Hello GRSciColl friends,
Thank you for subscribing to our mailing list. I wanted to let you know that we have changed the way we link occurrences published on GBIF to GRSciColl entries.
If you are unfamiliar with the idea, GBIF links occurrences published on GBIF with GRSciColl whenever possible. You can read more about it herehttps://data-blog.gbif.org/post/grscicoll-flags/. We used to attempt linking any occurrence on GBIF where a collection or institution code or identifier was provided. The rationale was that an institution should also be able to showcase their observation records. We also wouldn’t link occurrences to GRSciColl if the institution code provided didn’t match the owner code provided.
This generated a lot of confusion and mismatches (many observations had codes that happen to match to GRSciColl entries even though they were unrelated to those). So, we decided to only match the occurrences that are published with one of the following bases of record: Preserved Specimen, Fossil Specimen, Living Specimen, Material Sample. In addition to that, we now link records to GRSciColl entries even if the owner institution code differs from the institution code. This allows many records published by Tropicoshttps://www.gbif.org/dataset/e053ff53-c156-4e2e-b9b5-4462e9625424 to be linked to GRSciColl entries.
The behavior for the matching algorithm is documented herehttps://github.com/gbif/registry/blob/dev/registry-service/README.md. Don’t hesitate to send any feedback or question. All the best, Marie
-- Marie Grosjean (mgrosjean@gbif.orgmailto:mgrosjean@gbif.org) Data Administrator GBIF Secretariat Universitetsparken 15 DK-2100 Copenhagen, DENMARK URL: http://www.gbif.orghttp://www.gbif.org/