Re: [rtcweb] Proposal for dealing with CNAMEs and MSIDs for synchronization

On 2013-02-12 08:37, Harald Alvestrand wrote:
> On 02/11/2013 02:10 PM, Stefan Håkansson LK wrote:
>>>> If this is indeed a problem, then there must be other ways. A setting
>>>> "privacy=on" on PeerConnection, or do not expose CNAME to the app (do
>>>> we really need to signal it?; and [1] indicates you can use other
>>>> CNAME's than those signaled).
>>> What I'd like to have is to have no statement that forces (even at
>>> SHOULD level) all local sources to have the same CNAME, and perhaps even
>>> an explicit statement that the sender can use different CNAMEs for local
>>> sources if it wants to (for whatever reason it wants to do so), as long
>>> as they are in different tracks.
>>
>> The problem I have with this is that it makes it impossible for
>> someone building a service to know whether streams will be synced (and
>> I mean in situations when syncing makes sense - not when the video is
>> delayed by 500ms) or not. This is fully up to the browser - it may be
>> that sometimes they are and sometimes not.
>
> I don't understand what the problem is here. I may be dumb.

Or I am :)

> If syncing is important to the guy writing the service, (he can't live
> with them being unsynchronized), he should put them in the same stream.
>
> Under your suggestion, it would be impossible to generate two
> MediaStreams that have local content and are *not* synced - I don't see
> a reason to outlaw that.

But under your suggestion, it could be synced or not, the app would have 
no way to control or know. It is basically random (and implementation 
dependent). This is my main problem, I would like a consistent behavior 
(secondary is that I would like it to be the same CNAME).

>
> The practical issue I can think of would be a source that had its own
> clock, not the system clock, but is otherwise a local source. Having the
> same CNAME would require resynchronizing that source. I don't know if
> any such sources exist in practice, but again, I don't want to outlaw them.
>
>
>>
>> I would prefer that all streams used the same CNAME (if it is signaled
>> or not is another question). If this is a privacy issue I think we
>> should solve that instead of having the browser use, or not use, the
>> same CNAMEs at its own will.
>
> CNAMEs are signaled in RTCP SRs. We don't have the option of leaving
> them out.

Exactly, but what goes in the RTCP need not to be made available to the 
application in all cases - why not hide CNAME?

>
>
> _______________________________________________
> rtcweb mailing list
> rtcweb@ietf.org
> https://www.ietf.org/mailman/listinfo/rtcweb