Is more always better? Unveiling the impact of contributor dynamics on collaborative mapping

Open, collaborative mapping initiatives such as OpenStreetMap, a wiki-style map of the world, continually face concerns about the reliability and authority of its data. Based on harnessing the power of millions of volunteers globally, the data production process is decentralized and reflects a mosaic of individual contributors’ skills, motivations, and experiences. Linus’ Law, a widespread assumption within open-source communities, suggests that data quality increases with the number of contributors. In this paper, we evaluate Linus’ Law as applied to the co-production of volunteered geographic information using an agent-based model and examine the effects of knowledge level, variability, and prioritization on emergent production patterns and overall data quality. Our results demonstrate how diminishing returns and the experience of contributors limit Linus’ Law as an intrinsic assessment of data quality.