Persistent Identification of Instruments WG rev-002

The Persistent Identification of Instruments RDA Working Group (PIDINST WG) seeks to propose a community-driven solution for globally unique and unambiguous identification of instruments instances that are operational in the sciences.

Please see https://www.rd-alliance.org/sites/default/files/case_statement/rda-wg-pi… for the full case statement for the PIDINST WG.

In her recent book, entitled “Big Data, Little Data, No Data” [1], Christine Borgman writes “To interpret a digital dataset, much must be known about the hardware used to generate the data, whether sensor networks or laboratory machines.” Borgman further highlights that “When questions arise […] about calibration […], they sometimes have to locate the departed student or postdoctoral fellow most closely involved.” This is a striking account for the role information about instruments plays in science and the costs of not being able to find and access such information.

The need to uniquely identify an instrument instance is rapidly growing in many research communities. Indeed, persistent identifiers enable unambiguous reference to digital representations of instruments, which has many potential benefits:

• Metrics that quantify the use of instruments and the rationale for future funding
• Link data to the instruments that generated them (provenance), improving the interpretation and validity of data
• Aid equipment logistics and mission planning
• Facilitate interoperability and open data sharing, especially in advancing technologies that foster sharing of instruments
• Improve the discoverability and visibility of instruments and their data, published on the web.

Currently, there is no universal way to identify instrument instances. As the primary outcome, PIDINST WG contributes to establishing a cross-discipline, operational solution for the unique and lasting identification of active and decommissioned instruments. This case statement outlines the work planned for PIDINST WG.

Issues to be addressed

• Instruments as physical entities – What is an instrument? Implications of identifying the instrument instance as a physical object versus identifying a digital information object (metadata) about the instrument. What do instruments produce, their real-world configurations, their relations to platforms and deployments, and the implications of instrument modifications to identification (new versions).
• Granularity – Instruments can be parts of other (compound) instruments. For example, instruments can be manufactured with multiple bespoke sensor components, such as modular weather stations that simultaneously measure multiple meteorological variables. The granularity at which to reference and describe instrument instances (compound versus component) can vary for different stakeholders. How can these types of instruments be described in a generic way.
• Use cases – Support the analysis of community requirements and inform the work carried out by PIDINST WG.
• Metadata – Explore the types and sources of metadata that could be resolved under a PID and the difference between metadata registered at PID infrastructure provider (e.g. DataCite, ePIC, Crossref) vs. metadata at institutional instrument database provider. Develop a minimum common metadata schema for the registration of instruments with PID infrastructure providers.
• Machine readability, interoperability, and provenance – Investigate the need and the requirements involved to make metadata (at the institutional level) machine readable and compatible with existing interoperable technologies. Provenance, in particular the relation between data and instruments that generated them, is another aspect to be addressed.
• Landscaping – Explore the links, potential relationships and overlaps with instrument manufacturers, institutional instrument database providers, RDA groups and PID infrastructure providers.

Outcomes

The work of the PIDINST WG will contribute to the following outcomes. Note that these are long-term outcomes this WG aims at contributing to. This WG will not build a sustainable infrastructure for the persistent identification of instruments. It will merely contribute to specifying such infrastructure. The concrete deliverables of this WG are presented in the Work Plan.

• A sustainable infrastructure will support the registration of instrument instances by submitting metadata about them and allowing for minting an instrument instance PID. The PID must follow agreed standards for persistent identifiers, e.g. long-lasting actionable, descriptive digital identifiers.
• Improved understanding within research communities for how to describe instrument instances, including relations to other entities such as instrument model (type) or instrument deployment, the issue of identifying physical objects versus digital representations, and other related issues.
• Collaborations with one or more PID infrastructure provider interested in implementing the approach to persistent identification of instruments proposed by the PIDINST WG.
• Strong linkages to the activities of the RDA PID IG and other related RDA groups.

References

[1] Borgman, C.L. (2015). Big Data, Little Data, No Data. MIT Press.

Please see https://www.rd-alliance.org/sites/default/files/case_statement/rda-wg-pi… for the full case statement for the PIDINST WG.

rda-wg-pidinst-case-statement.pdf