Font Size:
Building a Biodiversity Information Network using Persistent, Resolvable Identifiers
Building: Grand Hotel Mediterraneo
Room: Sala dei Continenti
Date: 2013-10-30 02:41 PM – 02:50 PM
Last modified: 2013-10-07
Abstract
Our knowledge of biodiversity is built on the recording of instances of organisms, biological material, and events. These are then further aggregated and integrated to create a composite view of life on the planet, spanning multiple taxa, time-periods, and geographic scales. In biodiversity informatics, an instance of a specimen or observation is typically described by its taxonomic name, who identified it, and the date on which it is captured. However, these are typically just an assemblage of properties that are informally tied to the thing itself, lacking a robust mechanism for other systems to say things about the same instance of an object. Tracking unique instances of objects, whether physical or virtual, and processes that act on those objects is useful if we want to track them and their derivatives across systems and domains; for example, sequences from tissue or gut samples are related to a physical specimen, itself linked to the context from which it was derived. While the implementation of semantic approaches for linking biodiversity object instances covers formal models, semantic technologies, and identifiers, we focus here on a persistent identifier implementation called Biocode Commons Identifiers (BCIDs, see http://biscicol.org/bcid/). BCIDs are built on top of EZIDs (http://n2t.net/ezid), employing archival resource keys (ARKs) and the name-to-thing resolver (http://n2t.net/) to provide resolution services and a persistent back-bone for future stability. BCIDs use a suffix-Passthrough feature to allow any local identifier or UUID to resolve to a group level identifier, which provides a level of scalability and resolvability well suited to the needs for biodiversity science. Finally, the BCID approach will be demonstrated by applying them to samples as they are collected as part of a field information management system: newly collected field data being an especially useful point of entry for linking and integrating with downstream data aggregators.