Sharing Data and Resources

Arctos is a shared system, which means we share dictionary and authority files, as well as ideas and development, while individual collections exclusively own core catalog record data. This page will attempt to summarize how this works. Any user with access to shared nodes should have a thorough understanding of this information, and how it affects their collection and others.

Navigation: Manage Arctos > Roles/Permissions > Database Roles

What is not shared

Catalog records, and core catalog record data such at Attributes, Identifications, Collectors, and Citations, are wholly owned by individual collections.

However, relationships allow users to navigate between related (including “same individual as”) catalog records, and to perform cross-collection queries (such as those involving hosts and parasites). The Data Entry application allows related catalog record data (events, places, collectors, etc.) to be “seeded” into new catalog records. Various reports and alerts exist to facilitate maintaining these relationships.

Loans, accessions, and borrows are not shared, but can include objects from more than one collection. Projects are shared, but the transactions associated with them are not.

Catalog record events are not shared, but some of their underlying information (Events, Localities, and Higher Geography) are.

What is shared

Code tables with a “collection_cde” column are shared across collection types. That is, all “Herp” collections (and no “Mamm” collections) have access to an Attribute “carapace width” which in all collections requires a numeric value and units. All collections share non-partitioned code tables. That is, every collection with any numeric Attribute may access the “length units” code table.

Taxonomy is shared at the “source” level, which is set by collection under Manage Collection. All collections which use the “Arctos” classification should be consulted before changes are made; if no agreement can be reached, it may be necessary to split the classification. All classifications are available for querying data from all collections.

Agents are fully shared, and thousands of agents span collections – as collectors, borrowers, authors, and georeferencers. Changes to Agents should be coordinated by all users (available from Agent Activity). Major changes (such as mergers) prompt notifications and have a waiting period.

Publications and Projects are shared globally. Researchers borrowing objects from several collections may create shared Projects which produce shared Publications citing multiple collections.

Media are shared globally. PDFs of shared publications are one use case; field notes of hosts and parasites are another.

Containers are shared globally, facilitating the combined storage of e.g., parasites and “part parasites” belonging to vertebrate collections.

Higher Geography is shared globally.

Localities are shared. Changes are logged and reported.

Events are shared; Changes are logged and reported.

When to Edit

Some data (taxonomic names, for example) are structurally locked and cannot be edited when used; the view is that ANY change might be a fundamental change to the nature of the data, and so no changes may safely be allowed. (taxonomic classifications are not locked, however.) Other data (geography, for example) may be changed by Operators with appropriate access, and the safety of the data is left to careful operators. The general question to ask is, “Am I changing the fundamental nature of the data?” If the answer is “yes,” the data should not be changed. Examples:

Community Discussion

• Github Issue #123

Edit this Documentation

If you see something that needs to be edited in this document, you can create an issue using the link under the search widget at the top left side of this page, or you can edit directly here.