Agenda & Notes

• Ontology release tooling (SHACL validation, Github Actions, automated release notes etc.) - Philip Strömert 
  
  • see also:
    • https://nfdi4chem.github.io/VibrationalSpectroscopyOntology/odk-workflows/
    • https://oboacademy.github.io/obook/reference/odk/
  • What would need to be done, if VIVO were to be migrated to ODK?
    1. install requirements (ODK Docker container and Make)
    2. build import modules for all existing VIVO ontology dependencies → probably the most time-hungry step as this might need some tweaking in the custom Makefile
    3. copy native VIVO terms into vivo-edit.

Ideas for future meetings

• After that the release tooling needs to be discussed, ideally with Philip Strömert, Melanie Wacker, Anna Kasprzik and Susanne Arndt (TIB). Topic for a meeting in January.

To Do

•  Enter tasks here, use "@" to assign a user, and "//", to add a deadline.

• • 1. owl, imported terms will be there via the import modules
       1. maybe think about, if you need/want a ROBOT tsv template pipeline as used in VIBSO
    2. activate documentation build pipeline and edit docs to entail what you need for your contributors
    3. make a release using the automated release build pipeline of ODK (sh run.sh make prepare_release)
    4. publish new release
    5. complete OBO tutorial for ODK: https://oboacademy.github.io/obook/tutorial/odk-tutorial-2/
  • Philip started a migration in: https://github.com/vivo-ontologies/vivo-ontology-ODK
    • all edits are made on the /src/ontology/vivo-edit.owl
    • ODK configuration is done in the src/ontology/vivo-odk.yaml
      • possible parameters of this config yaml → https://github.com/INCATools/ontology-development-kit/blob/master/docs/project-schema.md
    • import modules:
      
      • Philip already declared some standard OBO import: BFO, OMO, RO, IAO, OBI
        • BFO and OMO should always be mirrored as we want all of what is defined in them
        • already scraped all BFO declarations from the /src/ontology/vivo-edit.owl → BFO dependency is clean
      • working on FOAF & VCARD import module
        • there are assertions on FOAF classes (annotations & subclassOf) living in the /src/ontology/vivo-edit.owl ← annotations can probably be deleted when already present in the FOAF source
        • VCARD:sound is declared as an DP but is originally an OP??? → best to wait on cleaned up VIVO version before continuing the migration
    • Problems noticed so far:
      • first push led the CI/CD to fail because vivo is outside of the DL profile, which is considered a bad smell → some axioms somewhere that can hopefully be dropped, see also https://douroucouli.wordpress.com/2021/03/24/avoid-mixing-parthood-with-cardinality-constraints/
        
        • owl:sameAs was  declared as an OP → deleted

        • there was a weird blank node axiom → AnnotationAssertion(owl:minCardinality _:genid2147483649 "1"^^xsd:nonNegativeInteger) → deleted

        • owl:DeprecatedProperty was  declared as an OP → deleted

        • owl:minCardinality was declared as an annotation property --> deleted

        • fixed in https://github.com/vivo-ontologies/vivo-ontology-ODK/pull/1
      • http://purl.obolibrary.org/obo/ARG_ ← terms with this namespace only exist in VIVO ontology → why not mint it there instead then and not use these unresolvable IRIs?
  • about version history
    • this workflow depends on GitHub, so the better the PR and issue titles the easier it is to automatically let GH create a "what has changed" list from the merged PRs since the last tag, which will then be your release description
  • http://robot.obolibrary.org/diff.html
• Q by Egon Willighagen on Mastodon: https://mastodon.social/@egonw@social.edu.nl/111937329072554484 - Christian Hauschke 
  • Who is using CiTO in VIVO, in production?
    • CITO's partially in VIVO
    • some actual uses of it mentioned in https://jcheminf.biomedcentral.com/articles/10.1186/s13321-024-00807-2 (in the references)
    • there was a lot of interest in the past around the launch of CiTO, but no productive use is known.
  • Could the data be parsed/queried/scraped for export to Wikidata?
    • potentially yes, but no prod instance is known, see above.

Ideas for future meetings

• Release of 1.17.X