search

EnvironmentOntology / envo

A community-driven ontology for the representation of environments
http://www.environmentontology.org
Creative Commons Zero v1.0 Universal
132 stars 51 forks source link

FoodOn "food material" placement under "environmental material" #1287

Open ddooley opened 2 years ago

ddooley commented 2 years ago

Currently FoodOn "food material" is positioned as:

material entity
  fiat object part
     environmental material
         food material

Yet litter, manufactured product etc. aren't under fiat object part or environmental material.

Can we just keep it simple and attach food material directly under material entity?

Also, ENVO has FoodOn product references from original handover of over 600 terms to FoodOn, but isn't importing all food products or food materials. Does ENVO really want to import all of those, or perhaps just keep a "food material" stub? This would probably solve a number of other FoodOn terms appearing at highest level of ENVO too.

dr-shorthair commented 2 years ago

See https://github.com/EnvironmentOntology/envo/pull/1252 for long thread on whether 'fiat' is useful or not.

wdduncan commented 2 years ago

@ddooley I agree that in general having food material under material entity is better. We might want to think if there are cases in which there are cases of "fiat" food entities.

cc @kaiiam @pbuttigieg @cmungall

kaiiam commented 2 years ago

Thanks @dr-shorthair very relevant here.

kaiiam commented 2 years ago

I believe this could simply be fixed in FOODON by asserting food material under material entity there.

wdduncan commented 2 years ago

@kaiiam I think this may be a good way forward to resolving this issue.

pbuttigieg commented 2 years ago

Yet litter, manufactured product etc. aren't under fiat object part or environmental material.

That is correct. (An arbitrary portion of) environmental material (water in the ocean, soil in the woods ...) is delimited by fiat.

Some, however, do have discontinuity-based boundaries: a bucket of water, a shovelfull of soil.

Some proposals in #1252 may offer a way to capture this deeper in a material entity > env material hierarchy, allowing us to infer if they're fiat.

Manufactured products and litter may be objects, object aggregates, or env materials, thus are bumped up a level of generality. The MP hierarchy is also populated by inference.

Can we just keep it simple and attach food material directly under material entity?

This is not simple, rather, it just avoids the semantic challenge of creating ontological hierarchies.

Food material (if you're talking about the material itself, see below) is very much an environmental material - composed of intended ingredients with other stuff in there too.

In FOODON, you have things under material that aren't materials

I would revise your hierarchies for internal consistency and upper-level alignment.

When classes were transferred from ENVO, it was under the assumption that FOODON would ontologise the very non-ontological thesauri and glossaries it harvested.

Also, ENVO has FoodOn product references from original handover of over 600 terms to FoodOn, but isn't importing all food products or food materials. Does ENVO really want to import all of those, or perhaps just keep a "food material" stub? This would probably solve a number of other FoodOn terms appearing at highest level of ENVO too.

No, we never considered importing all of FOODON, but we keep some classes we ceded over to FOODON to help communities that used/use ENVO for food terms. We're directing them directly to FOODON and other ontologies as their standards develop. We'll leave in a set of precomposed examples to help them (eventually) post-compose what they need as their ontological competencies increase.

We've solved the orphaned FOODON classes. Some of those were import artifacts due to equivalence axioms not being carried over and no subclass axioms being present.