Publish Web Audio API v1.1 FPWD

[hoch]

This PR is to publish FPWD of Web Audio API v1.1. The change includes followings:

• Update the version to 1.1 and the document stage to FPWD
• Collapse all the changes with mark-ups (proposed addition/correction) since Recommendation into the spec:
  • Use new Web IDL buffer primitives #2361
  • (OscillatorFolding): Oscillator folding considerations #127
  • Typo in VUMeterNode sample #2359
  • Fix typo (PR 2370) #2373
  • decodeAudioData and corrupted files #2321
  • The audio track to use in decodeAudioData isn't specified #2375
  • Explain that system-level audio callbacks exist #2475
  • API proposal to preemptively determine audio thread overload (Render Capacity) #2444
  • How to postMessage to AudioWorkletGlobalScope without creating a dummy worklet processor + node? #2456
  • Access to a different output device: AudioContext.setSinkId() #2400
  • AnalyserNode should recover from NaN? #2512
  • Allow user-selectable render quantum size #2450
  • Device-related error reporting via AudioContext.onerror #2567
• Remove JS code that handles the mark-ups.
• Remove the content change log section

References

• This change is approved by W3C: FPWD Request for Web Audio API 1.1 - web-audio-api-11 w3c/transitions#656
• This PR fixes Resolution: publish a new FPWD as version 1.1 #2602

---

Preview | Diff

[mjwilson-google]

LGTM

[mdjp]

Catching up a bit, but I am happy with this.

[hoch]

@padenot @svgeesus PTAL again!

[svgeesus]

This looks good to me

[svgeesus]

@padenot did you want to re-review, or ok to merge?

[svgeesus]

There are some broken links:

• https://www.w3.org/html/wg/drafts/2dcontext/html5_canvas_CR/
• https://pr-preview.s3.amazonaws.com/WebAudio/web-audio-api/pull/2603.html#c2359
• https://w3c.github.io/mediacapture-main/#dom-mediastreamtrack (lines 2058, 2789, 3157, 3160, 4434, 9276, 9357, 9385, 9386, 9387, 9400, 9404, 9414, 9465, 9490, 9536, 9541, 9543, 14570, 14979)
• https://w3c.github.io/mediacapture-main/#dom-mediastream (lines 2055, 2061, 2787, 3083, 3136, 3329, 9276, 9278, 9318, 9355, 9357, 9361, 9391, 9400, 9412, 9416, 9448, 9450, 9455, 9460, 14568, 14960, 14966, 14970)
• https://w3c.github.io/mediacapture-main/#dom-mediadevices (line 13069)
• https://w3c.github.io/mediacapture-main/#dom-mediadevices-enumeratedevices (line 3471)
• https://w3c.github.io/hr-time/#dom-domhighrestimestamp (lines 3548, 3557, 14592)

[hoch]

If that's okay, I'd like to keep this PR is about removing mark-ups. On a follow up PR, I can fix all the broken links.

[hoch]

Re @svgeesus:

• https://www.w3.org/html/wg/drafts/2dcontext/html5_canvas_CR/: [2dcontext] in spec ref is pointing an invalid link (404). Perhaps we can remove this link from the spec?
• https://pr-preview.s3.amazonaws.com/WebAudio/web-audio-api/pull/2603.html#c2359: Fixed
• These links don't seem to be broken. Not sure what to do:
  • https://w3c.github.io/mediacapture-main/#dom-mediastreamtrack
  • https://w3c.github.io/mediacapture-main/#dom-mediastream
  • https://w3c.github.io/mediacapture-main/#dom-mediadevices
  • https://w3c.github.io/mediacapture-main/#dom-mediadevices-enumeratedevices
  • https://w3c.github.io/hr-time/#dom-domhighrestimestamp

[mjwilson-google]

[2dcontext] in spec ref is pointing an invalid link (404). Perhaps we can remove this link from the spec?

It seems like this is now part of the HTML spec: https://html.spec.whatwg.org/multipage/canvas.html#2dcontext

Should we replace the reference with [HTML] instead?

[hoch]

I am not sure if that's a helpful change. If that's the only solution, linking "Canvas 2D" to the entire HTML spec is a bit misleading I guess?

I think the W3C link checker is a bit aggressive. It also complains about the document fragment (#). Not sure why this is not allowed.

[svgeesus]

I think the W3C link checker is a bit aggressive. It also complains about the document fragment (#). Not sure why this is not allowed.

It complains if it can't find the specific fragment in that document.

However it doesn't run JS on fetched resources so is vulnerable to false negatives.

[svgeesus]

I am not sure if that's a helpful change. If that's the only solution, linking "Canvas 2D" to the entire HTML spec is a bit misleading I guess?

We can add a local bibliography that defines [2DContext] and links to https://html.spec.whatwg.org/#2dcontext

specref should also be updated, but we don't have to be blocked on that.