OBOFoundry.github.io
OBOFoundry.github.io copied to clipboard
Make site with best practices for OBO ontologies that go beyond principles
Releases
- Keep your edit file (the file you change on a regular basis) and your release files strictly separate
- Imports should be merged into the main release file
Collaboration
- Tag your GitHub repository with the
obofoundry
tag so that people can find it: https://github.com/topics/obofoundry
@matentzn can you add instructions on the top of this issue how the best place to suggest new "recommendations"?
For example, I think all ontologies should disclose their funding situation/longevity plan before getting accepted. If development for an ontology will halt immediately following OBO Foundry acceptance, this should be considered before accepting. If a group commits to maintaining their ontology but then abdicates responsibility, then there should be some kind of note that shows this happened based on their original assessment
I think these kinds of recommendations are extremely difficult to proof - and it's also not so much a best practice as it would be an OBO foundry principle.. it seems like an obvious thing to add to the principles, but as far as I understand from many past discussions it's also misleading; for example PATO and RO have no core funding, but they get dragged along by other projects that rely on them (but not explicitly in a grant type of way). An alternative would be to ask for funding sources in the New Ontology Request.. not sure..
In any case, just making comments here is fine, I will collate the recommendations later and run it by the community for voting. Best practice though is not meant to be something that should be covered by OBO principles, just things that are great but sort of on the sidelines.
In terms of branding, bear in mind that we (and others) often refer to the principles as 'best practices', so keep that in mind when deciding what to call these. Straw man: 'principles extensions'.
This could be (and kind of already is) part of the OBOOK (https://oboacademy.github.io/obook/)
Also there are now two automated assessments I wrote:
- https://cthoyt.com/obo-community-health/ assesses how active repositories actually are (spoilers, most of them aren't active at all, and prove my theory that many become inactive immediately after acceptance)
- https://cthoyt.com/oquat/ assesses how semantically meaningful the references inside ontologies are (there's a lot to be desired)
Related issue collecting potential examples: #2453