Property talk:P106

ORCID identifiers provide one of the best unique, persistent, and resolvable identifiers for USGS people based on both how they are managed through the ORCID registry and the internal requirement the USGS has for contributors to USGS information products to be identified by their ORCID. We record ORCID identifiers for people mostly via discovery from the USGS Staff Profiles, but we also instantiate person records for other authors identified with ORCID identifiers in the Pubs Warehouse harvest.

Claims from ORCID

The ORCID registry provides a number of useful bits of information for people that are more or less useful depending on how well someone has maintained those aspects of the ORCID system they can directly manage and how well contributing processes have captured their work (e.g., creative works contributed to, peer reviews conducted, etc.).

Keywords

This is an optional field that is not controlled in any way. An individual ORCID owner can provide whatever keywords they want to describe some aspect of their work. These function in a similar way to the expertise keywords some people include in their USGS Staff Profile. Our method of using them in the GeoKB is to find any matches to groups of concepts such as subclasses of research method.

Biography

This is an uncontrolled text field that an ORCID owner can provide for themselves. It functions in a similar way to the "body text" that is scraped from USGS Staff Profile pages except that it is strictly text and will contain no HTML formatting. These are used in concert with other text blobs directly associated with a person to run a topic modeling pipeline where we link a person to specific concepts modeled into the GeoKB such as research methods, fields of study, and others.

Researcher URLs

ORCID owners can provide URLs that point to other information about themselves. Some USGS people record their Staff Profile URL here, but may also include other URLs that can be followed up on to gain further insights and topic modeling fodder. In the near term, we are recording these as additional URLs for the person.

Information Caching

To help facilitate further processing and a provenance trace, we cache select aspects of the ORCID JSON response in YAML within the item discussion pages for person records with ORCIDs. By "select aspects" we mean the following:

• History object is cached to help understand where the record came from and when it was last updated
• Person object is filtered to only include sections that have been populated
• Work-activities object is filtered to only include sections that have been populated

This process also updates the retrieved qualifier on the ORCID iD claim to indicate when that information was last pulled and cached.