Restructure cost and its subclasses

[LillyG901]

Summary of the discussion

See #1914

Type of change (CHANGELOG.md)

Update

• make cost (and all its subclasses) subclass of process attribute

Workflow checklist

Automation

Closes #1914

PR-Assignee

• 🐙 Follow the Pull Request Workflow
• 📝 Update the CHANGELOG.md
• 📙 Add #'s to term tracker annotation

Reviewer

• 🐙 Follow the Reviewer Guide
• 🐙 Provided feedback and show sufficient appreciation for the work done

[stap-m]

There is one axiom of social cost which can be deleted without discussion, which is the one to the obsolete class 'economic value'. Seems like we missed to remove it once we made the relation obsolete...

[madbkr]

I started the reasoner (used HermiT) and for me cost and all its subclasses are equivalent to owl:Nothing. Here is the explaination that Protegé offered:

So we have to make more changes if we want to move cost like this.

[stap-m]

Good catch @madbkr

[areleu]

I started the reasoner (used HermiT) and for me cost and all its subclasses are equivalent to owl:Nothing. Here is the explaination that Protegé offered:

So we have to make more changes if we want to move cost like this.

That is a tricky one, but I can imagine this will happen to any other quantity values that are directly converted to process attributes, at least those with the has unit axiom.

[stap-m]

We discussed in #1875 which former quantity values are actually quantities or process attributes. This PR is part of the restructuring. Both quantities and process attributes can be "measured" by quantity values, which are allowed for units. Therefore, I guess the solution is to change the axiomatisation: remove the "has unit" axioms for all process attributes and replace them with has quantity value linking to a corresponding quanity value.

[stap-m]

And the corresponding quantity value would be monetary value.

[LillyG901]

There is one axiom of social cost which can be deleted without discussion, which is the one to the obsolete class 'economic value'. Seems like we missed to remove it once we made the relation obsolete...

I found that monetary value and emission certificate price also have an axiom related to the obsolete class. Should I also remove them in this PR?
The axiom of emission certificate price is part of the axiom that defines it as an equivalent class. Can I remove this without issue?

[LillyG901]

I started the reasoner (used HermiT) and for me cost and all its subclasses are equivalent to owl:Nothing. Here is the explaination that Protegé offered:

Thanks for pointing it out. It seems my reasoner does not show inconsistencies until after it's synchronized. Just starting doesn't seem to work.

[madbkr]

I don't know if this is still a work in progress. But since you have some issues with the reasoner not working right I wanted to point out that my reasoner finds a lot of issues with the shared-axioms file now. It tells me the ontology is inconsistent with a lot of possible explanations. Everything, not just cost is now red in the file. I'll put one of the justifications here, maybe it helps.

[stap-m]

I removed some more obsolete axioms. The only classes below owl:Nothing are now social cost of emission and its subclass. Here the inference it hurt due to the axiom to emission rate, which we discussed to remove anyway, see #1914 (comment). I will remove them now, too.

One of the removed axioms was 'transport network component' SubClassOf 'obsolete has economic value' some cost. @LillyG901 could you open an issue to discuss a correct compensation axiom please?

[LillyG901]

I added issue #2009