Multiple Ontology Term URIs

[david4096]

From #694

Ontology Terms can be represented by multiple unique URIs. Perhaps we should allow multiple term_uris to be defined for any Ontology Term?

@jmcmurry added a line comment 4 hours ago

If multiple discrepant URIs exist and are known, what is the plan for describing / collapsing them?

[mbaudis]

-1 on this.

I am not sure how this would be of any benefit - iterating through the if one doesn't resolve? But then, which precedence? This should better be addressed captured through a service which remaps ontologies/URIs of moved/deprecated end points.

[mcourtot]

I would recommend not to. Instead, as per #695, we'll rely on an external registry. Note that this may imply we remove the termURI attribute at a next iteration, but would provide greater flexibility to redirect towards active resolvers.

[david4096]

@jmcmurry care to comment?

[jmcmurry]

No perfect solution but agree w/ @mcourtot that reliance on OLS is the way to go. Any objections? How close are they to having API access to their registry?

[mcourtot]

Pinging @simonjupp. Note this relates to the discussion during the Vancouver meeting and at #694.
In short, we decided to keep CURIE + label to denote ontology terms and remove other attributes.