2026-10-29 –, LUMC02
If you replace all add-ons, sensors and modules on an EM microscope, is it still the same microscope? In this short talk, find out how we approach identification and traceability of scientific research equipment with FAST.
FAST is a web application for facility access and management, used by the EXCITE2 network to provide access to European research infrastructure, with a database of facilities, equipment and metadata identified using standard approaches like ORCID and PID.
This presents both an opportunity and a challenge: linking data publications back to the equipment used to produce them is only reliable if you can account for how that instrument has changed over time.
With FAST, we place equipment in time using an audit trail and date markers, pinpointing instrument attributes at any given moment. We'll discuss current progress, open questions, and what we hope to learn at PIDfest.
FAST is a web application for facility access and management, developed for the EXCITE2 network, which coordinates access to European research infrastructure. Alongside proposal and transnational access management (TNA), FAST maintains a database of facilities, equipment, and metadata.
Research equipment can change over time, and aspects like add-ons and modules often change between TNA calls. An instrument identifier tells you what piece of equipment was used, but not what it looked like when the data was collected.
FAST has 2 of the pieces needed to solve this: equipment records with metadata, an audit trail of what changed when. We’re in the process of adding data publication features to FAST, including ‘access period’. The work we’ll present is about connecting data in a pragmatic and meaningful way.
FAST is in active development, so we’ll present our progress, what decisions were made, and what remains open. The open questions are part of why we're bringing this to PIDfest. What attributes are meaningful to capture for traceability? How granular should it be? How do you version a physical object?
This isn’t a unique problem and we're curious whether the PID community has encountered it elsewhere, and what approaches have worked.
