OpenSoundControl.org: /page-list /toc /spec/1.0 /spec/1.1 /implementations /publications /about
(contents of this file: links to each section)
John Lazzaro. Setting up OSC sessions using Voice-over-IP protocols. 2004. OSC Conference 2004.
Download: files/lazzaro-osc-voip.mov
Abstract: Performances are musical conversations. By extension, performances over a network can be viewed as musical conference calls. Ideally, we should be able to use the network protocols developed for voice-over-IP and video conferencing to do musical performance applications, by adding streams for musical gestural control to the audio and video streams. Thankfully, the network protocol suite developed by the Internet Engineering Task Force (IETF) for interactive media services is designed to make extensions of this sort easy to do. In the IETF stack, the Session Initiation Protocol (SIP) is used to do call management – to do the Internet equivalent of “playing the dial-tone” to the caller, letting the caller “dial the phone”, play the “busy signal” or “ringing tones” to the caller, and “putting the call through” if the callee picks up the phone. Consumer services that use SIP today include Vonage and Apple’s iChat AV. Behind the scenes, SIP (and helper protocols) solve messy problems. SIP lets phones negotiate the type of media to send, so that a videophone can take an audio-only call. SIP handles mobility, so a person with two offices can have phone calls follow him around the building. SIP also handles the modern realities of IP-life: user machines aren’t always on-line, they may not have a stable domain name, they may not have a stable IP address, they might be behind a Network Address Translator, etc. And SIP can do so securely. SIP does not carry media flows itself, and is agnostic to what type of media is being used. Instead, SIP ferries “session descriptions” between clients, written in the Session Description Protocol (SDP). By extending SDP to handle a new media protocol (like OSC), one makes the new protocol SIP-compatible. In my talk, I’ll describe the different ways that SDP could be extended to support OSC, and how the OSC community could work with the IETF to make it happen. As work in adding SDP support for MIDI 1.0 cable emulation (RTP MIDI) is already in progress in the IETF, adding OSC support to SDP would let SIP clients negotiate the use of OSC (if both clients were aware of each others OSC Address Space) or MIDI (if the clients were OSC-unaware) at the start of a call.
Context: This was a featured publication on the legacy (pre-2011) opensoundcontrol.org website, ported to the new site by Matt Wright in early 2021
Submitted to opensoundcontrol.org by Legacy at 03/26/2021 17:03:32
This page of OpenSoundControl website updated Mon May 24 11:20:32 PDT 2021 by matt (license: CC BY).