As said, the implications of such a decision have been discussed and there are
pros and cons to the approach.
"evil" is probably not a correct clarification.
On Mar 30, 2011, at 10:18 AM, Eric Burger wrote:
And the Proxy <-> Browser interaction is 100% out of IETF scope. For that
matter, the IETF should be pointing out how dangerous and evil such a
proposal is, as it means the end of consumer choice and a competitive
marketplace for clients.
On Mar 30, 2011, at 9:14 AM, Hannes Tschofenig wrote:
Dave,
I explain the change with two figures in order not to be misunderstood
(again).
I use SIP as an example; Jonathan gave a nice presentation.
Working Assumption previously:
............................ ..............................
. . . .
. +-------+ . . +-------+ .
. | | . SIP . | | .
. | Proxy |------------- | Proxy | .
. | 1 | . . | 2 | .
. | | . . | | .
. / +-------+ . . +-------+ \ .
. / . . \ .
. / . . \ SIP .
. SIP / . . \ .
. / . . \ .
. / . . \ .
. / . . \ .
. / . . \ .
. +-------+ . . +-------+ .
. | | . . | | .
. | | . . | | .
. | UA 1 | . . | UA 2 | .
. | | . . | | .
. +-------+ . . +-------+ .
. Domain A . . Domain B .
............................ ..............................
Figure 1: The SIP trapezoid
We have lots of standardization efforts that focus on the UA<->Proxy leg in
the RAI area.
Suggested new working assumption:
+-----------+ +-----------+
| Web/ | | Web/ |
| SIP | SIP | SIP |
| |-------------| |
| Server | | Server |
| 1 | | 2 |
+-----------+ +-----------+
/ \
/ \ Proprietary over
/ \ HTTP/Websockets
/ \
/ Proprietary over \
/ HTTP/Websockets \
/ \
+-----------+ +-----------+
|JS/HTML/CSS| |JS/HTML/CSS|
+-----------+ +-----------+
+-----------+ +-----------+
| | | |
| | | |
| Browser | ------------------------- | Browser |
| | Media | |
| | | |
+-----------+ +-----------+
Figure 2: Browser RTC Trapezoid
The server-to-server interaction I was referring to in my previous mail is
the interaction between server 1 to server 2. With cross-domain usage there
still a standardization need. This is what I mean by "the interoperability
need shifts".
We had spoken about the implications of that change already.
Ciao
Hannes
On 3/29/2011 1:31 PM, Hannes Tschofenig wrote:
Correct.
The interoperability need shifts away from the client-to-server side (for
example, to the server-to-server side;
No, that's wrong and I believe it is not what Eric said at all.
THERE IS STILL A CLIENT/SERVER PROTOCOL, HANNES.
ALL THAT CHANGES IS THAT THE CLIENT/SERVER PROTOCOL IS NOW PROPRIETARY.
I apologize for shouting. I'm shouting for the classic reason that I'm
taking your continuing to misunderstand this multiply-repeated and very
basic point as a hearing problem.
Server-server is an entirely different task and different part of the
architecture.
d/
--
Dave Crocker
Brandenburg InternetWorking
bbiw.net
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf