Re: [xrblock] publish FPWD of Identifiers for WebRTC's Statistics API; respond by 10 Oct 2014.

[Varun Singh <[e-mail]>]

Hi Rachel,

Comments are inline.

>> Thank you Colin for the feedback. I would like to hear other
participants'
>> opinions as well.
>>
>> Another thing to note: most metrics proposed in the current draft report
>> cummulative values, meanwhile the XR metrics can also report over shorter
>> interval. The IPPM registry combines the same metric with different
>> measurement intervals under one numerical identifier, while the xrblock
may
>> choose to distinguish between cumulative values and interval values. For
>> example: totalPacketsLost, intervalPacketsLost.
>
> [Rachel]: Correct me if I'm wrong. IPPM registry does not distinguish
> measurement duration. It regards interval or cumulative pattern as the
> fixed parameters which could be determined and input in the
> measurement system when needed. I think it makes sense since there's

True, the IPPM registry serves as input to designers, and therefore
combines different measurement intervals into one identifier.

> no need to create two different metrics name for different measurement
> duration. While in our draft, we only consider cumulative values.

If we consider adding RTCP SR and RR metrics, then two metrics fractionLost
and Jitter would not be cumulative, the interval is then inferred from the
reception
timestamp of the corresponding report. In which case our document would have
to explicitly state the measurement interval for each metric.

>So I think maybe we should just explicitly say it in the draft that the
fixed parameter "I" is set to 11 (Cumulative Duration)?
>

Yes this is possible, but is this sufficient?

>>
>> Cheers,
>> Varun
>>
>> On Thu, Oct 9, 2014 at 4:00 PM, Colin Perkins <[e-mail]> wrote:
>> > Hi Varun,
>> >
>> > My impression is that the IPPM metrics registry is broader than what
this
>> group would need for WebRTC. I expect
>> draft-ietf-xrblock-rtcweb-rtcp-xr-metrics will need to include a
registry of
>> metrics that are exposed via the WebRTC statistics API. That registry
might
>> refer to the metrics via the IPPM metrics registry, but I think it’s
separate to
>> (and a subset of) the IPPM metrics registry.
>> >
>> > Colin
>> >
>> >
>> > On 1 Oct 2014, at 16:22, Varun Singh <[e-mail]> wrote:
>> >> Thank you Dan for forwarding this. Feedback on the spec, metrics is
>> appreciated.
>> >>
>> >> In reference to WG's other work on RTCPXR-statistics, I would like to
>> >> hear feedback on setting up a registry, i.e., should we setup a new
>> >> one, reuse an IPPM registry. IIRC this was discussed briefly at the
>> >> last meeting (Al Morton sent a pointer earlier, details here:
>> >>
>> http://www.ietf.org/mail-archive/web/xrblock/current/threads.html#01660)
>> >>
>> >> Cheers,
>> >> Varun
>> >>
>> >>
>> >> On Wed, Oct 1, 2014 at 8:57 AM, Romascanu, Dan (Dan)
>> <[e-mail]> wrote:
>> >>> Please see below the W3C Call for Consensus for the first public
draft of
>> the WebRTC stats API. This work is related to the RTCWeb work in XRBLOCK.
>> >>>
>> >>> If you wish to discuss aspects related to the XRBLOCK work you are
invited
>> to do it here. If you want to send comments or participate in the W3C CfC
>> please send these directly to the W3C mail list.
>> >>>
>> >>> Thanks and Regards,
>> >>>
>> >>> Dan
>> >>>
>> >>>
>> >>>> -----Original Message-----
>> >>>> From: Stefan Håkansson LK [mailto:[e-mail]]
>> >>>> Sent: Wednesday, October 01, 2014 11:50 AM
>> >>>> To: [e-mail]
>> >>>> Subject: CfC: publish FPWD of Identifiers for WebRTC's Statistics
>> >>>> API; respond by 10 Oct 2014.
>> >>>>
>> >>>> Hi all,
>> >>>>
>> >>>> This is a Call for Consensus (CfC) to publish a FPWD of
>> >>>> 'Identifiers for WebRTC's Statistics API'.
>> >>>>
>> >>>> The draft is here: http://w3c.github.io/webrtc-stats/
>> >>>>
>> >>>> Please review and note any concerns by responding to this mail.
>> >>>> This CfC ends next week, Friday, 10 Oct 2014. If you need more
>> >>>> time, please indicate so.
>> >>>>
>> >>>> Please note that a FPWD is just that, a first public draft, so it
>> >>>> need not be perfect and subsequent change is possible; publishing
>> >>>> this draft should give it further review and attention.
>> >>>>
>> >>>> Please respond to this CfC (even a +1 is useful). Silence will be
>> >>>> considered agreement.
>> >>>>
>> >>>> Thanks
>> >>>> Stefan
>> >>>>
>> >>>> [1] http://w3c.github.io/webrtc-stats/
>> >>>
>> >>> _______________________________________________
>> >>> xrblock mailing list
>> >>> [e-mail]
>> >>> https://www.ietf.org/mailman/listinfo/xrblock
>> >>
>> >>
>> >>
>> >> --
>> >> http://www.netlab.tkk.fi/~varun/
>> >>
>> >> _______________________________________________
>> >> xrblock mailing list
>> >> [e-mail]
>> >> https://www.ietf.org/mailman/listinfo/xrblock
>> >
>> >
>> >
>> > --
>> > Colin Perkins
>> > https://csperkins.org/
>> >
>> >
>> >
>> >
>>
>>
>>
>> --
>> http://www.netlab.tkk.fi/~varun/
>>
>> _______________________________________________
>> xrblock mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/xrblock

--
http://www.netlab.tkk.fi/~varun/