Re: Separate proposal on layering/simulcast (was part of the "big proposal")

On Sat, Apr 19, 2014 at 10:09 AM, Bernard Aboba <Bernard.Aboba@microsoft.com
> wrote:

>   However​, you breach the question of scope. What's the scope of the
> layerId? I was assuming it was per RtpSender/RtpReceiver. But now you're
> suggesting that they have basically a global scope (across all
> RtpSender/RtpReceivers). That's an important question to answer. One one
> hand, it the JS more power. On the other, it is more complex. I'm usually
> inclined to stick with simple, not complex. How compelling do you think
> having dependent layers on different ports really is?
>
>
> [BA] This will only arise with Multi-Session Transport. It will not arise
> with Single-Session Transport, with either single or multiple sources.  As
> I understand it, VP8/9 is Single-Session Transport (and single source) and
> all H.264/SVC implementations I am aware of are Single-Session (some are
> single source, others multiple source).  So maybe we can say Multiple
> Session Transport (though not multiple source) is out of scope?
>

​If there isn't a really good reason to put it in scope, I'd suggest we
leave it out.  But it may be worth it to leave the layerId as a string
instead of an int because that would make it easier to change the scope
later if we need to.
​