W3C home > Mailing lists > Public > public-webrtc@w3.org > June 2012

Re: [ACTION-43] (sdp related objects and global namespace) - way forward

From: Dominique Hazael-Massieux <dom@w3.org>
Date: Thu, 14 Jun 2012 17:18:22 +0200
Message-ID: <1339687102.3217.447.camel@altostratustier>
To: Adam Bergkvist <adam.bergkvist@ericsson.com>
Cc: Anant Narayanan <anant@mozilla.com>, "public-webrtc@w3.org" <public-webrtc@w3.org>
Le jeudi 14 juin 2012 à 17:08 +0200, Adam Bergkvist a écrit :
> Perhaps we could settle with just one object. It would have a type 
> attribute that could be "candidate" as well as the current 
> SessionDescription types.

Are you suggesting to use "candidate" to replace the IceCandidate
interface? I'm not sure why we need that interface at all. It seems to
only encapsulate a string — why not pass just the string?

>  I believe a requirement on the factory method 
> is to be able to take a string received from the network. Otherwise it 
> wouldn't be a direct replacement for the constructor (if that's what we 
> need).

The only reason why I thought we wanted a constructor (or an interface
more generaaly) for SessionDescription was that we thought we would want
to add methods on top of that object to enable manipulations; the
discussions at the F2F didn't seem to indicate this was that useful (or
at least not useful enough to deserve a JavaScript native object, vs a
ad-hoc library maybe?).

But given your point above, I'm not sure anymore — was there any other
point about having an object for SessionDescription?

Dom
Received on Thursday, 14 June 2012 15:18:53 UTC

This archive was generated by hypermail 2.3.1 : Monday, 23 October 2017 15:19:28 UTC