Build a JSON Schema Landscape

[benjagm]

It's great that we have the implementations and tooling page but that is just a portion of JSON Schema ecosystem. There are other players like adopter organizations or maybe partners that need's to be mapped.

It would be great to have a JSON Schema Landscape web space to visualize the real dimension of JSON Schema and the linux foundation actually provide a tool for that: https://lfx.linuxfoundation.org/tools/landscape/

We can make it very simple like the GraphQL Landscape:
https://landscape.graphql.org/

I'd essentially have these categories:

• Everything present in our implementations page:
  • Adopters
  • Sponsors
  • Partners
  • The tools will be managed in the implementations page

Resources:

• Linux Foundation landscape project that should make it easy to start mapping our ecosystem.
• List of published landscape: https://landscapes.dev/

Work Detail:

• List of work
• To be done

[Relequestual]

This is the repo provided by CNCF for creating new landscapes: https://github.com/cncf/landscapeapp

[Julian]

I think we're quickly gonna run into the fact that basically no tools have logos, no?

That repo says:

the hardest part of building a landscape is getting hi-res images for every project

Which I can believe, but even worse yeah e.g. my implementation is 15 years old and has no logo (though I'd love one at some point) -- but yeah I bet the norm is the vast majority not having them.

What would we do there?

[benjagm]

What would we do there?

Agreed on your concerns regarding logos. As an alternative we can use something like this Inner-source Portal Patterns to show projects and search on them.
https://patterns.innersourcecommons.org/p/innersource-portal

This is what Data on Kubernetes Community did: https://dok.community/landscape/

[benjagm]

Now that we launched adopters, this issues is going to be replaced by #160

[Relequestual]

Reopening.
Landscape diagram includes...
adopters, implementations, sponsors, projects, consuming standards, and probably a few other things I forget.

I think eventually we might use a single data source, but that will take some time.