Indexers will translate GitHub/Gitlab/Gogs/Gitea/...'s API response into this ontology.
The grammar itself would mostly likely be JSON-LD so it is compatible with our other metadata (CodeMeta/schema.org)
Current options:
* create our own
* pick an existing forge's ontology, and map everything to it -> this would eventually turn into inventing our own ontology anyway
* [[ https://forgefed.peers.community/vocabulary.html | ForgeFed's ontology ]] (and ActivityPub) -> not suitable for us; at the moment it's a disjoint set from what we want (ForgeFed only cares about issues and PRs)
* GHTorrent -> I am told they are working on a generic way to represent data that isn't tied to GitHub, but I cannot find it
* Wikidata (eg. https://www.wikidata.org/wiki/Property:P9100 )
* DOAP