Consolidate human- and machine-readable Living Standard/Workstream lists

[domenic]

An ideal setup would be a single machine-readable (JSON) document listing:

• Workstreams
  • Name
  • Scope
  • Short, "database" name (used by participate.whatwg.org code)
  • Editor
    • Name
    • Email address
  • Living standards. Per living standard:
    • Authors (usually a superset of workstream editor; used for bibliographies and attribution, not decision-making power)
    • Short all-caps name for cross-references
    • Full citable title that includes "Standard" at the end
    • Description (currently seen on spec.whatwg.org)
• Non-workstream things people might cite
  • See BOOKS, DOMPARSING, FIGURES, JAVASCRIPT, LOADER in https://resources.whatwg.org/biblio.json

Right now this data is scattered across:

• spec.whatwg.org as HTML
• idea.whatwg.org as HTML
• whatwg/sg's Workstreams.md
• participate.whatwg.org config.json
• https://resources.whatwg.org/biblio.json

The new resource would live in whatwg/sg since the SG controls scope changes/the creation of new workstreams/etc.

From this new machine-readable data source, we would then

• Generate spec.whatwg.org
• Generate idea.whatwg.org
• Generate whatwg.org/workstreams
• Generate https://resources.whatwg.org/biblio.json
• Update participate.whatwg.org code to consult this canonical resource
• Update labels.py to consult this canonical resource

[guest271314]

What needs to be done? Where is help needed to realize the goal of this issue?

[annevk]

@domenic one problem with putting this in whatwg/sg is that changes to whatwg/sg can take a long time.

(Having this would also help with further automating our label setup.)

[annevk]

"whatwg/sg's Workstreams.md" and "participate.whatwg.org config.json" disagree on how Workstreams are named. Per discussion with @domenic the latter is likely wrong so those names will subtly change.

[domenic]

(The effect of this would be changes visible on https://participate.whatwg.org/agreement.)

[annevk]

whatwg/sg#111 and whatwg/whatwg.org#303 are my attempt at solving this. (I also added idea.whatwg.org as target.)

[domenic]

Amazing stuff; looking forward to reviewing!

You haven't done participate.whatwg.org's config.json yet, right? I'd be happy to take that on myself; in particular figuring out a way that uses Node.js tools and works for both local development and deploy.

[annevk]

Yeah, that's the one that's remaining. Thanks for taking that!

[annevk]

@domenic so for labels.py I need all repositories that host a standard. I think it's safe to assume that standards and repos are 1:1 and that the shortname of a standard in its URL is 1:1 with the repo name. Do you agree or do we need a specific shortname or reponame or both field in addition to the URL?

[tobie]

Let me know when we need to update specref.

[annevk]

@tobie I've made it so that no changes are required for you. I'll update OP. (My rationale is that biblio.json has a semi-standardized format and Steering Group's db.json does not and I'm not sure to what extent we want to commit to its stability, especially initially.)