%%%
xml2rfc (http://xml2rfc.ietf.org/)
Title = "RealTime Internet Peering for Telephony (RIPT) Compatibility with webRTC"
abbrev = "RIPT webRTC" category = "std" docName = "draft-rosenberg-dispatch-ript-webrtc-00" ipr= "trust200902" area = "Internet" keyword = ["RTP"]
[pi] symrefs = "yes" sortrefs = "yes"
[[author]] initials = "J." surname = "Rosenberg" fullname = "Jonathan Rosenberg" organization = "Five9" [author.address] email = "jdrosen@jdrosen.net"
%%%
.# Abstract
The Real-Time Internet Peering for Telephony (RIPT) Protocol defines a technique for establishing, terminating and otherwise managing calls between entities in differing administrative domains. The RIPT Inbound extension brings this to end clients, such as a browser. However, it defines a different technique for media that cannot directly use the webRTC APIs, and require a change to them. This specification provides an extension to RIPT for webRTC compatibility, enabling media to flow from browser to server as is done with RIPT, or from browser to browser as is done with webRTC. It also discusses techniques for sending e2e encrypted media.
{mainmatter}
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [@!RFC2119].
Basic idea: The TG indicates that this compatibility mode is supported on the TG. The handler is also indicates that this mode is supported. The handler includes its own ICE candidates. This means we provide the ICE candidates at "registration" time and not before the call. This is necessary to facilitate the many call move and other operations in RIPT-inbound. This also means the browser needs to keep them fresh all of the time, rather than just before the call [[is this posible with current API??]].
Since the media is sent by DTLS-SRTP and not embedded as media chunks in a client-to-server HTTPS connection, the browser includes its fingerprint in the handler as well.
To initiate this compatibility mode for media, the server indicates as such in the directive. It can only put it in a directive if the handler that is selected, supports the mode. The directive includes the ICE candidates from the peer. This will trigger the client to perform ICE and send media (which will be DTLS-SRTP).
RIPT itself doesnt convey the ICE candidates in the server to server link, since its only through handler whih is static for a device and not per-call. So we'd either need to move them, develop a separate way to convey them, or assume SIP or some other technique is used for server to server calls.
Suggest we also require a well-known port for media, and we'll need an RTP headr extension to convey the callID since its included inband in RIPT.
TODO
TODO
Thanks to Justin Uberti and Cullen Jennings for the discussion on this concept.
{backmatter}