Expand authorship of spec to all contributors

[dstansby]

Currently the zenodo record only lists one author: https://zenodo.org/records/4282107, but many others have contributed to the specification. It would be nice to recognise their contributions by expanding the authorship to anyone who's contributed to the specification.

[joshmoore]

See #175 which stalled out. Thoughts and opinions (and help) there welcome.

[jo-mueller]

@lubianat I'll hijack this issue for my suggestion about athuorship display from last week. To summarize current status:

• People (at Hackathon) agreed to have contributors to spec/schemas/RFCprocess as authors to the spec document over at ngff-spec
• I started adding author information to the project in WIP: Add authors to project metadata ngff-spec#31
• This author information renders nicely if the ngff-spec pages are built as a standalone project with Jupyter book
• The author information can be directly parsed into a citation.cff file
• Pure sphinx (which is used here) doesn't natively parse/render the author data

Starting from this, one could:

• Use Subproject ngff-spec under specfications directory #404 as a starting point
• Add some Python code to the pre_build.py over at ngff-spec that parses the author data and spawns a html object
• Insert the html at the top of the spec document.

This way, people would only need to keep the author metadata up to date in the project's myst.yaml and the authorship info would then be built and rendered each time when the pages are built and pre_build.py is executed.

[lubianat]

Thanks @jo-mueller.

I'll take a look at that. I wonder if a .zenodo.json is relevant for more fine grained things too (https://help.zenodo.org/docs/github/describe-software/zenodo-json/)

I'll go through the tickets and issues and try to get something on that tomorrow. I'll check:

• WIP: Add authors to project metadata ngff-spec#31
• Add CFF citation support and contribution instructions ngff-spec#25
• Add CITATION.cff #175

I'll also go through sphinx extensions (https://github.com/useblocks/awesomesphinx and https://github.com/sphinx-contrib to see if I can find something).

It should be straighforward to parse + inject the html, the hard part is doing it cleanly :)

Building a small extension (without guarantees and overfit for our case) could be cleaner than just a simple script.

[lubianat]

@jo-mueller I wonder if we can first get the #404 PR in and then get this authorship sorted?