Review answers to conceptdescription

[JoergNeidig]

I have the feeling that some answers concerning conceptDescriptions, isCaseOf, etc. are older and do not reflect v3.0 properly. E.g. the answers of the two following questions appear to be in conflict:

• How to use isCaseOf to indicate 'alternative' semanticIDs of AAS elements?
• How shall properties be entered which are defined both in ECLASS and in CDD?

[StenGruener]

Update 2024 with the v3 meta-model:

• isCaseOf: should be used when the concept description is "slightly modified", e.g., using a different unit contrary to ECLASS defintion or a different value domain, e.g., string instead of boolean. In this case the supplied concept descriptions should have a different semantic ID as the ECLASS IRDI.

• supplementarySemanticID: should be used by "exactly same" semantic meaning, as in the use-case of 3 semantic id in the first example: ECLASS, CDD, and TechnicalData submodel template. In this case the supplied concept descriptions should have a same semantic ID as one of the referenced IRDI/semanticIds.

@BirgitBoss would you sign this off?

Proposal - remove both questions and provide this one as a new answer.

[BirgitBoss]

@StenGruener @JoergNeidig thanks for pointing this out. (The new) supplementalSemanticIds belong to elements having semantics (inheriting from hasSemantics), isCase of belongs to concept descriptions (not inheriting from HasSemantics). This is useful if maintenance of concept description repositories is not coupled to maintenance of submodel templates or submodels.
I personally would say that all isCaseOf values could also be entered as supplementalSemanticIds although I would not recommend to do it.
We need a more in depth discussion on the usage of the two attributes. But I would agree to remove these two answers from the FAQ or explicitly state that the answers are valid for V2.0 only until we provide answers valid for V3.x.

[BirgitBoss]

See also How to define concrete semantics for arbitrary submodel elements based on generic semantics?

[StenGruener]

I will mark both questsions deprecated as proposed by Birgit
@BirgitBoss any news on this topics?

[BirgitBoss]

admin-shell-io/aas-specs-metamodel#479 deals with supplementalSemanticId
In general matching strategies for semantic IDs might be relevant, see also admin-shell-io/aas-specs-metamodel#471

But I agree: mark answers as valid for V2.x only.

[StenGruener]

Marked as v2.x only

[StenGruener]

admin-shell-io/aas-specs-metamodel#479 (comment)