skybristol / geokb

Data processing workflows for initializing and building the Geoscience Knowledgebase
The Unlicense
3 stars 3 forks source link

ORCID processor for person entities #7

Closed skybristol closed 10 months ago

skybristol commented 11 months ago

As I'm slowly porting functionality into the GeoKB that was built previously for something called iSAID, I need to work up a processor for gleaning useful claims from the ORCID system. These include additional name aliases that can be useful, keywords pointing to subject matters, affiliations, and other works that we might want to leverage.

One important decision point is on whether to cache information from the ORCID system. It's not so much stability or the technical capability of the interface (those are pretty good), but we might want to have a history somewhere of what someone had in their ORCID record over time. The ORCID platform likely has this, but I don't think it's available anywhere. Fundamentally, if someone once "claimed" to have some type of expertise in a subject matter by including it in their ORCID record, we incorporate that into the GeoKB as a referenced claim, and then it goes away, our evidence base has disappeared. If we cache it, then we can always get there. But that might also be overthinking things.