Re: [rtcweb] Filling in details on "trickle ICE"

[Ted Hardie <[e-mail]>]

On Wed, Oct 17, 2012 at 1:40 PM, Matthew Kaufman
<[e-mail]> wrote:

> 2)      Are we seriously considering defining the allowed SDP for
> RTCWEB/WEBRTC by referencing an I-D and not an RFC, or are we going to delay
> defining the allowed SDP until trickle ICE is in an RFC, or are we going to
> drop trickle ICE from RTCWEB 1.0?
>

It's fairly normal for an I-D to reference other I-Ds spawned by the
work.  There's an obvious race condition, and a common set of
responses to that.  Typically, if the I-D making the reference is
ready to advance before the one it refers to is published, there's
discussion at that point just prior to IETF Last Call of whether the
reference is stable enough for a down reference.  If it is not,
there's a reference hold at the RFC publication stage (Bill Fenner has
produced some lovely/scary graphs of what can occur with these holds).

Basically, we don't need to decide which path to take now, but it's
clear that paying attention to the drafts in other working groups  is
a good thing for the advancement of our work.

regards,

Ted