Re: [xrblock] Proposed change to summary discard draft to address discard type issue

Hi Glen,

See my replies inline.

Best Regards!
Rachel

-----Original Message-----
From: Glen Zorn [mailto:glenzorn@gmail.com] 
Sent: Friday, September 21, 2012 11:36 AM
To: Huangyihong (Rachel)
Cc: xrblock; glenzorn@gmail.com
Subject: Re: [xrblock] Proposed change to summary discard draft to address discard type issue

On 09/21/2012 08:41 AM, Huangyihong (Rachel) wrote:
>
> Hi all,
>
> In the last IETF meeting, one open issue of summary statistics draft 
> (http://tools.ietf.org/html/draft-ietf-xrblock-rtcp-xr-summary-stat-00) raised 
> is to add some words to point out discard type chosen. Here is the 
> proposed text to address this issue.
>
> In section 3.2.2, the description of Gap Discard Rate field, the third 
> paragraph.
>
> OLD TEXT:
>
> "
>
> where "number of packets discarded" is obtained from the RTCP XR
>
> Discard Block [DISCARD] and Packets Expected is calculated as the
>
> difference between "extended last sequence number" and "extended
>
> first sequence number" (Interval or Cumulative) provided in the
>
> Measurement Information block [MEASIDENT].
>
> "
>
> NEW TEXT:
>
> "
>
> where "number of packets discarded" is filled with number of packets 
> due to both
>
> early and late to be played outwhich is obtained from the RTCP XR
>
> Discard Block [DISCARD], and Packets Expected is calculated
>
> as the difference between "extended last sequence number" and "extended
>
> first sequence number" (Interval or Cumulative) provided in the
>
> Measurement Information block [MEASIDENT].
>
> "
>

I would rewrite this as

"where "number of packets discarded" is filled with the number of 
packets discarded due to both early and late arrivalwhich is obtained 
from the RTCP XRDiscard Block [DISCARD], and Packets Expected is 
calculatedas the difference between "extended last sequence number" and 
"extendedfirst sequence number" (Interval or Cumulative) provided in 
The Measurement Information block [MEASIDENT]."

But even so, it would imply that two instances of the RTCP XR Discard 
Block are required, since a single Discard block can only report 
discards due to either early or late arrival, not both. Is that your 
intent? If so, then I would suggest making it explicit; if not, the 
maybe just leave the text as it is.

[Rachel]: I am okay with your proposed change. Also one instance is sufficient
since Discard draft supports reporting discards due to both.

On a different topic (but somewhat related), should the references to 
draft-ietf-xrblock-rtcp-xr-discard and 
draft-ietf-xrblock-rtcp-xr-meas-identity be normative? It seems to me 
that it would be impossible to fully understand this draft w/o 
understanding them, too.

[Rachel]: Good point and will change them into normative.

> Best Regards!
>
> Rachel
>
> Best Regards!
>
> Rachel
>
>
>
> _______________________________________________
> xrblock mailing list
> xrblock@ietf.org
> https://www.ietf.org/mailman/listinfo/xrblock