National PID Strategies

Letter from the Executive Director, August 2024

The knowledge graph that represents the world of scholarly literature is vast, complex, and in many respects difficult to navigate. There are many confusing turns, alcoves, and dead ends. There is also a wide range of scholarly outputs and nodes in the graph that aren’t visible because they aren’t uniquely identified, or if they are, that identification and the associated metadata isn’t exposed to the community. One might not necessarily be certain that one is following the correct path. Assessment of the graph is incomplete at best.

Persistent identifiers (PIDs) can help this navigation process by connecting the nodes, be they people, papers, datasets, research tools, facilities, grants, or projects. Much has been written about the value of PIDs in research communications. They have been proven to reduce administrative burden on scientists, improve discovery of research objects, and reduce overall costs of research, and can lead to transformations in research assessment. These efficiencies can only be achieved through consistent use across all elements of the research enterprise, from grant applications to assignment of facilities, project management, research evaluation systems, numerous publication processes, and repository management. It isn’t so much that there is a shortage of identifiers for these objects. Rather, the problem lies in inconsistent practice, misapplication of existing standards, and the ever-present belief that one has an “even better solution,” which leads to the development of something new instead of improvement of an existing process.

For the past several years, under a project led by the Research Data Alliance, a group focused on national PID strategies has been working to advance consistent application of persistent identifiers in scholarly communications. The national PID strategies recommendations were approved in early 2023, and several national research organizations have adopted variations, including in Australia, Canada, the UK, Ireland, and Japan. Many of these projects gave updates during the PIDFest conference in Prague in June. Alice Meadows, Phill Jones, and I wrote about the meeting at the Scholarly Kitchen blog a few weeks ago.

During the meeting, I reported on a project that I co-led with John Chodaki at California Digital Library beginning in 2022. Organized by the Open Research Funders Group, the team set out to develop a national persistent identifier strategy for the US. Building on the work of RDA’s national PID strategies outputs, the efforts to create and share national strategies from other countries, and other guidance documents issued by the US federal government, the group produced a model for developing a ational PID strategy. That report was published in February 2024. With the support of the RDA-US team at Indiana University’s Pervasive Technology Institute, John and I have put together a series of short videos describing the report and the progress, which have been posted on the RDA-US YouTube channel. The report has elicited a lot of conversations and ideas about how it could be improved.

This was part of our ideal process, since no single group working without broader input can be expected to capture all the nuance of the entire US research ecosystem. Judging from the comments and responses that have been submitted so far, there is certainly much that could be improved, including how areas of existing practice should be represented or how new identifiers for as-yet unidentified objects should be vetted. The feedback is welcome and full of many excellent points. Early in the process, we anticipated an iterative process that could yield a consensus standard, which could be adopted across the ecosystem. We’re now moving in that direction.

The proposal to launch this as a NISO standards project was submitted this spring. It was approved last month by the Information Policy and Analysis Topic Committee, and the NISO Voting Members are currently balloting a question on whether the project should be launched. If the vote passes, then a Working Group will be organized this fall to incorporate the existing feedback into the report and solicit additional input from a broader constituency, then redraft the guidance. Ideally, the result will be a NISO/ANSI standard that can advance this issue. 

There is so much potential in our community for persistent identifiers to be even more widely used to facilitate scientific communication. In an era of greater machine interaction and content navigation, the robustness of the scholarly graph is vital. It will reduce costs, improve assessment, reduce integrity concerns, and hopefully speed science, resulting in better outcomes. PIDs are relatively small things, but they can be extremely powerful if adopted and implemented consistently, hence the need for standards.

Sincerely,

Todd A. Carpenter. Executive Director