Re: [rtcweb] consent freshness [was RE: STUN for keep-alive]

[Randell Jesup <[e-mail]>]

On 9/22/2011 11:54 AM, Magnus Westerlund wrote:
> On 2011-09-22 17:09, Randell Jesup wrote:
>> There certainly may be use-cases where media in one (or both)
>> directions are inactive or DTX-muted for considerable periods of time
>> - remember, many of the use-cases (and unknown interesting future
>> uses) aren't "phone calls".
>>
>> In particular, I have some interesting use-cases bubbling in the back
>> of my brain where the data channel remains open at low/no activity
>> and the media channels are inactive or non-existant for long periods,
>> or where data traffic predominates and media are often inactive.
> Yes, I definitely believe you. Which only points that we will need to
> consider issues like how much background traffic that occurs just from
> having the PeerConnection up and being ready for sending media. Maybe
> mandate maximum levels when in practice inactive.
>
> It also has implication I think on the continued media consent. As the
> point raised Muthu is that when none sends media, there is in fact zero
> entropy in the RTCP RR compound packets. Thus RTCP can't be used for that.
I wonder: could we create a default reliable data channel in all
PeerConnections that's used (solely?) for a heartbeat/"consent freshness"?
In that case, we get to define the heartbeat and can mandate enough entropy.
(Receiving data on reliable SCTP (or TCP) over DTLS might be enough right there.)

-- 
Randell Jesup
[e-mail]