RDA Recommendation on PID Kernel Information

Abstract

Global middleware infrastructure is insufficient for robust data identification, discovery, and use. While infrastructure is emerging within sub-ecosystems such as the DOI ecosystem of services purposed for data and literature objects (i.e., DataCite, CHORUS, CrossRef), in general the layers of abstraction that have made the Internet so easy to build on, is lacking for data especially for computer (machine) automated services. The goal of the PID Kernel Information recommendation is to advance a small change to middleware infrastructure by injecting a tiny amount of carefully selected metadata into a Persistent ID (PID) record. This carefully chosen and placed information has the potential to stimulate development of an entire ecosystem of third party services that can process the billions of expected PIDs and do so with more information at hand about an object (no need for costly link following) than just a unique ID.

The key challenge of the PID Kernel Information working group was to determine which from amongst thousands of relevant metadata elements are suitable to embed in the PID record. This recommendation lays out principles to guide in the identification of information suitable for inclusion in the PID record.

The information contained in a PID record is represented by a PID Kernel Information profile which must be publicly and globally available. For PID Kernel Information to be effective in stimulating an ecosystem of data services, the number of different profiles of PID Kernel Information must be small and their content stable. The recommendation includes a draft profile with illustrating examples and cases for adoption in practice.