[BUG] Issues with periglacial semantics

pbuttigieg commented 4 years ago

A bug report assumes that you are not trying to introduced a new feature, but instead change an existing one... hopefully correcting some latent error in the process.

Description

SWEET's phenCryo TTL file includes :

###  http://sweetontology.net/phenCryo/Glaciation
sophcr:Glaciation rdf:type owl:Class ;
                 rdfs:subClassOf sophcr:Accumulation ;
                 owl:disjointWith sophcr:GlacierRetreat ;
                 rdfs:label "glaciation"@en .

###  http://sweetontology.net/phenCryo/Periglacial
sophcr:Periglacial rdf:type owl:Class ;
                  rdfs:subClassOf sophcr:GlacialProcess ;
                  rdfs:label "periglacial"@en .

###  http://sweetontology.net/phenCryo/Periglaciation
sophcr:Periglaciation rdf:type owl:Class ;
                     rdfs:subClassOf sophcr:Glaciation ;
                     rdfs:label "periglaciation"@en .

Our work in the Semantic Harmonization Cluster and with the GCW / Polar Semantics team notes that periglacial processes or phenomena - by definition - cannot be subclasses of glacial counterparts.

What would you like to see changed?

Periglacial phenomena should be siblings of their glacial counterparts; however, note that periglacial processes need not be accumulation processes. This suggests that periglacial should be a subclass of "physical process". This change makes the class more general, which may cause some issues for users who have used it in a precise manner.

Note that this is part of the ENVO/SWEET alignment. Relevant ENVO classes include: http://purl.obolibrary.org/obo/ENVO_01001659 http://purl.obolibrary.org/obo/ENVO_01001641

Attribution

If you would like a nano-attribution, please use the following guidance.

The guidance doesn't seem to be up. The ESIP Semantics Harmonization Cluster can be credited.

pbuttigieg commented 4 years ago

Also, we note that "periglaciation" does not map to any phenomena we encountered. By definition, it cannot be a subclass of a glaciation.

pbuttigieg commented 4 years ago

Further, "periglacial" is a subclass of "glacial process" in SWEET, which cannot be (by definition). This is also a strange use of an adjective.

Also, "accumulation" is a subclass of "glacial process", which seems backwards.

Garybc commented 4 years ago

The links http://purl.obolibrary.org/obo/ENVO_01001659 http://purl.obolibrary.org/obo/ENVO_01001641 gave me a "term not found."

graybeal commented 4 years ago

I don't see them in the ontology. Maybe in an earlier version? (The copy I have is 2019.03.18 into BioPortal. Doesn't appear they were in the previous version either, 2018.05.18, based on a quick diff check.)

rduerr commented 4 years ago

No they are in an unreleased Editor’s version so haven’t been released

charlesvardeman commented 4 years ago

The semantic harmonization cluster meeting on 04/08/2020 worked on this topic. We note that the following concerns need to be discussed and addressed with the broader community. 1) How to deprecate terms that are not in agreement with scientific community use. 2) Provenance to account for harmonization cluster activity and procedure. This should be documented in the wiki. Issue #179.

pbuttigieg commented 4 years ago

Also linked to #126.

We're wrestling with which Annotation Property is appropriate for our harmonisation and mapping.

ESIPFed / sweet