A new data visualizer centered on CPAN Testers data

[GeekRuthie]

As I was talking to folks about CPT data, I learned that there were a lot of potential uses of the data that aren't happening, because of the limitations of the current "matrix" format.

The matrix as it stands now is very useful for module authors, in seeing that their tests are failing on some platform or some version of perl at a given version of the module. It's a good thing, and it shouldn't go away.

But for someone who is, for instance, trying to decide which of the many YAML read/write modules we have on CPAN, can we build a tool that will let them explore them more? There's more data out there: CPANCover, Kwalitee, the upcoming security things, and more. If a system administrator or developer could visualize more of that, with good filtering tools that let them see potential pain points in an install or upgrade, you start to approach a "decision tool," which would be a huge enhancement.

Take a look at https://github.com/vasturiano/3d-force-graph?tab=readme-ov-file. This JS library, MIT-licensed, lightweight and feature-rich, could be used to visualize the data in a new application (maybe repurpose the Perl Magpie name here, since it would be collating data from lots of places?) Allow the user to select the meanings of features (colors, sizes, shapes, meaning things like "failing tests", "river depth," "has security metadata" and so on).

I'd love to hear thoughts on how to make that usable, and other places you can think of where useful information might be grabbed.

[preaction]

It might be interesting to see a single module's dependency tree as a directed graph like this. It might be a good visualization for seeing failures in the tree, like one of the dependencies failing its own tests for a given set of criteria.