Re: [rtcweb] Proposal for a JS API for NoPlan (adding multiple sources without encoding them in SDP)

Christer Holmberg <> Tue, 18 June 2013 16:24 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id B4D6221F9B08 for <>; Tue, 18 Jun 2013 09:24:47 -0700 (PDT)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -5.58
X-Spam-Status: No, score=-5.58 tagged_above=-999 required=5 tests=[AWL=0.368, BAYES_00=-2.599, HELO_EQ_SE=0.35, HTML_MESSAGE=0.001, MIME_8BIT_HEADER=0.3, RCVD_IN_DNSWL_MED=-4]
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id p9HFXcZerPjM for <>; Tue, 18 Jun 2013 09:24:42 -0700 (PDT)
Received: from ( []) by (Postfix) with ESMTP id 52ED521E808E for <>; Tue, 18 Jun 2013 09:24:34 -0700 (PDT)
X-AuditID: c1b4fb25-b7f4c6d000004656-a9-51c089bb0c61
Received: from (Unknown_Domain []) by (Symantec Mail Security) with SMTP id E3.E8.18006.BB980C15; Tue, 18 Jun 2013 18:24:28 +0200 (CEST)
Received: from ([]) by ([]) with mapi id 14.02.0328.009; Tue, 18 Jun 2013 18:24:27 +0200
From: Christer Holmberg <>
To: "Matthew Kaufman (SKYPE)" <>, =?iso-8859-1?Q?I=F1aki_Baz_Castillo?= <>, "" <>
Thread-Topic: [rtcweb] Proposal for a JS API for NoPlan (adding multiple sources without encoding them in SDP)
Date: Tue, 18 Jun 2013 16:24:27 +0000
Message-ID: <>
References: <> <> <> <> <> <> <> <>, <> <>
In-Reply-To: <>
Accept-Language: en-US
Content-Language: fi-FI
x-originating-ip: []
Content-Type: multipart/alternative; boundary="_000_7594FB04B1934943A5C02806D1A2204B1C3ABABCESESSMB209erics_"
MIME-Version: 1.0
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFnrGLMWRmVeSWpSXmKPExsUyM+Jvre6ezgOBBgfWMllM32djMePWWRaL tf/a2R2YPc41vGf3WLLkJ5PHrQeT2AKYo7htkhJLyoIz0/P07RK4M3oO/2Mr2FRTcfb5FvYG xk85XYycHBICJhJ35y9khbDFJC7cW8/WxcjFISRwmFGi6fwkdpCEkMAiRolFdyy7GDk42AQs JLr/aYPUiAjMZJTY/rqJCaRGWCBf4n77E2aQGhGBAolDOxkhzDyJ1xc1QCpYBFQlZr46CLaK V8BX4vrVP0wQq2ayShx8uYcZJMEpkCgx78A0FhCbEeie76fWgI1nFhCX+HDwOjPEnQISS/ac h7JFJV4+/gd1v5LEjw2XWCDq8yV+HT7LDLFMUOLkzCcsExhFZiEZNQtJ2SwkZRBxPYkbU6ew QdjaEssWvoaq15WY8e8QC7L4Akb2VYzsuYmZOenlRpsYgdF0cMtv1R2Md86JHGKU5mBREuf9 eGpXoJBAemJJanZqakFqUXxRaU5q8SFGJg5OEMEl1cBYMSs93Pm9esnhV/rGLVMfP5UMijuW e0RG/7PhGa3bZ+5XMCz6JyVi3Le7eIbG2z1736VOe3/9/dVI1+23X8zdN2uOb83nuduaG8NT LJ97BHpmzf7XWX7n686w+uv6yiG3rUv72LPY9GebFPHoOLa33knivmu7aovKfOtcJcWos4qr 1v7+JnNIiaU4I9FQi7moOBEAWcUuvHkCAAA=
Subject: Re: [rtcweb] Proposal for a JS API for NoPlan (adding multiple sources without encoding them in SDP)
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Real-Time Communication in WEB-browsers working group list <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Tue, 18 Jun 2013 16:24:48 -0000


The fact that you may not need ICE, or DTLS-SRTP, for your use-case is not related to SDP O/A, is it? SDP O/A itself doesn't mandate you to do ICE or DTLS-SRTP :)



Lähettäjä: [] Puolesta Matthew Kaufman (SKYPE)
Lähetetty: 18. kesäkuuta 2013 18:43
Vastaanottaja: Iñaki Baz Castillo;
Aihe: Re: [rtcweb] Proposal for a JS API for NoPlan (adding multiple sources without encoding them in SDP)

It isn't SDP that's the problem. It is the offer/answer semantics.

Here's a real simple example of the problem: If I have a web browser that can do RTP A/V, I should be able to send a form post or XHR to ask my security camera to start streaming me RTP video to a specific address and port. The browser doesn't need to do ICE connectivity tests, because it is only going to receive video.

I should be able to use a few lines of JavaScript to initialize the UDP/RTP receive port and wire it to a video window, set up the codec mapping, and ask it for its local address and port so I can tell the camera where to send things. But that isn't how it works at all. Instead I need to run, in JavaScript, the entire offer/answer exchange from the security camera's point of view, extract the relevant information from the SDP blob, and then send it off. (Never mind that I also need to have DTLS-SRTP added to my camera, even though I'm sending unencrypted RTP from it all over the rest of my LAN to other receivers that aren't browsers)

Same thing if I have a two-way radio system that can talk RTP and ICE and which has its own proprietary way of setting up connections... again, need to write a whole SDP parser *and* state machine to run the "fake" offer/answer exchange.

Pain. In. The. Ass.

Matthew Kaufman

From:<> [] on behalf of Iñaki Baz Castillo []
Sent: Tuesday, June 18, 2013 7:48 AM
Subject: Re: [rtcweb] Proposal for a JS API for NoPlan (adding multiple sources without encoding them in SDP)
2013/6/18 Robin Raymond <<>>
SDP is clearly the WRONG technical choice. It was wrong from the start but I think there was a great misunderstanding that it was required or SIP wouldn't be compatible with WebRTC. Since the strong majority at the table were SIP guys because they are the "established" industry it became the 'way to do it' despite how horrible it is for the future. So here we are today...

Dear WG Chairs,

With all due respect, IMHO there is enough material to reopen the "SDP or not SDP" debate so I would like to request it to the WG.

I would also appreciate that those in favour of mandating SDP as the core communication for WebRTC explain their rationale again (given the number of arguments against SDP and the frustration SDP is causing), and also that they give arguments and responses to all the SDP related issues exposed in this thread, that are nicely summarized in this mail:

Really thanks a lot.

Iñaki Baz Castillo