The dataARC project aims to encourage interdisciplinary collaboration throughout the research process in projects focused on the long-term human ecodynamics of the North Atlantic. An inherent challenge of interdisciplinary work is that it requires us to engage with the slightly alien ways of thinking and organizing information used by our colleagues who specialize in different types of data analysis. Doing so can be laborious, requiring extended explanation and discussion of each domain’s assumptions, and may seem unprofitable as it is clear each researcher will not become expert in the other’s area. dataARC is attempting to support interdisciplinary work by developing a data discovery tool that provides intentionally interdisciplinary search result sets and contextualizes the results from each domain within a shared conceptual model. In practice, the project is creating a data discovery tool that brings back both results directly related to a search and connected results, and provides explanations for the links between them. The creation of structured and contextualized, intentionally interdisciplinary result sets requires that we define the levels of granularity at which data elements operate, connections between data elements to form basic ideas, and connections between specific ideas and broader, overarching concepts. This paper presents the data structure and knowledge model developed within the dataARC project and discusses the challenges encountered in operationalizing these structures to produce useful, intelligible search results that can enable interdisciplinary teams to share their data and understandings of that data with more confidence and clarity. (Rachel Opitz, Colleen Strawhacker, Philip Buckland, Gisli Palsson, Peter Pulsifer, Lynn Yarmey, Emily Lethbridge, Ingrid Mainland, Anthony Newton, Richard Streeter, Tom Dawson, Jackson Cothren)
0 Comments