Required controlled vocabularies

[mesemus]

For NMA, we need to have some values from the controlled vocabularies to be able to provide meaningful statistics/dashboards.
As NMA can not perform identifier deduplication, this will mean for example for institutional agent / affiliation:

1. if no identifiers are provided for the institution, NMA will harvest the record but the record will not be a part of institutional dashboard
2. if identifiers are provided and can not be mapped to the internal vocabulary of institutions, the record will not be accepted. The reason is that we want to avoid confusion "but my institution has an identifier, why is it not shown on the dashboard".

We need to document this and make it either part of CCMM documentation or NMA documentation.

[michal-klodner]

In terms of wider organizational processes, NMA or NRP documentation can specify requirements for vocabularies somehow independently, but there is different capacity for communicating the metadata guidance outside of main CCMM and EOSC-CZ WG Metadata channels, which are meant for these purposes.

My +1 to provide mandatory and "recommended" vocabularies as part of CCMM documentation for clear guidance of implementors.

In CCMM 1.0, there is no PID scheme and vocabulary for institutions specified, which gives no answer what is best practice for PIDs at least and leaves open space for incompatible interpretations with no mapping possible.