Re: [rtcweb] Where to specify ICE usage and the common transport

[Magnus Westerlund <[e-mail]>]

On 2012-07-12 06:31, Cullen Jennings wrote:
> 
> On Jul 6, 2012, at 24:29 , Magnus Westerlund wrote:
> 
>> The last part points to that we have 2 or 3 (depending on how you
>> see DTLS-SRTP) users of the datagram transport flow. As co-author
>> of one of those documents I would like to have a single point to
>> reference.
> 
> Why does your document need a reference to something that says if we
> have a UDP over HTTP mechanism or not? I'm worried that we are
> creating some ugly interdependencies in the layering that are not
> needed.

I want to have something to reference that defines the datagram
transport the RTP uses. Independently if the actual datagrams are direct
ICE establish path or over any variety or combination of relays over
HTTP or not.

I personally think we do need something that sufficiently describing and
normatively specifying the datagram transport that both RTP and
DTLS/SCTP uses. This is after all usage of ICE, likely variant. The
requirement to support STUN and some TURN version. The requirement to
handle a varaint of the demultiplexing that DTLS-SRTP discusses due to
the simultanous occurrance of STUN, RTP and DTLS packets at the same
level in the stack.

Cheers

Magnus Westerlund

----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM
----------------------------------------------------------------------
Ericsson AB                | Phone  [phone]
Färögatan 6                | Mobile [phone]
SE-164 80 Stockholm, Sweden| mailto: [e-mail]
----------------------------------------------------------------------