External References & API Interaction

[dpwrussell]

Hi,

Especially for fields where there are vast amounts of data and/or non-file-based representations, I think there is a requirement to reference globally unique IDs in a external repository instead of importing the data into Dataverse. This would be up to the owner of the Dataverse to determine which external repositories were eligible to fulfil the contract required by Dataverse to maintain data integrity.

An example is @openmicroscopy OMERO which is for managing microscopy data. This data is very large and very complex and requires advanced tools to manage and visualise and it would be great to take advantage of those external systems to manage that data, but be published with versioning through Dataverse.

In addition, it would be great if it were possible to build modules to allow these external tools to be used through APIs from Dataverse. For example, in the OMERO case I can imagine that a microscopy metadata query could use the OMERO API to run a query against the OMERO server which could then return some results to Dataverse.

The other obvious use case I could think of would be for source code. Instead of uploading a zip file of source code, a reference to a specific GitHub commit/tag could be specified. As I said before, this relies on the owner of the Dataverse trusting these external sources.

Thanks

[dpwrussell]

I realise that there is probably some crossover with #2245

[4tikhonov]

Yes, this is exactly the same topic of discussion how to store API endpoints with metadata descriptions and technical specification in Dataverse.

[dpwrussell]

Closing in favour of #2245