Re: [webrtc-stats] Add RTCInboundRtpStreamStats:decoderFallback (#724)

@fippo thanks you for the update. As I already mentioned, on #725, please consider the following aspects of the new metric.

I agree that new metric could also expose the HW capability at some instance of the time. 

However I would like to note that what the fingerprint guidance concerns is not only the hardware exposer of the user agent(browser) but risk of the fingerprinting of the user agent as defined by <https://w3c.github.io/fingerprinting-guidance/> as follows:

_**..., browser fingerprinting is the capability of a site to identify or re-identify a visiting user, user agent or device via configuration settings or other observable characteristics.**_

To share the transitions of this metric, it may look like following:
- In case of no HW Decoder:  stays false
- In case of HW Decoder: starts with false and turns to true only when the decoder gets fallback and changes to false again when the decoder gets recovered to HW.
 
So the current metric could not be used for identifying the user, user agent or device because of ephemeral characteristics of decoder fallback. 

And when we consider the cloud gaming's general use case, the camera and microphone permission are not required  because it works as one-way streaming from the gaming server to the browser client, so it does not require user permission for the hardware access.

And we could detect any streaming abnormalities through the decoder fallback from decoder implementation metric and it is currently blocked by the fingerprint concern by [#712](https://github.com/w3c/webrtc-stats/issues/712).

That's why I proposed this new metric which might not expose the finger print of the user agent directly unlike decoder implementation and power efficiency metrics.

-- 
GitHub Notification of comment by xingri
Please view or discuss this issue at https://github.com/w3c/webrtc-stats/issues/724#issuecomment-1397407628 using your GitHub account


-- 
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config

Received on Thursday, 19 January 2023 18:12:15 UTC