make regions roles instead of immaterial entities

viktorwichern commented 8 months ago

Description of the issue

Part 1

While discussing Issue #1489 with the OVGU-Group, we agreed that the current structure of spatial region should be slightly changed. One reason for that is the fact that study region considered region and interacting region are currently on the same hierarchical level as one-two-three and zero-dimensional region.

Part 2

Right now, subregion is a subclass of spatial region and has only one child, study subregion.

study subregion is defined as: A study subregion is a subregion of a study region. but has no relation to study region, neither through inheritance nor axiom.

Are subregion and study subregion important for the ontology? If we decide to make the regions into roles, study subregion has to be edited as well.

This is a little bit confusing, I offer a solution below but I am not sure, since subregion is defined as a spatial region that is part of a spatial region, so theoretically, a subregion could be bearer of the study region role as well, which would make the distinction obsolete. This may be related to the circular definition of subregion.

Ideas of solution

Make the different regions into roles of spatial region instead.

study region role: definition: A study region role is a role of a spatial region that is under investigation and consists entirely of one or more subregions.

interacting region role: definition: An interacting region role is a role of a spatial region that interacts with a spatial region that has the study region role. It is part of a spatial region that has the considered region role, but not a study region role.

considered region role: definition: A considered region role is the role of a spatial region that is used in an analysis.

Part 2

Change study subregion into a role.

study subregion role definition: The study subregion role is the role of a subregion that is under investigation and is part of a spatial region with the study region role.

'subregion' and 'part of' 'spatial region' and 'has role' some 'study region role'

The current examples of usage, alternative labels and the term tracker items of the regions should be included in the new roles as well.

Workflow checklist

[ ] I discussed the issue with someone else than me before working on a solution
[ ] I already read the latest version of the workflow for this repository
[ ] The goal of this ontology is clear to me

I am aware that

[ ] every entry in the ontology should have a definition
[ ] classes should arise from concepts rather than from words

l-emele commented 8 months ago

Why two issues for the same topic? Please bring these in one issue, else the discussion on one topic will be scattered across the issues.

viktorwichern commented 8 months ago

@l-emele I united the two issues to one. The reason I made two separate ones in the first place was because I thought the discussion about the definition/use of subregion could maybe be worth its one issue, but I agree that it seems kind of unnecessary in hindsight.

Regarding your suggestion from the other issue: I think having those role is a good idea. However, I think we should add these additionally and not instead of the existing classes. We should keep then the existing classes as equivalent classes.

I somewhat disagree. As explained in the issue, having the regions at the same hierarchical level as BFO-metacategories seems poorly constructed.

Counter-Proposal: Implement one meta-class under spatial region that contains the equivalent region-classes, for example (rough draft):

region of relevance: definition: A region of relevance is a spatial region that is mentioned in any study or analysis.

l-emele commented 8 months ago

As explained in the issue, having the regions at the same hierarchical level as BFO-metacategories seems poorly constructed.

I agree that the current structure could be improved.

Counter-Proposal: Implement one meta-class under spatial region that contains the equivalent region-classes, for example (rough draft):

region of relevance: definition: A region of relevance is a spatial region that is mentioned in any study or analysis.

I am fine with this.

viktorwichern commented 8 months ago

Okay, so should I implement? @stap-m

stap-m commented 8 months ago

@adelmemariani does the OEKG break if the classification of study region etc change?

OpenEnergyPlatform / ontology