Persistent Identifier (PID) - A journey of making data machine-actionable

We are in an age of digital convergence. This has resulted in a growing demand for data and services that can be accessed from any device at any time. However, finding and linking the data reliably has been a continuous challenge. FAIR principles states findability as one of the key pillars in the journey of making data machine-actionable. This journey in not new. A good example are published books with its International Standard Book Number (ISBN). However, in this age of rapid digitization it is also imperative to uniquely identify digital objects. In order to distinctly identify a digital object, it needs to have an universally unique identifier. Some of the well know techniques used in industry to uniquely identify digital object is Digital Object Identifiers (DOI) 1 and Handle System 2 . The Handle System includes an open set of protocols, an identifier space and a way to modify the associated digital data. The protocol enables distributed computer systems to store identifiers of digital resources and resolve those identifiers to the information necessary to locate and access that resource. Furthermore, the associated information can be changed as needed to reflect the current state of the identified resource without changing the identifier, thus allowing the ID of the item to persist over changes to its location or content. In NFDI4Cat project [1], we have leveraged the flexibility of the Handle system to not only provide Persistent Identifier (PID) to digital objects but enable linking different digital objects that use the PID. Our unique solution combines group and permission based access control and domain specific digital object subscription. This combination ensures strict data privacy, a must for both academia and industry, thereby making sure only those PID's that are made explicitly public viewable can be accessed. Furthermore, to realize our journey towards machine-actionable data we are leveraging on linkML 3 . LinkML [2] allows to define standardized meta data schema and express PID meta data and its semantics in a machine-readable format. This ensure data consistency, integrity, and interoperability across different systems, tools, and domains. In our pilot phase, we have deployed the Handle server [3] and the REST API [4] on public cloud infrastructure. The handle prefix is registered with Corporation for National Research Initiatives (CNRI) hence, any publicly readable values stored in the handle system will be accessible to all. NFDI consortia members can request for a domain specific space (4cat, 4chem, 4earth, 4ds, 4hum). Using REST API they can mint and manage their PIDs. The REST API provides secure access to domain specific space enabling flexible PID integration into existing or future solutions. Furthermore, it allows linkML based model [5] to store metadata in the handle records. This ensure consistent tracking, integration, and accessibility of resources . This paper highlights our current progress towards our journey of making data machine-actionable.