and more. A number of important scientific computing resources are developed through Github, including astronomical software, genetic sequencing tools, and key components of the Compact Muon Solenoid experiment’s data pipeline. A “Github for science” initiative has been launched and Github is becoming the dominant service for open scientific development. Github provides rich public data on team activities, including when new teams form, when members join existing teams, and when a team’s project is updated. These open collaborations evolve in an entirely self-organized manner and are not driven by hierarchical management as one may encounter at a commercial organization. Github also provides social media tools for the discovery of interesting projects. Users who see the work of a team can choose to flag it as interesting to them. This allows us to accurately quantify the impact of the team, in a manner analogous in many ways to citations of research literature. In this study, we analyze the memberships and activities of approximately 150,000 self-organized teams, as they perform real-world tasks, to uncover the blend of features that leads to success. To the best of our knowledge this is the largest study of real-world team performance to date. We present results that demonstrate (i) how teams distribute work activity across their members, (ii) the mixture of experiential diversity and collective leadership roles in teams, and (iii) how high-impact teams are different from other teams.”