Re: Control of stuff that is not offer/answer (Re: Update of "Sorting issues into categories")

I would agree with Stefan that a lot of this can be addressed by something
like constraints mechanism, but this would require defining of additional
constraints or ability to provide/modify constraints without adding or
removing MediaStreams or MediaStreamTracks. For instance if all I want to
change the target bitrate for the video codec, why would I want to do
anything else except change the target bitrate. I would not want to do a
complete O/A exchange. I would not want to add and remove MediaStream
either. Stefan's proposals are good but it is in no way a complete solution
and would require more discussion.
_____________
Roman Shpount


On Thu, Oct 4, 2012 at 5:21 AM, Harald Alvestrand <harald@alvestrand.no>wrote:

> On 10/03/2012 07:40 PM, Roman Shpount wrote:
>
>> Once again, I have mentioned my concern for the shortcomings of the
>> current API in
>>
>> http://lists.w3.org/Archives/**Public/public-webrtc/2012Sep/**0102.htm<http://lists.w3.org/Archives/Public/public-webrtc/2012Sep/0102.htm>
>> <http://lists.w3.org/**Archives/Public/public-webrtc/**2012Sep/0102.html<http://lists.w3.org/Archives/Public/public-webrtc/2012Sep/0102.html>
>> >
>> It looks like some additional API surface is needed to address my
>> concerns ( http://lists.w3.org/Archives/**Public/public-webrtc/2012Sep/**
>> 0109.html<http://lists.w3.org/Archives/Public/public-webrtc/2012Sep/0109.html>).
>>
>> I am not sure why this is being ignored...
>>
> Because you didn't reply to http://lists.w3.org/Archives/**
> Public/public-webrtc/2012Sep/**0109.html<http://lists.w3.org/Archives/Public/public-webrtc/2012Sep/0109.html>where Stefan suggested that the present work on applying and modifying
> constraints seems like a reasonable approach to follow with regard to the
> issues you raised?
>
> Seriously, when the chairs say that "we think this is being addressed by
> ongoing work", the ball is back in your court to say whether you think this
> is correct or not.
>
>                      Harald
>
>
>