Infrastructures for Managing and Publishing Large, Heterogeneous Linked Datasets

Abstract

• A Linked Data approach in Humanities research is likely to produce large, rich, heterogeneous datasets with huge research potential, but how can these datasets be managed and published in a form that is flexible, scalable, interoperable and, ultimately, sustainable? Several infrastructures exist that aim to address this issue, but come with challenges of their own, including the balance between the need for off-the-shelf front-end solutions, and for APIs that can support custom user interfaces, including explorable data visualisations. For this activity, we want to hear about your experiences with such infrastructures, to contribute to discussions about available solutions and community requirements for an ‘ideal’ infrastructure of this kind - a speculative design - such as might be created in the future. Our asynchronous component will start with an introduction to the Tools of Knowledge project, where we will share and evaluate our experience of managing and modelling data using the Arches platform. We will then open up a Google Doc for you to share your own experiences of working with similar platforms, best practice, and lessons learned. An in-person component will follow, where we hope contributors will join us for a panel discussion on this important topic.