Re: [Dart] draft-dart-dscp-rtp

It would be nice if we could close on the multi-stream question really soon now. I don't think I see a consensus here so far.

Perhaps this is an area where we should punt and indicate it needs further study?

Thanks!

Ben.

On Aug 29, 2014, at 10:14 AM, Black, David <david.black@emc.com> wrote:

> Harald,
> 
> I'm about to submit a -05 with the indication that the single DSCP recommendation
> for SCTP and DCCP may be revised.  The RTCP multi-stream optimization text will
> still be in there with Colin's clarification about "received" streams.  I'm about
> to vanish for about 3 weeks, but could put in a revised -06 over the weekend 
> if you can quickly convince Colin.
> 
> Thanks,
> --David
> 
>> -----Original Message-----
>> From: Harald Alvestrand [mailto:harald@alvestrand.no]
>> Sent: Friday, August 29, 2014 6:49 AM
>> To: Ben Campbell; Black, David
>> Cc: dart@ietf.org
>> Subject: Re: [Dart] draft-dart-dscp-rtp - way forward
>> 
>> I think the text in -04 looks reasonable.
>> 
>> I would like to do two things:
>> 
>> - The justification for "single DSCP" in SCTP and DCCP is "we don't know
>> what will happen".
>> A logical consequence is that this recommendation might change as the
>> result of further study, so we might want to add near the bottom of
>> section 5 a sentence saying "This recommendation may be revisited if
>> experiments and analysis shows compelling reasons to change it."
>> 
>> - I still think the ban against RTCP multi-stream reporting optimization
>> is completely wrong. The reports will correctly reflect the differing
>> packet loss behaviours of the incoming RTP streams, and that's all they
>> should be required to do.
>> 
>> I regard the latter as somewhat important. The former is more a matter
>> of taste (if we're arguing based on ignorance, allow for the case where
>> ignorance becomes knowledge).
>> 
>> 
>> On 08/28/2014 10:25 PM, Ben Campbell wrote:
>>> Okay, works for me.
>>> 
>>> Thanks!
>>> 
>>> Ben.
>>> 
>>> On Aug 28, 2014, at 3:15 PM, Black, David <david.black@emc.com> wrote:
>>> 
>>>>> If we think we can close this as not a real issue (or at least not an
>> issue we
>>>>> plan to address), but there's some risk it might come back up at IETF LC,
>> then
>>>>> that's fine. But if we are saying there is an open issue that needs
>> closure,
>>>>> but we can close it during IETF last call, then I have reservations. The
>>>>> version that goes to IETF LC should be one that the working group believes
>> to
>>>>> be as complete is it's going to get.
>>>> In private discussion, both Harald and I favored saying less (rather than
>> more)
>>>> about RMCAT in this DART draft.  For that reason I think "at least not
>>>> an issue we plan to address" is a valid characterization of this item.
>>>> 
>>>> Thanks,
>>>> --David
>>>> 
>>>>> -----Original Message-----
>>>>> From: Ben Campbell [mailto:ben@nostrum.com]
>>>>> Sent: Thursday, August 28, 2014 3:44 PM
>>>>> To: Black, David; Harald Tveit Alvestrand
>>>>> Cc: dart@ietf.org
>>>>> Subject: Re: [Dart] draft-dart-dscp-rtp - way forward
>>>>> 
>>>>> (as chair)
>>>>> 
>>>>> On Aug 28, 2014, at 1:12 PM, Black, David <david.black@emc.com> wrote:
>>>>> 
>>>>>> [E] Open issue: Harald Alvestrand's concerns about "differential
>> treatment"
>>>>>> wrt Section 5.1 .  I don't completely understand these concerns, and
>> suspect
>>>>>> that they may require an email discussion across the DART and RMCAT WGs
>>>>>> to sort through.
>>>>>> 
>>>>>> --> Deferred - Harald wanted to think about this, and I've seen nothing
>>>>>> --> further.  I'd suggest that IETF Last Call as an appropriate
>> opportunity
>>>>>> --> to share any further thoughts, so I don't think there's anything to
>> be
>>>>> done
>>>>>> --> about this now.
>>>>> I have mixed feelings on this.
>>>>> 
>>>>> If we think we can close this as not a real issue (or at least not an
>> issue we
>>>>> plan to address), but there's some risk it might come back up at IETF LC,
>> then
>>>>> that's fine. But if we are saying there is an open issue that needs
>> closure,
>>>>> but we can close it during IETF last call, then I have reservations. The
>>>>> version that goes to IETF LC should be one that the working group believes
>> to
>>>>> be as complete is it's going to get.
>>>>> 
>>>>> David, thoughts?
>>>>> 
>>>>> Harald, do you care to comment further?
> 
> _______________________________________________
> Dart mailing list
> Dart@ietf.org
> https://www.ietf.org/mailman/listinfo/dart