Re: [rtcweb] Why are we asking the wrong questions?

[Iñaki Baz Castillo <[e-mail]>]

2011/10/20 Roman Shpount <[e-mail]>:
> On Thu, Oct 20, 2011 at 4:09 PM, Iñaki Baz Castillo <[e-mail]> wrote:
>>
>> Hi Roman, could you please check this short document and specify which
>> "signaling protocol" (of the three appearing in the document) you
>> mean?:
>>
>>  http://dev.sipdoc.net/projects/oversip/wiki/RTCweb_Signaling_Components
>>
> This document is under password protection and I cannot access it.

Hi Roman, it's not under password protection. But the web has no
favicon and the document root has no index permission, so when your
browser requests the "standard" favicon icon it finds a 401 response.
That's a bug of some web browsers. Please just press "cancel" when
user:passwd is requested.


> P.S. I suggest to stop redefine standard terms for the sake of the working
> group. Signaling protocol is defined in
> http://en.wikipedia.org/wiki/Signaling_protocol and is something like SIP --
> definition of data interactions over a wire.

The document I've written is more that just the definition of
"signaling protocol".


>  A set of JavaScript methods is
> not a signaling protocol -- it is an API. State machine that describes these
> JavaScript methods is still part of the API specification. In my personal
> opinion,

I can agree with this. The important point here is that all the WG
agrees with the term to use in each case.


> there are should be no "signaling protocols" in RTC.

In your previous comment you said:

  "Standard signaling protocol is a convenience feature for RTC."

Could you please clarify if you are in favour of mandating the
in-the-wire signaling protocol or not?


Thanks a lot.


-- 
Iñaki Baz Castillo
<[e-mail]>