Use case: maintaining a prescriptive context

[matentzn]

I want to be sure while I am promoting bioregistry around that the following use case can be captured.

We should be able to:

• [ ] define a project specific context, i.e. obo context
• [ ] associate some orcids with that context that declare us as the maintainers
• [ ] have some kind of default order for which prefix sources (prefix commons, OBO etc) we want to prefer, and the rest being suggested as "defaults" by bioregistry
• [ ] define the preferred prefixes for each URI expansion (example: we want to use DOI: for dois, not doi:)
• [ ] define the preferred URI expansions for each prefix (example: we want to expand DOI: to "https://doi.org/", not any other URL)
• [ ] we need a mechanism to fix preferred prefixes for each URI expansion (i.e. when we say: "preferred" somewhere and no matter what happens, we need to be able to rely on the fact that this preferred prefix cannot be changed without the explicit agreement of the maintainers documented above.
• [ ] we need a mechanism to fix the preferred URI expansions for each prefix (i.e. we say: "preferred" somewhere and no matter what happens, we need to be able to rely on the fact that this preferred prefix cannot be changed without the explicit agreement of the maintainers documented above.

Can we confirm whether this extensive set of requirements can be met by bioregistry or not? Thanks :)

cc @dosumis @cmungall

cf. https://obo-communitygroup.slack.com/archives/C01R2D66249/p1645718367637119

[cthoyt]

I'm still having a hard time wrapping myself around the ideas here.

Should this kind of thing even be in the Bioregistry? I'm very worried to have parts of the Bioregistry that should only be edited by a certain set of people. Will these people get onboarded to the Bioregistry review team? Then what happens if they're not responsive? Consider the various fiefdoms within the OBO Foundry project that have this issue.

Should I personally spend time supporting the development of this kind of thing? How much bloat should I allow into the Bioregistry to support people from a specific community?

Can we even plan for the Bioregistry to be prescriptive without massive prior buy-in? I don't want to commit to something that can't be enforce, either (consider the issues with enforcing OBO standards).

All that being said, I'm sure it would be possible to build something like this with the Bioregistry. The following code is what produces the OBO-context prefix maps:

https://github.com/biopragmatics/bioregistry/blob/6a978c441025c3eb97f8341a1a2eca56cb01081e/src/bioregistry/export/prefix_maps.py#L110-L124

This could be generalized to be stored in an external configuration file like this

```json { "obo": { "name": "Open Biomedical Ontologies Context", "description": "The OBO Context exists for the following reasons: ...", "maintainers": { "github": "matentzn", "name": "Nico Matentzoglu", "orcid": "0000-0002-7356-1779" }, "priority": [ "obofoundry", "default", "prefixcommons", "miriam", "ols" ], "use_preferred": true, "prefix_remapping": { "umls": "UMLS", "snomedct": "SCTID", "ensembl": "ENSEMBL", "doi": "DOI" }, "custom_prefix_map": { "DOI": "https://doi.org" } } } ```

That might already constitute 95% of what you want

[matentzn]

Lets have a quick catch up call when you have time and discuss how to move forward. Your suggestion sounds good, but I would like to talk it over - in case I am accidentally stretching the point of bioregistry here.