Use an "enhancement upvote" issues page

[choldgraf]

Over in executablebooks/ @chrisjsewell just merged in a pretty cool single-page javascript app that uses GitHub's GraphQL interface to collect all of the issues across executablebooks/ repos and sort them by the number of 👍 responses they've gotten. Here's the page:

https://executablebooks.org/en/latest/feature-vote.html

I think this could be a cool way to collect information across JupyterHub-land as well, are folks interested in something like this? Then we could encourage more people to 👍 things if they agree they'd like to see them happen.

I bet @hamelsmu would think this is cool too :-)

[chrisjsewell]

Thanks @choldgraf 😄 One disclaimer to note, I'm not completely satisfied with how the data is updated; originally I went for a "full" JS solution that updated the data dynamically (see https://jsfiddle.net/chrisjsewell/xjw5623y/latest/). However, (a) this meant it was possible to hit the rate limit of GraphQL requests and (b) I couldn't see an obvious way to keep the Auth Token secret. So now I've gone with a scheduled fetch workflow (https://github.com/executablebooks/meta/blob/master/.github/workflows/retrieve_issues.yml). The outstanding issue with this is that GitHub does not return a CORS header from the asset download, so I have to fetch via a proxy: https://cors-anywhere.herokuapp.com/https://github.com/executablebooks/meta/releases/download/issues/issues.json which feels a bit hacky

Any improvement suggestions let me know 👍

[consideRatio]

@chrisjsewell excellent work on that! It is brilliant! I'd love to see such functionality for JupyterHub repo's!

---

@choldgraf I also remember seeing the Sphinx 3 compatibility issue through this list, and reading in it I saw your "State of the situation" heading or similar which was written as an edit in the original issue not created by you - I thought that was a good way to inject something relevant from the maintainers.