Abstract

Online collaboration communities (OCCs) enable geographically distributed individuals, groups, and organizations to self-organize and contribute to community-owned artifacts. The significance of these artifacts is underscored by recent advancements in large language models, which leverage community content for training sophisticated models across diverse domains, including productivity, healthcare, and education. This study investigates star contributors—individuals making disproportionately large contributions to focal OCC artifacts. Drawing on theories of collective action and strategic interactions, we hypothesize a curvilinear relationship between star contributors’ contributions and both artifact quality and popularity. Utilizing data from over 21,000 open-source software projects between 2015 and 2019, we find: (a) an inverted U-shaped relationship between the number of star contributors and artifact quality; (b) an inverted U-shaped relationship between the number of star contributors and artifact popularity; (c) that a higher proportion of star contributors’ contributions enhances artifact quality but not popularity; and (d) that environmental dynamism moderates the relationship between the number of star contributors and both artifact quality and popularity. This research advances the conceptualization of star contributors, offering a more nuanced understanding aligned with the fluid boundaries of OCCs compared to traditional core-periphery models. A key implication is that while star contributors positively impact artifact quality and popularity, an excessive proportion of their contributions negatively affects artifact quality.

DOI

10.17705/1jais.00970

Download

Share

COinS