Re: [rtcweb] Agenda time request for draft-marjou-rtcweb-audio-codecs-for-interop-01

[Hadriel Kaplan <[e-mail]>]

On Mar 14, 2013, at 5:09 PM, Andrew Allen <[e-mail]>
 wrote:

> Koen is right that there are many more obstacles to RTCweb and legacy network interop than just a common codec. First there will need to be RTCweb signaling gateways to interface between the RTCweb signaling and the legacy networks (SIP, H.323 etc) and there will need to be in place mechanisms for peering, federation and address resolution between networks as well as service agreements in place between the players.
> 
> Until those are resolved supporting codecs used in those networks is pointless.

There are already WebRTC-SIP gateways, as I think you know.  I know of two SBC vendors who are in customer labs with it, for example, and there are probably more.  And yes at least one of them, and probably both, can transcode the audio... though it's by no means free.  I don't know if the business case justifies the cost or not for WebRTC use-cases - it appears to in some cases, not others.

It costs far more than has been discussed on this list though, because we think like engineers and consider the cost as a CAPEX issue; whereas in practice the OPEX costs often dwarf the CAPEX ones.  In some data centers, for example, the local Real-Estate costs dwarf CAPEX too, and using less rack-space can pay for a product in no time. (there's an entire company out there that built its business leveraging this fact, replacing POTS switches with much-higher-density ones purely based on rack-space usage savings)

Anyway... moving on to your other point of federation/peering/etc. - I think the use-case that mobile operators have in mind is fairly obvious, and it clearly doesn't have that problem.  Whether you agree with helping that use-case or not is up to you and the rest of this WG, of course.  I'm just noting that peering isn't an obstacle for it.

-hadriel
p.s. Transcoding video, on the other hand, is untenable for pretty much anyone.  I have never heard of a business-case which justifies it, because it simply doesn't scale.  Except for maybe companies that live on other revenue sources, for whom all of this stuff is a rounding error in their IT budget. ;)