The following slides were presented by Adèle Barclay, Susan Brown, and Jentery Sayers at HASTAC 2013 (York University) on 27 April 2013, as part of an INKE research paper titled, “The Key to All Ontologies?: The Long Now of Linked Data.” The paper corresponds with ongoing work by the INKE Modelling and Prototyping Team, including Susan Brown, Harvey Quamen, Jentery Sayers, John Simpson, Adèle Barclay, Jon Bath, Alex Christie, Lisa Goddard, Jon Saklofske, and Mandy Elliott, in partnership with the Text Mining and Visualization for Literary History Project and the Modernist Versions Project. Please note: additional slides were presented during the HASTAC 2013 paper but are not included here. Should you have any questions about the slides, please contact Susan Brown (firstname.lastname@example.org) or Jentery Sayers (email@example.com).
Digital scholarly communications are increasingly dynamic, collaboratively-produced texts that emphasize interlinkages across unique, distributed resources. The now popular Resource Description Framework (RDF) offers considerable potential for supporting these aspects of digital scholarly production through the creation, publication, and harvesting of public RDF in the form of Linked Data.
However, RDF and Linked Data have been mobilized largely in the sciences, and very little humanities research has been conducted on either. As such, this INKE paper asks what the humanities have to learn from RDF and Linked Data, and—more specifically—how each may allow scholars to explore “the kinds of humanistic phenomena” that “appear only at scale” (Liu 2012). This paper surveys an array of existing humanities projects involving RDF, organizing them into the following categories: 1) domain-centric projects, which build upon previously established preservation projects and extend them online; 2) aggregator projects, which gather contextual information from disparate sites around the web and afford access to millions of scholarly materials, often through advanced visualization techniques; and 3) tools, which leverage the synergistic integrations promised by the growth of semantic web activities in the humanities and help scholars navigate, describe, and interpret large sums of data.
Based on this survey, the INKE Research Team has concluded that humanities applications of RDF and Linked Data generally differ from those in the sciences. Whereas science-based applications tend to privilege a single structure or ontology, humanities applications focus on user-based knowledge creation and customized ontologies and approaches.
Yet this conclusion also acts as a cautionary tale for the future of RDF and Linked Data in humanities projects, namely because customized ontologies and approaches pose a number of difficulties where accessibility and interoperability are concerned. Transparency of knowledge representation and ease of use will have a major influence on how effectively Linked Data will help humanities scholars explore phenomena that appear only at scale. As such, this paper ultimately recommends that digital humanities practitioners consider the “long now” of their RDF and Linked Data projects (Eno 2003). A form of long-term thinking and responsibility, working in the long now involves designing, building, and maintaining domain-centric collections, aggregation projects, and tools that think seriously about the audiences, developers, and archivists who are well off in the distance, in 2023 and beyond.