review data model for multiple classifications

[kcranston]

@cgendreau created a data model for holding multiple classifications for a single DINA instance. Can folks review in the context of the taxonomy use cases.

https://github.com/DINA-Web/taxonomy/blob/master/docs/taxonomy_concept_diagram_multiple.png

Key points about the model:

• name: shared across classifications; contains nomenclatural information
• taxon: shared across classifications; contains reference information (is a taxon concept) and links to a name
• node: specific to a classification; contains parent info and links to taxon
• synonyms: synonym defines taxonomic synonyms, while each nomenclatural synonym is a separate entry in the name table

Terms defined in the use case glossary.

[dshorthouse]

I see rank in the taxon table whereas I would put that in the node table.

[dshorthouse]

Vernacular names could do with additional terms. See https://tools.gbif.org/dwca-validator/extension.do?id=gbif:VernacularName

[dshorthouse]

Where might flags for taxonomic status of synonyms be stored (eg homotypic, heterotypic)?

[dshorthouse]

Is an adjacency list sufficient or should the node table be additionally extended to nested sets?

[cgendreau]

• The reason why rank is on taxon (instead of node) is based on the fact that if it would be at another rank, it would be a different taxon concept (based on my current understanding).
• The original idea for taxonomic status is in the previous diagram but I think to simply store the taxonomic status would probably be enough. What do you think?
• Regarding the storage/query of the tree it is still undefined but I would try adjacency list with Postgres RECURSIVE CTE.