Expose specification status

[foolip]

It would be useful to expose the specification status of features, to distinguish features with a new and unstable spec from ones with a specification that's gone through a lot of review and has consensus. Roughly, to answer the question "is it standards track?"

There are at least two places we can get this information from:

• The release.status and nightly.status fields in browser-specs
• The standard_track boolean in BCD

I would expect that we couldn't infer everything correctly from these sources, and that we'd still need some manual overrides.

[tidoust]

I would expect that we couldn't infer everything correctly from these sources, and that we'd still need some manual overrides.

There are specs whose status is incorrectly captured as "Editor's Draft" in browser-specs. We haven't prioritized reporting a better status for them because no one has really requested something more precise for now. We'd be happy to improve things in that space if web-features needs it! :)

A related question is whether the status property should contain some kind of normalized value. Today, it more or less contains whatever the spec says its status is, but that means specs come with different statuses for different organizations (and/or groups within these organizations) that roughly mean the same thing, for instance "Final Deliverable" for AOM specs, "Proposed Standard" for IETF RFCs or "Recommendation" for W3C.

But perhaps the question "is it standards track?" first needs to be expanded a bit. I see BCD defines standard_track as "part of an active specification or specification process", which can mean many things. For W3C specs, more than the status, "standards track" could perhaps be defined as "in scope of a Working Group" (as opposed to "under incubation in a Community Group" or "not adopted by any group"), which can also be inferred from the data in browser-specs.

[foolip]

Perhaps we do the work in browser-specs and then just look it up there via the spec URL that web-features already has. I've filed w3c/browser-specs#1181 for further discussion.

[ddbeck]

This come up in today's WebDX call (alongside the related issue on browser-specs). I wanted to know what are the intended applications for this?

The discussion on the call suggested that answering "is this eligible for Interop?" is one such application. I'm wondering what other concrete uses this might have, which might influence whether and how we expect consumers of this data to show this information to web developers.

[captainbrosset]

Here is the minutes from the WebDX call where this was discussed: https://docs.google.com/document/d/1ree75ImLZjf60lTZ3BhCaLHygxgywr7SBXp-q0xPs8A/edit#heading=h.1awrxbghdhfj