Using git for manuscripts

I think it would be interesting to share more of the experience of collaborating on manuscript writing in github.
I find that collaborating with Word with more than one person is a real pain: file versions, track changes becoming a jungle, comments getting lost, version control via email, people forgetting to send (push) revised draft to the main author, people revising the wrong version, merging two different revisions of the same version (branch merge), bibliographies...
Everyone knows this.
How does git+github make it better? You discuss this in _Use cases 3._, but maybe give an actual example from that project - an event in which the main author asked for contribution via an issue, two authors branched, edited, pushed, and the main author merged. Something like that.
I think that this is where you can really sell git for manuscripts (and of course the `draft_v1_v2_221012_final_final.doc` thing...), even for non-programming scientists.

Start by reading the existing Use cases 3 discussion and identify where manuscript-collaboration examples are documented. Add a concrete GitHub workflow covering a contribution request, author branches, edits, pushes, and the main author's merge, with enough context for non-programming scientists to follow it.

Written by the indexing model from the issue text.