During a recent vocabulary update, I discovered that there are three Many-to-Many relationships which prevents loading data without removing the constraints.
Domain<–>Concept
Vocabulary<–>Concept
Concept_Class<–>Concept
I don’t think this is a good long-term strategy. Loading data should not require structural changes to the data model.
I’d suggest, instead of a model which looks like this:
It should look something like this:
Or, perhaps combining them into one junction table.
It appears that has already done this with Concept<–>Relationships with the Concept_Relationships table.
I understand that changing the data model after the fact is a MESS, and will require some significant changes throughout. The question is if correcting the model is worth the effort, and how much trouble will failure to correct it cause down the line.