- From: Martin Thomson <martin.thomson@gmail.com>
- Date: Fri, 9 May 2014 09:33:03 -0700
- To: Justin Uberti <juberti@google.com>
- Cc: Alex Gouaillard <alex.gouaillard@temasys.com.sg>, Eric Rescorla <ekr@rtfm.com>, IƱaki Baz Castillo <ibc@aliax.net>, "public-webrtc@w3.org" <public-webrtc@w3.org>
On 8 May 2014 23:36, Justin Uberti <juberti@google.com> wrote: > I don't think an event is supposed to be fired for "new". That is the > initial state. One might also argue that close(), having been invoked by the application, does not need an event either. But I think that for the purposes of consistency, the model should be that all state transitions generate an event. In the case of new, that transition occurs in the constructor, prior to returning a reference to the object, so there is no way for that change to be observed by an application.
Received on Friday, 9 May 2014 16:33:30 UTC