Build a matrix of all projects that implement the doctools specification.

[zivbk1]

I am not sure what the source data of the matrix should be. It could be as simple as a markdown table, or maybe a .json file where each project is an object in an array that is rendered into a nicer table, or set of pages on the good docs website.

One way or another, we need to create a mini dataset of records about various documentation template projects that are compatible with the goals of this project.

[mgan59]

I am not sure what the source data of the matrix should be. It could be as simple as a markdown table, or maybe a .json file where each project is an object in an array that is rendered into a nicer table, or set of pages on the good docs website.

Kicking around some ideas here. My one concern is if it were a single .json file it could get unmanageable and prone for it breaking if someone left out a comma or mis-placed something... And yes we could have an action doing linting on the json but again it is still easily broken.

A single markdown file also could get super large and markdown tables aren't that great.

I'd suggest that the source-data-matrix could be an aggregate of multiple data-files. Maybe we have a set a folder that represents a project and inside of that could be a json shape or a markdown file. If using markdown fan of yaml as front-matter.

I kinda lean towards the idea that the matrix is markdown with yaml and we define or technical specifications are front-matter/yaml. And the markdown body could be information around the implementation and other notes, etc.....

Thoughts?

[bwklein]

I really like the markdown file approach. It makes it easy for a starter kit project to submit a PR to add in their offering to the matrix and we can use front matter (I also prefer yaml) to store any boolean values and additional 'data' about the project that would be used in the project page and to drive the check boxes and details within the summary matrix.