Re: [rtcweb] #27: Section 6.2 FEC

Magnus Westerlund <magnus.westerlund@ericsson.com> Tue, 27 August 2013 06:03 UTC

Message-ID: <521C414D.2080802@ericsson.com>
Date: Tue, 27 Aug 2013 08:03:57 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Mozilla/5.0 (Windows NT 6.1; rv:17.0) Gecko/20130801 Thunderbird/17.0.8
MIME-Version: 1.0
To: Bernard Aboba <bernard_aboba@hotmail.com>
References: <066.f62f1912f660dbc0c28343d2955a2ef5@trac.tools.ietf.org> <081.60ad42bfcba9b4972ad06bf3f62ca73e@trac.tools.ietf.org>, <5A2C089B-BBFF-42D1-949A-AD2984EE3E90@csperkins.org> <BLU169-W916B4ECB57EB451B8CC70393490@phx.gbl>
In-Reply-To: <BLU169-W916B4ECB57EB451B8CC70393490@phx.gbl>
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Cc: "rtcweb@ietf.org" <rtcweb@ietf.org>, Colin Perkins <csp@csperkins.org>
Subject: Re: [rtcweb] #27: Section 6.2 FEC
Precedence: list

On 2013-08-26 20:13, Bernard Aboba wrote:
> Colin said: 
> 
> "Again, that all seems reasonable, but doesn't address the issue of what
> type of FEC to use. Like Magnus, I don't see working group consensus to
> recommend a particular FEC scheme. If you believe there is consensus on
> using a particular FEC scheme, make a proposal, and we'll incorporate it
> if accepted. "
> 
> [BA] I assumed (rightly? wrongly?) that the FEC section of the Unified
> Plan document was put there with the intent to enable FEC to be
> indicated in the (SDP-based) API. Assuming that was the intent, and that
> the capability is developed further,  it would be useful for the
> supported FEC schemes to be documented somewhere, so as to enable
> interoperability.  I'll leave it to those who worked on the document to
> clarify where the RTX/FEC section is headed.  

To my understanding unified plan includes this because it is considered
a general requirement to support this type of functionality for any
multi media stream usages of RTP sessions as indicated by SDP. Thus the
discussion makes sense in the context of unified plan as being important
to show that it can be done.

When it comes to the rtp-usage document I don't see any issues with
negotiating the usage of FEC mechanism of the implementors choice using
what ever negotiation mechanism one has, at this stage presumably SDP
O/A. However, unless the WG can come to a consensus on what the FEC
mechanisms and their usages is to be, describing all the options and
deployment choices is a significant task that I frankly don't see as
improving interoperability. Thus, unless the WG do come to consensus on
some specific choices I find it difficult to write anything in RTP usage.

Bernard, if you have an idea of what you like to see here, can you
please propose some text.

Cheers

Magnus Westerlund

----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM
----------------------------------------------------------------------
Ericsson AB                | Phone  +46 10 7148287
Färögatan 6                | Mobile +46 73 0949079
SE-164 80 Stockholm, Sweden| mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------

[rtcweb] #27: Section 6.2 FEC rtcweb issue tracker
Re: [rtcweb] #27: Section 6.2 FEC Magnus Westerlund
Re: [rtcweb] #27: Section 6.2 FEC rtcweb issue tracker
Re: [rtcweb] #27: Section 6.2 FEC Colin Perkins
Re: [rtcweb] #27: Section 6.2 FEC Bernard Aboba
Re: [rtcweb] #27: Section 6.2 FEC Roni Even
Re: [rtcweb] #27: Section 6.2 FEC Bernard Aboba
Re: [rtcweb] #27: Section 6.2 FEC Magnus Westerlund