experiment vs model DOI #6
Description
Right now, the code gives the DOI of the model + experiment.
Using data['container']['identifier'] we could access the more general model DOI. This would be better suited to the model table that we are asking for. For a paper using historical + 4 SSPs, 5 citations x N models might be overkill.
I suggest that we add the option to choose which DOI with default to model DOI.
(I can do a PR with this and support for paths.)