Re: [rtcweb] A few questions on draft-ejzak-dispatch-webrtc-data-channel-sdpneg-00

[Paul Kyzivat <[e-mail]>]

On 2/25/14 5:29 PM, Christer Holmberg wrote:
> Hi,
>
>>>> So these parameters are needed to properly initialize a channel when
>>>> the subprotocol is not specified.
>>>
>>> Why would someone NOT specify the subprotocol?
>>
>> Because there is no defined subprotocol for what they are doing?
>>
>> In the common webrtc use case with two browsers talking to the same web service, and the web
>> service feeding the needed javascript to the browsers, they just "know" the protocol they are using. > So they may just "make it up", and have nothing registered.
>>
>> It is when they need to interoperate with something else that it becomes important to name and
>> register the protocol.
>>
>>> Especially considering Richard's draft, I thought the whole idea was to be able to negotiate the
>>> subprotocol using SDP.
>>
>> That's for us, not them. :-)
>>
>> They will be using DCEP.
>
> Sure, but my question is about the usage of the attributes in SDP (Richard's draft).

Are you then suggesting that the subprotocol MUST be supplied in the SDP?

I think we need some careful consideration of what rules are common to 
all data channels, and which ones can be different when using SDP rather 
than DCEP. That should consider how something first built using DCEP can 
be migrated to using SDP when it is discovered that it needs to 
interoperate with a sip server.

Unclear if that discussion should take place on rtcweb or mmusic.

	Thanks,
	Paul