Support annotating existing datasets with ontology concepts

[leinfelder]

Assuming a workable annotation model (see proposed OA-based approach), students could attach measurement concepts to existing data packages that have already been documented with some structural attributed metadata and free-test descriptions (e.g., EML).

TBD:

• How to annotated 1,000s of datasets that are not manually annotated?
• Suggest terms? Automatically categorize with natural language processing?

[mbjones]

Use case 50 and 51 cover this, and are complete. https://github.com/DataONEorg/sem-prov-design/tree/master/docs/use-cases/semantics