[Jingle] Proposed message to send to the IETF rtcweb and W3C WebRTC working groups.

Philipp Hancke fippo at goodadvice.pages.de
Mon Jul 22 11:37:57 UTC 2013


On Mon, 22 Jul 2013, Dele Olajide wrote:
> I applaud the effort as well, but I hope it is not too late. I suppose Google abandoning Jingle was a wake-up call (no pun intended).

Well, I recall a message to the rtcweb wg from someone who claimed to work 
a large jingle based system.

> When I first proposed what has now been labelled SOX (SDP over XMPP) over a year ago, as an extension to Jingle and even wrote a draft XEP, I was shot down in flames. The point I made then is still valid right now. Jingle did not make itself attractive to all the web developers looking for a protocol back then to implement WebRTC signalling and most of them have since adopted SIP over Websockets with JsSIP or did some custom SDP invite/offer/answer/terminate signalling with JSON and NodeJs.

Right. Note that currently there's alot of discussion about SDP (and 
babies and bathwater) over at the w3c. Evaluation of things like sox or 
doing sdp-content inside jingle is still on the table if you want to 
discuss that.

> Jingle is fast becoming irrelevant on a few WebRTC projects I have been working on since. We now integrate WebRTC directly into the chat and group-chat sessions. We have no need for another out-of context signalling session. It also solves the issue of forking calls to multiple XMPP user sessions. In fact, we are moving away from legacy telephone calls to natural human audio and visual communication. As in "Sisko to O'Brien" from the Star Trek DS9 TV show.

Right. This tight integration is one of the reasons why i'd like to 
evaluate using <message/> instead of <iq/>. If you already have an 
established chat session (a <thread/>) doing call-style ringing seems 
rather old-fashioned.


More information about the Jingle mailing list