Agenda & Notes
- Ontology release tooling (SHACL validation, Github Actions, automated release notes etc.) - Philip Strömert
- see also:
- What would need to be done, if VIVO were to be migrated to ODK?
- install requirements (ODK Docker container and Make)
- 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
- copy native VIVO terms into vivo-edit.owl, imported terms will be there via the import modules
- maybe think about, if you need/want a ROBOT tsv template pipeline as used in VIBSO
- activate documentation build pipeline and edit docs to entail what you need for your contributors
- make a release using the automated release build pipeline of ODK (
sh run.sh make
prepare_release) - publish new release
- 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
- Philip already declared some standard OBO import: BFO, OMO, RO, IAO, OBI
- 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?
- 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/
- 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
- 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.
- Who is using CiTO in VIVO, in production?
Ideas for future meetings
- Release of 1.17.X
To Do
- Enter tasks here, use "@" to assign a user, and "//", to add a deadline.