Workers v2 (Was: Refactoring SharedWorkers out of Web Workers W3C spec)
On Tue, 10 Dec 2013, Jonas Sicking wrote: However I'd really like to see us start a level 2 of the spec. The synchronous messaging channels is something else I'd like to see done there. There's seven features I'm aware of that people have asked for that aren't in Workers currently, or are specced in a way people don't want: - Synchronous message channels This has been proposed several times on this list, but so far I've only seen interest from Mozilla. This is currently not on my radar, since there's no outstanding e-mail on this topic that was sent to the WHATWG list, and no bug is assigned to me on this topic as far as I can tell. The last proposal that I am aware of is: http://lists.w3.org/Archives/Public/public-webapps/2012JulSep/0686.html http://lists.w3.org/Archives/Public/public-webapps/2013OctDec/0142.html - Inline workers (inline as in specified by script in HTML) Waiting for implementation interest: https://www.w3.org/Bugs/Public/show_bug.cgi?id=22700 - Canvas in Workers There's been various proposals, including one in the spec that hasn't met with implementor approval; I'm waiting for something to get traction amongst the competing proposals. - Being clearer about what features are visible in workers Blocked on: https://www.w3.org/Bugs/Public/show_bug.cgi?id=22646 - Cross-origin workers Waiting for implementations to implement the other features first: http://lists.w3.org/Archives/Public/public-webapps/2012OctDec/0617.html - Real-time support Waiting for implementation interest: http://lists.w3.org/Archives/Public/public-whatwg-archive/2012Dec/0272.html - A worker to intercept the fetch logic Alex is working on this; I haven't been following it: https://github.com/slightlyoff/ServiceWorker/blob/master/README.md If any of them have multiple vendors on board, let me know, and I'll spec them. I try to keep the spec not too far ahead of the browsers. Incidentally, I found this interesting: https://gist.github.com/tobeytailor/2693804 ...especially in the context of: http://lists.w3.org/Archives/Public/public-webapps/2012JanMar/0678.html http://lists.w3.org/Archives/Public/public-webapps/2012JanMar/0695.html If this kind of thing is indeed feasible (I haven't studied it closely), it might make the need for sync APIs more moot. -- Ian Hickson U+1047E)\._.,--,'``.fL http://ln.hixie.ch/ U+263A/, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Re: Workers v2 (Was: Refactoring SharedWorkers out of Web Workers W3C spec)
- Canvas in Workers There's been various proposals, including one in the spec that hasn't met with implementor approval; I'm waiting for something to get traction amongst the competing proposals. - Being clearer about what features are visible in workers Blocked on: https://www.w3.org/Bugs/Public/show_bug.cgi?id=22646 I have proposed several times about allowing to create PeerConnection and DataChannel objects from inside a Worker, don't know if that request falls into the what features are visible topic or if it's a special case like the canvas... https://groups.google.com/forum/m/#!topic/discuss-webrtc/-bOW_hhs28E
Re: Workers v2
On Wed, 11 Dec 2013, pira...@gmail.com wrote: - Canvas in Workers There's been various proposals, including one in the spec that hasn't met with implementor approval; I'm waiting for something to get traction amongst the competing proposals. - Being clearer about what features are visible in workers Blocked on: https://www.w3.org/Bugs/Public/show_bug.cgi?id=22646 I have proposed several times about allowing to create PeerConnection and DataChannel objects from inside a Worker, don't know if that request falls into the what features are visible topic or if it's a special case like the canvas... https://groups.google.com/forum/m/#!topic/discuss-webrtc/-bOW_hhs28E It's in the what features are visible topic, unless there's anything specific about the API that needs changing in workers. As far as the WebRTC stuff goes, though, I'll let the WebRTC group decide what should happen. The issue of being clearer about what features are visible in workers is mostly about getting some IDL-level keyword that we can use to make it easier to specify (right now it can be done but has to be done in prose, and I haven't been consistent about it in my specs). -- Ian Hickson U+1047E)\._.,--,'``.fL http://ln.hixie.ch/ U+263A/, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Re: Workers v2
Thanks for the clarification :-) It shouldn't be dificult, specially for DataChannels: WebWorkers has already support for WebSockets and their API is the same (and the security context fairly similar...). I agree on let this discussion to the WebRTC group. Send from my Samsung Galaxy Note II El 11/12/2013 22:18, Ian Hickson i...@hixie.ch escribió: On Wed, 11 Dec 2013, pira...@gmail.com wrote: - Canvas in Workers There's been various proposals, including one in the spec that hasn't met with implementor approval; I'm waiting for something to get traction amongst the competing proposals. - Being clearer about what features are visible in workers Blocked on: https://www.w3.org/Bugs/Public/show_bug.cgi?id=22646 I have proposed several times about allowing to create PeerConnection and DataChannel objects from inside a Worker, don't know if that request falls into the what features are visible topic or if it's a special case like the canvas... https://groups.google.com/forum/m/#!topic/discuss-webrtc/-bOW_hhs28E It's in the what features are visible topic, unless there's anything specific about the API that needs changing in workers. As far as the WebRTC stuff goes, though, I'll let the WebRTC group decide what should happen. The issue of being clearer about what features are visible in workers is mostly about getting some IDL-level keyword that we can use to make it easier to specify (right now it can be done but has to be done in prose, and I haven't been consistent about it in my specs). -- Ian Hickson U+1047E)\._.,--,'``.fL http://ln.hixie.ch/ U+263A/, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'