Re: On babies and bathwater (was Re: [rtcweb] Summary of Application Developers' opinions of the current WebRTC API and SDP as a control surface)

On Mon, Jul 22, 2013 at 1:37 AM, Rob Manson <roBman@mob-labs.com> wrote:

> On the topic of the gdocs survey that Peter T setup...for my feedback I've
> tried to use little more positive language e.g. not "strongly dislike" -
> but instead "strongly like a change" 8)
>
> I'm actually really impressed with and enthusiastic about what has been
> achieved for WebRTC so far.
>
> Yet I also think there's a compelling case for making a change now before
> 1.0 is finalised that will make WebRTC significantly better (less fragile
> and more extensible).
>
> I think the issues laid out in http://tools.ietf.org/html/**
> draft-raymond-rtcweb-webrtc-**js-obj-api-rationale-00<http://tools.ietf.org/html/draft-raymond-rtcweb-webrtc-js-obj-api-rationale-00>are very clear and well thought out.
>
> Also, in the gdocs survey I think the "OK/Good for simple things" question
> is a bit misleading.
>
> Firstly, "OK" and "Good" really are two separate things.
>

Yes, I originally had two separate buckets, but that ended up with too many
buckets, so I merged them.  Basically, I found that distilling all that
feedback into a readable summary was hard :).   Some precision was lost in
the simplification.


>
> Secondly, this initially led me to think about just a basic video call and
> I answered "yes". But when you think about "mute" or "on hold", etc. which
> are pretty simple things too then the answer clearly has to be no.
>
> I also think how this all relates to the whole RTCDataChannel model is
> important. Both for channel setup and for the ability to use
> RTCDataChannels for signaling and how that relates to both SDP and O/A.
>
> Just a few thoughts.
>
> roBman
>
>
>
> On 21/07/13 17:51, Stefan Håkansson LK wrote:
>
>> On 7/21/13 8:28 AM, Rob Manson wrote:
>>
>>> Thanks Stefan.
>>>
>>> Yep...will do.
>>>
>>
>> I can see your input now, thanks!
>>
>>
>>> roBman
>>>
>>>
>>> On 21/07/13 15:46, Stefan Håkansson LK wrote:
>>>
>>>> Rob,
>>>>
>>>> Peter T has a spreadsheet at
>>>> https://docs.google.com/**spreadsheet/ccc?key=**
>>>> 0AuaKXw3SkHMSdHlZdV9RN0xSWFhyb**Vl4anJLRkVPV0E#gid=1<https://docs.google.com/spreadsheet/ccc?key=0AuaKXw3SkHMSdHlZdV9RN0xSWFhybVl4anJLRkVPV0E#gid=1>
>>>> collecting the experiences made when using the WebRTC APIs. Could you
>>>> enter what you have learned? The more people that contribute the better.
>>>>
>>>> Stefan
>>>>
>>>> On 7/21/13 1:08 AM, Rob Manson wrote:
>>>>
>>>>> +1
>>>>>
>>>>> As a web developer that's spent a lot of time experimenting with the
>>>>> currently specified version of the WebRTC related APIs and that's been
>>>>> following the mailing list debates closely this really does seem like
>>>>> the best resolution.
>>>>>
>>>>> It provides a more extensible and flexible architecture that can evolve
>>>>> at "web developer speed" not "aligned browser release speed". And at
>>>>> this speed it will also be less fragile.
>>>>>
>>>>> It provides a clear separation of concerns so people can use SDP where
>>>>> they want, but not everyone is restricted by the timelines of other WGs
>>>>> that are required to evolve SDP.
>>>>>
>>>>> And it would enable even more experimentation and future facing
>>>>> development too.
>>>>>
>>>>>
>>>>> Also, in terms of timing I think getting this right is more important
>>>>> than the current commitment to a deadline.
>>>>>
>>>>> This is from the perspective of a web developer that has gone to all
>>>>> the
>>>>> effort of just finishing a book on "Getting started with WebRTC" using
>>>>> the existing API and who is also working on several commercial projects
>>>>> based on the current API.
>>>>>
>>>>> So if anyone should be promoting "just get the first version out" then
>>>>> it should be someone in my position. But I think you really will find
>>>>> that most web developers would rather we got this abstraction right
>>>>> first so we can avoid all of the extra support issues and application
>>>>> re-work that will be required down the track if we don't.
>>>>>
>>>>> roBman
>>>>>
>>>>>
>>>>> On 20/07/13 23:51, Iñaki Baz Castillo wrote:
>>>>>
>>>>>> Let W3C experts to define a good JS API for WebRTC (with no SDP), let
>>>>>> MMUSIC WG to define a SDP format for WebRTC, and then let JavaScript
>>>>>> SIP
>>>>>> experts to build JS libraries on top of it to play the SDP game, and
>>>>>> we
>>>>>> all will be happy. And telcos will be much more happy than they think.
>>>>>> Let's get rid of all the SDP O/A stuff in the browser. The browser is
>>>>>> not a phone and "fixed logic + fixed code" does not work here.
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>
>>>
>>
>>
>