Issue 621: Update synchronization btw CRMarchaeo and CRMbase
The CRMarchaeo maintainers have proposed to perform a systematic check of its classes and properties' hierarchy against CRMbase. This issue is for relevant discussions.
Post by Christian-Emil Ore (4 December 2022)
Some homework concerning superproperties of
- AP4 produced surface (was surface produced by): A10 Excavation Interface
- AP5 removed part or all of (was partially or totally removed by): A8 Stratigraphic Unit
- AP10 destroyed (was destroyed by): S22 Segment of Matter
can be found issue folder for the 55th meeting, here: https://docs.google.com/document/d/1CUNT2Z2ZDFgTXf0BAqQ-9DGIvBRDPIq8gL5x...
In the 55th joint meeting of the CIDOC CRM and SO/TC46/SC4/WG9; 48th FRBR/LRMoo SIG meeting, the SIG reviewd HW by CEO, on the properties of of A1 Excavation Processing Unit (IsA S1 Matter Removal, and IsA S4 Observation) that are problematic (in the sense that they assume contradictory semantics for class A1):
AP4 produced surface (was surface produced by)
AP4 is declared a subproperty of both P92 brought into existence and P108 produced. However, P108 IsA P92, which means that one of the two superproperty declarations is superfluous. Deciding which one to keep depends on the choice of superclass relation for A1 (the domain of AP4).
Proposal: drop the AP4 produced surface IsA P92 brought into existence, which means that we also have to drop that A1 Excavation Processing Unit IsA E63 Beginning of Existence in favour of A1 Excavation Processing Unit IsA E12 Production.
- the production of the surface in the course of an instance of A1 is necessary
- the added value of declaring AP4 IsA P108 (and A1 IsA E12) is not clear, in the sense that the surface produced is a by-product of the instance of A1.
- A1 can be made a subclass of almost all classes. It’s core to CRMarchaeo, a fact which can be captured through multiple instantiation rather than excessive subclassification.
There was no consensus on the proposal. Proceed as follows:
HW to the CRMarchaeo editors to revisit and come up with a new proposal (or same proposal supported by new evidence). Evidence for/against the proposal:
- How does A1 benefit from using properties that are particular to E12 that are not available for E63?
- How does making A1 IsA E63 (hence not E12) prevent unintended modelling?
AP5 removed part or all of (was partially or totally removed by)
Ap5 is currently declared IsA P31 has modified (with domain E11 Modification). However, the domain of AP5 is A1 (that IsA S1 Matter Removal (IsA E7 Activity) and IsA S4 Observation (IsA E13 Attribute Assignment IsA E7))
Decision: drop the AP5 IsA P31 has modified and make it IsA O1 diminished [D:S1 Matter Removal, R:S10 Material Substantial]
AP10 destroyed (was destroyed by)
AP10 is currently declared IsA P93 took out of existence (with domain E64 End of Existence). This means that A1 needs to be declared IsA E64 as well. Alternatively, the superproperty relation needs to be dropped.
Decision: Keep the superproperty relation (AP10 IsA P93 took out of existence [D:E64 End of Existence, R: E77 Persistent Item])
Belval, December 2022