collection object storage location: to share or not to share?

[garymotz]

Wouter Addink (Naturalis) is interested in sharing inventories of collections based upon storage locations (lists of drawers and jars) with some metadata about what they contain (e.g. on species, genus, or higher taxonomic information). Is it possible to use "storage location" as a basis for a checklist?

DISSCo is interested in using a similar workflow for generating an inventory of all collections in Europe as the first step of digitization (@garymotz has done this as their initial digitization protocol as well)

@tucotuco has indicated that this might be a good option for an extension

Andrea Hahn (GBIF Secretariat) speaking at DwCH 7 indicated this might also be a good case for a sampling dataset, because the inventory could be considered an atypical kind of event. Taxonomy at any level (e.g. "a box of beetles" can be good enough, for starters).

See Wouter's example: GBIF example

Alex Thompson (iDigBio) thinks this could be a good potential use case for NCD (Natural Collections Description). Watch NCD on TDWG's GitHub for development.

@tucotuco asked Wouter if ABCD has the ability to support collection-level storage information

[ekrimmel]

discussed in DwC Hour 7: Aggregators - a Darwin Core View

[debpaul]

Need to get @wouteraddink and @mswoodburn to look at this too. And I'll link this ticket in the NCD github repo

[rondlg]

Yay this will be great to get stuck into. It follows exactly the work we want to see happen with NCD to document collections that have not had funding/time to accomplish specimen level digitization.

[stanblum]

There are two concepts that have the potential be conflated here: location and quantity of storage space. If space is regular and location address is a linear(-ish) metric (cabinet number, bank number -column number - shelf number), there might be a close enough correspondence between location and quantity of collected material. I would guess, however, that irregular situations are common enough to keep these concepts separate in a data standard. (Storage volume is really a proxy for the number of items that need to be cataloged, so why not use a direct representation -- e.g., estimated_number_of_items?)

I've heard that people managing high-value collections might not be willing to share storage locations for certain items or classes of items. Does storage location have a use in the context of combined collection descriptions, or is it only to assist local collection management? (I'm assuming both Darwin Core and NCD are designed for the context of multiple data sources.)