Re: [MMUSIC] SDP Directorate: Review of draft-ietf-xrblock-rtcp-xr-qoe-08

["Ali C. Begen (abegen)" <[e-mail]>]

- I don't agree with the following statement:

   Information is recorded about QoE metric which provides a
   measure that is indicative of the user's view of a service.

The application cannot possibly know a user's view of service unless the user explicitly enters that info in some fashion to the application.

[Qin]:  The application can rely on QoE evaluation model to calculate MoS value and use such MoS value to reflect how end user feel or experience the media quality. That’s what this statement want
to convey.

That is not what the sentence says, though, at least to me. The "user's view of a service" is the troubling part to me. What you do is that you apply an algorithm with some observed values and you try to estimate some sort of a mean opinion score. That may or may not be indicative of the end-user's QoE.


- Section 4.1: "extmap" is already a registered attribute name. I did not understand why and in what way you want to use it.

[Qin]: We need a new attribute defining the mapping from the different parameter names (i.e., the different calculation-algorithm parameter names) to the numeric values used in the packets Unlike RFC5285 extmap usage, we map name rather than URI to the numeric values used in the packet.
So I assume “extmap” can be reused, however if not, I believe we should define new attribute similar to “extmap” in this document.

I'd refer to some SDP experts here as I could not understand what you wanna do and how you wanna do it.