The FAIR Signposting Profile

Herbert Van de Sompel
Chief Innovation Officer
Data Archive and Networked Services

Martin Klein
Scientist
Los Alamos National Laboratory

The FAIR Principles [1] aim at enhancing the ability of machines to find and use scholarly objects on the web. Landing pages support humans with such tasks by providing descriptive metadata, listing identifiers, and linking to actual content. But landing pages are not optimised for use by machine agents that navigate the scholarly web, among others because they come in many shapes and forms. Signposting, an ad-hoc and intentionally lightweight interoperability effort launched in 2014, provides this type of information in a uniform machine-friendly way. It builds on widely implemented web protocols specified in IETF RFCs, yielding interoperability that is not restricted to the scholarly landscape but encompasses the web at large. The recently published FAIR Signposting Profile [2] is a concrete implementation guideline for Signposting. It details how, with little investment, repositories and publishing platforms can help machine agents accomplish their task by providing them with a map that guides their travels across hosted resources and beyond. Implementation of the FAIR Signposting Profile improves all aspects of FAIRness of scholarly objects on the web. This presentation will revisit the basic technologies leveraged by Signposting: typed web links, IANA registered link relation types, and Link Sets. It will introduce the FAIR Signposting Profile that was devised to support broader adoption of Signposting. A demo will show an implementation of the Profile for a common repository platform, turning it into an environment in which machine agents can feel welcome.