Re: [rtcweb] Use Case draft - legacy interop

[Harald Alvestrand <[e-mail]>]

On 05/05/2012 06:32 PM, Iñaki Baz Castillo wrote:
> 2012/5/5 Dan Wing<[e-mail]>:
>>> If I have a SIP phone implementing ICE and DTLS-SRTP (which is
>>> standarized for SIP regardless it has null impementation), will my SIP
>>> phone be able to *directly* talk in the media plane with a WebRTC
>>> browser? or not?
>> That would work.
> Ok, so then, taking into account that SIP defines the usage of
> DTLS-SRTP in RFC 5763 (regardless no one device implements it), why
> does this WG assumes that "interop with non WebRTC endpoints will be
> made via *gateways*"?
>
> Let me please repeat my question: if my SIP phone implements RFC 5763,
> can my SIP phone directly interop at media plane with a WebRTC
> browser? (I know you already replied this, but I want to be very very
> sure) :)
Inaki, your statement that "the WG assumes" is, as far as I can see, a 
straw man made out of red herrings.

The WG's opinion is what's captured in the WG's documents, and (if 
needed) announced by the chairs. Statements by participants are 
statements by participants. If you have issue with the WG's documents, 
quote the text you want changed, don't just make unfounded statements 
like you do above.

I've quoted the overview document before:

    As for all protocol and API specifications, there is no restriction
    that the protocols can only be used to talk to another browser; since
    they are fully specified, any device that implements the protocols
    faithfully should be able to interoperate with the application
    running in the browser.

Since you did not specify the product number of your hypothetical phone, 
and since the IETF has not finished specifying the set of protocols 
needed, there is no guarantee that interworking will happen in practice. 
That is, there's no guarantee that ICE + RFC 5763 is sufficient for 
interoperation.