Send iNat user id to GBIF to make observations searchable by unique user

Feature Requests
1

Platform(s): NA, applies to the GBIF export and the GBIF search

URLs (aka web addresses) of any pages, if relevant: see below

Description of need:
On iNat, users are identified by their user id (unique, a number), their user login (unique, often a casual nickname), and their user name (not necessarily unique, often a real name). When iNat sends records to GBIF, they send the non-unique user name, which is searchable in the GBIF interface, for example https://www.gbif.org/occurrence/search?dataset_key=50c9509d-22c7-4a22-a47d-8c48425ef4a7&recorded_by=Tony%20Iwane. This field has some issues because user names are not unique (and also not permanent – users can change them). For example, https://www.gbif.org/occurrence/search?dataset_key=50c9509d-22c7-4a22-a47d-8c48425ef4a7&recorded_by=John%20Smith contains results for both cicadadayz and also sqeeaawk408, both with user name John Smith on iNat. A very small fraction of iNat users have set an ORCID, which makes their records uniquely tied to them, for example: https://www.gbif.org/occurrence/search?dataset_key=50c9509d-22c7-4a22-a47d-8c48425ef4a7&recorded_by_id=https://orcid.org/0000-0002-1096-6066. But outside the users with ORCIDs set, there is no unique, permanent field in GBIF to identify the observer.

Feature request details:
In the same field where iNat sends the ORCID (the DwC recordedByID term), please add the iNat user id. This field can take multiple values, so for most users, it would read something like https://inaturalist.org/users/28, and for users with an ORCID, it would read https://orcid.org/0000-0002-1096-6066 | https://inaturalist.org/users/1138587. If that were added, in addition to the already valid search
https://www.gbif.org/occurrence/search?dataset_key=50c9509d-22c7-4a22-a47d-8c48425ef4a7&recorded_by_id=https://orcid.org/0000-0002-1096-6066
you would also be able to use
https://www.gbif.org/occurrence/search?dataset_key=50c9509d-22c7-4a22-a47d-8c48425ef4a7&recorded_by_id=https://inaturalist.org/users/1138587
to find the same observations.
This would allow searches for observations by iNat users who have changed their user name, or who have a non-unique user name.

I think it’s rarer that people use the identifiedByID field (in part because it only holds the ID of the first identifier, not all identifiers), but it would be reasonable to also populate this field at the same time.

Why this rather than asking GBIF to honor the existing inaturalistLogin field?

As discussed in this earlier thread, iNat already exports the user login as foaf:nick (inaturalistLogin), but GBIF doesn’t currently ingest that term. The natural “ask GBIF to add foaf:nick support” approach has two problems compared to populating recordedByID:

  1. Logins are mutable — users can change their iNat login, the numeric user id never changes.
  2. recordedByID already works end-to-end in GBIF — parsing, faceting, and search are all live today. Adding the iNat user id to that field requires no coordination with GBIF.
2

I’m surprised an unique identifier isn’t sent already. In any case, you may want to think about using https://inaturalist.org/users/28 (a URI) rather than https://www.inaturalist.org/users/28 (a URL).

3

done :+1:

4

Side question, i checked my profile settings and cant find it. Where do i set my ORCID?

6

Account Settings → Applications there will be a “connect” to ORCID

7

This is also not a Darwin Core field, so I guess it will anyway not be as easy for GBIF to simply “ingest that term.”

8

Thanks that helped!

9

Good point! Maybe the best solution is to encourage users to link their ORCID.

10

And what about users who do not have an ORCID because they are not in academia?

11

I didn’t know that was a requirement.

12

Users who do not publish in scientific/academic journals are unlikely to have an ORCID. (Quite a few of us fall into this category – it’s sort of the point of citizen science.) It seems a bit unreasonable to expect them to register for one just so that their iNat identity can be integrated into GBIF; thus, an alternative would seem desirable instead of assuming that using an ORCID is an option for everyone.

13

There are also those of us who do have an ORCID but it has nothing to do at all with biology even in the loosest sense. It’d be inappropriate to link that to GBIF.