How Much is the Whole Really More than the Sum of its Parts? 1 + 1 = 2.5: Superlinear Productivity in Collective Group Actions

Abstract

In a variety of open source software projects, we document a superlinear growth of production ($R\sim c^{\beta }$) as a function of the number of active developers $c$, with $\beta \simeq 4/3$ with large dispersions. For a typical project in this class, doubling of the group size multiplies typically the output by a factor $2^{\beta }=2.5$, explaining the title. This superlinear law is found to hold for group sizes ranging from 5 to a few hundred developers. We propose two classes of mechanisms, {\it interaction-based} and {\it large deviation}, along with a cascade model of productive activity, which unifies them. In this common framework, superlinear productivity requires that the involved social groups function at or close to criticality, in the sense of a subtle balance between order and disorder. We report the first empirical test of the renormalization of the exponent of the distribution of the sizes of first generation events into the renormalized exponent of the distribution of clusters resulting from the cascade of triggering over all generation in a critical branching process in the non-meanfield regime. Finally, we document a size effect in the strength and variability of the superlinear effect, with smaller groups exhibiting widely distributed superlinear exponents, some of them characterizing highly productive teams. In contrast, large groups tend to have a smaller superlinearity and less variability.

arXiv comment29 pages, 8 figures

References in corpus (17)

• Power-law distributions in empirical data
• Robust dynamic classes revealed by measuring the response function of a social system
• Discrete Hierarchical Organization of Social Group Sizes
• Sub-critical and Super-critical Regimes in Epidemic Models of Earthquake Aftershocks
• Quantifying reflexivity in financial markets: towards a prediction of flash crashes
• Endogenous Versus Exogenous Shocks in Complex Networks: an Empirical Test Using Book Sale Ranking
• Importance of direct and indirect triggered seismicity
• Empirical Tests of Zipf's law Mechanism In Open Source Linux Distribution
• The Dynamics of Book Sales: Endogenous versus Exogenous Shocks in Complex Networks
• Power Law Distributions of Offspring and Generation Numbers in Branching Models of Earthquake Triggering
• Anomalous Power Law Distribution of Total Lifetimes of Branching Processes Relevant to Earthquakes
• Quantification of deviations from rationality with heavy-tails in human dynamics
• Generation-by-Generation Dissection of the Response Function in Long Memory Epidemic Processes
• Super-linear scaling of offsprings at criticality in branching processes
• Generating Functions and Stability Study of Multivariate Self-Excited Epidemic Processes
• Hierarchy of Temporal Responses of Multivariate Self-Excited Epidemic Processes
• Predicted and Verified Deviation from Zipf's Law in Growing Social Networks

Cited by in corpus (11)

• Given Enough Eyeballs, All Bugs Are Shallow? Revisiting Eric Raymond with Bug Bounty Programs
• The non-linear health consequences of living in larger cities
• Super-linear scaling of offsprings at criticality in branching processes
• Micro-foundation using percolation theory of the finite-time singular behavior of the crash hazard rate in a class of rational expectation bubbles
• Massive Multi-Agent Data-Driven Simulations of the GitHub Ecosystem
• Aristotle vs. Ringelmann: On Superlinear Production in Open Source Software
• A Theory of Discrete Hierarchies as Optimal Cost-Adjusted Productivity Organisations
• The dynamics of leadership and success in software development teams
• Computational Diplomacy: How "hackathons for good" feed a participatory future for multilateralism in the digital age
• Collective Intelligence Heuristic: An Experimental Evidence
• A collaborative path to scientific discovery: Distribution of labor, productivity and innovation in collaborative science