Re: Future feedback

Le 14/05/2013 15:03, Boris Zbarsky a écrit :
> On 5/14/13 9:04 AM, David Bruant wrote:
>> http://lists.w3.org/Archives/Public/public-script-coord/2013AprJun/0167.html 
>>
>
> I should note that the description of the browser event loop in that 
> message is wrong.  It does not have only two FIFO queues in the specs, 
> or in implementations.  In particular, see task sources.
I will. This part is quite dense and complicated but important 
especially with the upcoming ES7 work on the event loop to make sure it 
is sufficiently low-level to implement (in spec and/or code) the browser 
event loop model.

> I would be strongly opposed to specifying something that requires only 
> two FIFO queues.
I have to admit that the idea of forging in ES7 the "2 nested event 
loops" model seems that it may rule out valuable use cases. I would be 
in favor of some spec freedom here in the message ordering/reordering 
policy. Let's see where this go.

Thanks for your message,

David

Received on Tuesday, 14 May 2013 14:15:04 UTC