Re: Bug in section 12.3

On 12/18/13, 12:40 AM, Stefan Håkansson LK wrote:
> On 2013-12-17 22:05, Nils Ohlmeier wrote:
>> In section 12.3 it says:
>>
>> ' A track in a |MediaStream|, received with an ||RTCPeerConnection|
>> <http://dev.w3.org/2011/webrtc/editor/webrtc.html#idl-def-RTCPeerConnection>|,
>> /MUST/ have its |readyState| attribute [GETUSERMEDIA
>> <http://dev.w3.org/2011/webrtc/editor/webrtc.html#bib-GETUSERMEDIA>] set
>> to |muted| until media data arrives.'
>>
>> But the enum for MediaStreamTrackState in
>> http://www.w3.org/TR/mediacapture-streams/ does not have a state 'muted'.
> Looking at
> http://dev.w3.org/2011/webrtc/editor/getusermedia.html#mediastreamtrack,
> I think what is meant is that the track should have the property "muted"
> being true until data arrives.
Fair enough. Then the whole sentence above should then be changed.
And BTW the same applies to section 12.2.2, step 2., bullet point 5.
"Initialize tracks' readyState attribute to muted."
>> I assume the sentence in the WebRTC draft should be changed to say "...
>> set to to new until media data arrives."
> I think it should be set to "live" actually - there is a source (the
> PeerConnection) attached.
>
As someone who started reading these documents quite recently I have to
say this is not obvious at all to me.
The whole section 4.3.1 seems to mostly consider the "local stream set" 
perspective.
If you look at it from a "remote stream set" perspective I would have 
expected that
a MediaStreamTrack starts initially in "new" once it was received via 
the RTCPeerConnection
and transition to "live" once you start to receive RTP on it.
Otherwise the sentence " In the |live| state, the track is active and 
media is available
for rendering at a ||MediaStream| 
<http://dev.w3.org/2011/webrtc/editor/getusermedia.html#idl-def-MediaStream>| 
consumer 
<http://dev.w3.org/2011/webrtc/editor/getusermedia.html#dfn-consumer>." 
from section 4.3.1 does not make much sense to me.

Best regards
   Nils Ohlmeier
   WebRTC QE
   Mozilla