Community and Code: Nine Lessons from Nine NESCent Hackathons

Abstract

In recent years, there has been an explosion in the popularity of hackathons — creative, participant-driven meetings at which software developers gather for an intensive bout of programming, often organized in teams. Hackathons have tangible and intangible outcomes, such as code, excitement, learning, networking, and so on, whose relative merits are unclear. For example, a frequent complaint is that code is abandoned when the hackathon ends, and questions like, “which outcomes are produced most reliably?” and, “how valuable are they for participants, organizers, and sponsors?” remain open. As a first step in giving “hackology” a more rigorous footing, this paper describes the NESCent hackathon model, developed over the course of a decade to serve the academic discipline of evolutionary biology, with the dual goals of augmenting the community’s shared software infrastructure, and fostering a diverse community of practice in open scientific software development. The paper presents a detailed guide to staging a NESCent-style hackathon, along with a structured information set on nine events involving 54 team projects. NESCent hackathons have produced tangible products with downstream impacts, including prototypes that were leveraged for major funding, incremental additions to production code bases, and creative drafts (designs, standards, and proofs-of-concept) that contributed to further work. The impacts of intangible outcomes could not be assessed objectively, but the subjective experience suggests that hackathons have a positive impact by (1) providing individuals with valuable experiences, networking, and training, and (2) fostering a more cohesive community of practice by enhancing awareness of challenges and best practices and by building links of familiarity between and among resources and people. Future research that recognizes the diverse outcomes of hackathons might enable evidence-based decisions about how to design hackathons for effectiveness.