Comments on: WebRTC Signaling Protocols and WebRTC Transport Protocols Demystified https://bloggeek.me/webrtc-signaling-transport/ The leading authority on WebRTC Sat, 28 Dec 2019 15:14:31 +0000 hourly 1 By: Sergey Markovich https://bloggeek.me/webrtc-signaling-transport/#comment-118521 Tue, 17 May 2016 01:27:02 +0000 https://bloggeek.me/?p=10339#comment-118521 In reply to Shair Khan.

Shair Khan,

The problem is that if your Browser is behind the NAT, it has a private address like 192.168.*.*, etc. Sharing only this address (we call it “local candidate”) with the peer (via socket.io or something else) does not make any sense because the peer will not be able to use this address to contact you. So ICE is needed to discover more “candidates” (addresses) – like your public address on the public side of the NAT and probably allocating another address on the TURN server. Then you share ALL your candidates (addresses) with the peer via socket.io or any other mechanism. This is done by a special SDP extension. The peer does the same and shares the list of his candidates with you. After that the connectivity testing starts – the goal of this process is to find the pair of candidates (one yours and one of your peer) that will work TOGETHER, i.e. media will flow in both directions.
This is very brief and simplified description 🙂

]]>
By: Shair Khan https://bloggeek.me/webrtc-signaling-transport/#comment-118520 Mon, 16 May 2016 23:36:41 +0000 https://bloggeek.me/?p=10339#comment-118520 Let me ask a stupid question.

Why do we need an elaborate icE and Turn and Stun servers to discover each other address? Why we cannot use socket.io to exchange the address with the peers?

]]>
By: Tsahi Levent-Levi https://bloggeek.me/webrtc-signaling-transport/#comment-118519 Mon, 16 May 2016 16:39:21 +0000 https://bloggeek.me/?p=10339#comment-118519 In reply to Jeff – VoipDIY.

Thanks for the complete answer Jeff 🙂

]]>
By: Jeff - VoipDIY https://bloggeek.me/webrtc-signaling-transport/#comment-118518 Mon, 16 May 2016 16:37:15 +0000 https://bloggeek.me/?p=10339#comment-118518 As a ham radio operator since the age of 13, answering the question to the above question to the photo caption:
“Can you guess the signaling and transport here?”
didn’t take me more than a millisecond to recall.

As the photo is of a telegraph hand key, I would presume the “protocol” is International Morse Code – more commonly known as just – “Morse Code”.

In the day of Thomas Edison, the “transport” would typically be be just a pair of wires hung between telephone/telegraph poles connected to a telegraph key and “sounder” at each end of the wires.

Later, the main transport for transmitting and receiving Morse Code became the HF radio transmitter/antenna pair at the near end, and an antenna/radio receiver pair at the far end.

In my early days of ham radio, Morse Code was the primary methodology I used to communicate with other ham radio operators around the world. It was also used by the military, merchant marines, and not to mention Western Union. 20 wpm was my top transmission rate speed.

Morse Code over radio was also known as “CW” (continuous wave modulation). The Morse Code hand key essentially modulates a “continuous wave” radio signal with its on/off action producing a coded message encoded with Morse Code.

One might consider the human/telegraph-key pair as the Encoder and the human/radio receiver pair as the decoder, while the radio transmitter/receiver antenna pairs as the transport system.

Sorry if I was too long winded answering that question to the photo caption… 🙂

]]>
By: Tsahi Levent-Levi https://bloggeek.me/webrtc-signaling-transport/#comment-118517 Mon, 16 May 2016 14:07:24 +0000 https://bloggeek.me/?p=10339#comment-118517 In reply to Matthew.

Matthew,

Thought about you and Matrix (https://matrix.org) multiple times when writing this one. You know I love what you’re doing and also suggesting it to people when it makes sense.

What you might not like is where I have Matrix compartmentalized – in my mind, this is still in the domain of proprietary even if that isn’t the intent with it 🙂

]]>
By: Matthew https://bloggeek.me/webrtc-signaling-transport/#comment-118516 Mon, 16 May 2016 13:22:40 +0000 https://bloggeek.me/?p=10339#comment-118516 You forgot Matrix.org as a less hateful alternative to SIP, XMPP and proprietary SaaS for signalling 🙂

]]>