[Standards] Re: Jingle bootstrapping
thiago at jivesoftware.com
Mon Mar 5 11:54:20 UTC 2007
Our Proposal at Jive is to still using ICE bases, but without:
* Using TURN with SDP negotiation. In other words, we want to get the relay candidates from our XMPP server. (Because we want to negotiate with the client using XMPP.)
* Trying to get public address without STUN use when it's possible. (Because we prefer XMPP instead of other protocols.)
* Use UDP ECHO instead of STUN connectivity check. (Because UDP ECHO is pretty much easy to implement and very reliable.)
From: standards-bounces at xmpp.org [mailto:standards-bounces at xmpp.org] On Behalf Of Kai Vehmanen
Sent: segunda-feira, 5 de março de 2007 08:25
To: 'XMPP Extension Discussion List'
Subject: RE: [Standards] Re: Jingle bootstrapping
On 03 March 2007, Scott Ludwig wrote:
>Once the relay and stun services are available, the complex
>piece is the ice transport implementation. While other
>transports are possible (and encouraged), we do need a basic
>ICE design that is agreed on. I don't think we need to track
>ICE X standard, where X is always increasing in number. The
>basic ICE algorithm hasn't changed much for awhile. We make a
it's unfortunate that ICE is still not finalized in IETF (in development
since ~2003), but it is getting _very_ close now. I definitely
agree with you that the basic algorithm has remained the same
for a long time now, and all that is left is finetuning the details.
So you should soon have a RFC number to reference to (though I won't
promise to eat my hat in the remote case it's delayed again ;)).
>I don't think it is necessary for the xmpp community to "be
>compatible with company X's ICE/STUN/TURN design" as a design
>requirement. From my experience, if we can build momentum
>around a standard, the companies are incented to build in
>compatibility with that standard.
I understand the concerns over ICE schedule, but I'd hope XMPP/jingle
would still opt to use IETF-ICE as the basis. I also understand that
some extensions/adaptations might be needed (although I think
the SIP/SDP specific parts of ICE are quite well isolated). Still for
implementors, being able to build the media transport subsystem
from fairly standard building blocks (RTP, ICE, codecs), is a big
help. And as getting the ICE RFC out is now so close, it'd be great
to have the XMPP community using it from start.
first.surname at nokia.com (Kai Vehmanen)
More information about the Standards