Identifier and IdentifierScheme should be different node objects - this is now the case.
Implementers should not assume that multiple identifiers listed for a given product/facility/organisation are semantically identical. AlsoKnownAs is not the same as owl:sameAs."
It's been really hard to try to defend the adms:identifier approach in a number of data modeling contexts. People simply want to state that an "Entity has an ID" - and if there are several IDs then the Entity has properties like "thisID" and "thatID" and "anotherID" etc. Sic!
Now I'm getting even more confused since this discussion seems to separate the Identifier-class into a Identifier -class and Identifier System -class.
Would be interesting to hear your opinion on this: https://github.com/uncefact/spec-untp/issues/137
The conclusion of the discussion is:
"I think the outcome of all this is
It's been really hard to try to defend the adms:identifier approach in a number of data modeling contexts. People simply want to state that an "Entity has an ID" - and if there are several IDs then the Entity has properties like "thisID" and "thatID" and "anotherID" etc. Sic!
Now I'm getting even more confused since this discussion seems to separate the Identifier-class into a Identifier -class and Identifier System -class.