Axiomatise origins

[l-emele]

Description of the issue

The definitions of origin and its subclasses indicate relations between origins and portions of matter and/or energies. However, these are not axiomatised.

Ideas of solution

• Origin is a quality of a portion of matter or energy based on where it comes from. It is inherited from its primary sources. => origin 'has bearer' some ('portion of matter' or energy)
• Anthropogenic is an origin of portions of matter or energies created by human activity. => anthropogenic 'has bearer' some ('portion of matter' or energy)
• Synthetic is an anthropogenic origin of portions of matter created artificially by a chemical process. => synthetic 'has bearer' some 'portion of matter'
• Biogenic is an origin of portions of matter made by or produced from life forms. => `biogenic 'has bearer' some 'portion of matter'
• Conventional is an origin of energies that don't replenish when transformed / consumed. => `conventional 'has bearer' some energy'
• Geogenic is an origin of portions of matter or energies that are the result of geological processes. => geogenic' has bearer' some ('portion of matter' or energy)
• Fossil is an geogenic origin of portions of matter created from organic material by geological processes lasting thousands or millions of years. => fossil 'has bearer' some 'portion of matter'
• Renewable is an origin of energies that replenish on a human time scale. => renewable 'has bearer' some energy'

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]

The other option would be to axiomatise this the other way round: energy 'has origin' some origin 'portion of matter' 'has origin' some origin

After thinking a bit, this seems to be better, as we already have for most subclasses of energy and portion of matter the axioms in that structure.

[l-emele]

If we implement as proposed in my last comment, the structure would look like this (two red arrows)[^1].

If we do it, as proposed in the proposed solution of the issue start, we would create a cycle of relations. But according to @jannahastings' Primer on ontologies such cycles should be avoided.

[^1]: For the third red arrow, see #1167

[l-emele]

@stap-m : Any thoughts on this? Would be nice to solve this issue soon that we can finalise the picture for the paper.

[jannahastings]

@l-emele In the older OBO-format ontologies, it was considered best practice to avoid cycles of relationships in order to avoid impacting algorithms that traversed ontologies as graph structures recursively and would be sensitive to such cycles. In the OWL language, this problem does not usually arise with object properties so I think we can certainly use cycles of relationships where they make sense.

[l-emele]

Okay, if that is not a problem, than there is also a third option: Axiomatising the relations in both directions by e.g. adding

• origin 'has bearer' some ('portion of matter' or energy)
• energy 'has origin' some origin
• 'portion of matter' some origin

The pro is that this would make it easier to understand these concepts for non-ontology experts. The con is that would be a bit more work for further maintenance of the OEO. But I think in this case the pro might weigh more than the con.