Consider cross-spec device IDs

[jyasskin]

https://w3c.github.io/mediacapture-main/#idl-def-MediaDeviceInfo defines the groupId field to identify a physical device. http://webaudio.github.io/web-midi-api/#idl-def-MIDIPort defines the id field to identify the device. https://webbluetoothcg.github.io/web-bluetooth/#bluetoothdevice defines the instanceId field to identify a device. A single physical device could appear in more than one of these APIs, and it'd be nice to let web pages figure out that it's a single device, to the extent the browser knows.

@cwilso @slightlyoff (I'm not sure if it makes to file this in a couple of the affected specs, or if it should be a TAG issue.)

[jan-ivar]

Unfortunately, "device" is used to describe all sorts of different things. For instance, MediaDeviceInfo differentiates between "device" (deviceId) and "physical device" (groupId)!

Looking at your list:

• groupId is a data-less grouping token for mic and camera belonging to the same "physical" hardware. Not persisted to disk, but stable within a browser-session and not origin-unique (I'm surprised no-one's picked up on this fingerprinting issue). Did you mean deviceId?
• deviceId is per audio/video etc. device, origin-unique, and persistent only after certain criteria. Persist deviceIds only after gUM permission grant. #218
• id for MIDIPort - after a quick read this also seems not origin-unique. Has it been privacy vetted?
• instanceId for bluetooth seems origin-unique and scoped to the grant-window of the device, which I assume means it's persistent.

These all seem tailored to their domain, and not easily interchangeable to me.

[jyasskin]

Thanks @jan-ivar. Since this is a cross-spec issue, @slightlyoff suggested I open w3ctag/design-reviews#64, which I believe incorporates your comments.

I'll file a new issue about groupId's origin-uniqueness. MIDIPort.id is addressed in WebAudio/web-midi-api#48.