Re: [rtcweb] interworking with non-WEBRTC endpoints [was RE: Use Case draft]

["Fabio Pietrosanti (naif)" <[e-mail]>]

On 5/2/12 7:50 PM, Dan Wing wrote:
http://www.ietf.org/proceedings/83/slides/slides-83-rtcweb-3.pdf
> However, when I presented slide 7, there were objections at the 
> microphone that this model 'is broken'.  I would like to understand 
> the objections so we can reach consensus on how interworking from
> WEBRTC to non-WEBRTC is expected to occur.

IMHO it would be much easier, as described in tons of previous email, to
use different Key Management for different requirements:

- SDES + SRTP for end-to-site (peer to gateway)
- DTLS-SRTP for end-to-end (peer to peer)

It would be a simpler approach for:

- Security (the user know the security level)
- Interoperability (Use standard protocols for the need to interoperate)

That way the "simpler, legacy, standard" technology would be used for
all voip "server applications" while the new burning (yet not used by
anyone) DTLS-SRTP will be used for peer-to-peer calls.

Please DO NOT reinvent the wheel for what's already existing and deployed.

Doing so, it's like going against the natural human behavior, it would
just represent a failure.

Fabio