Re: Standby API Specification Proposal

Hi Marcos,

I’ve placed a pull-request on github, to be more clear with my suggestions:
https://github.com/w3c/screen-wake/pull/3

Thanks,
Ilya

On 29 мая 2014 г., at 21:21, Marcos <marcos@marcosc.com> wrote:

> 
> 
> 
> On May 29, 2014 at 1:08:39 PM, Ilya Bogdanovich (bogdanovichiy@yandex-team.ru) wrote:
>> Hi Marcos,
>> 
>>> 
>>> This could also happen with the GC-based model, as the object needs to be kept alive and  
>> explicitly deleted to be useful.
>>> 
>>> this.wakeLock = screen.getWakeLock();
>>> 
>>> function youCanSleepAgain(){
>>> doSomething(); // exception thrown
>>> delete this.wakeLock;
>>> }
>> 
>> That’s true, but choosing between two possible errors - "app should not dim but it does",  
>> and “app should dim but it doesn’t” - the second one seems more harmful to me.
> 
> Yes, I agree. 
> 
>>> Anyway, the point is that there are lots of ways to manage this - and the underlying OS  
>> may give the user means to override allowing apps stopping the screen from going to sleep  
>> or dimming.
>> 
>> I didn’t mean modal dialog, but of course browser or OS should have a chance to notify user  
>> of such web app behavior and allow him to reject this. And it’s better to notify him before  
>> the action starts, than interrupting process when OS would try to dim the screen for the  
>> first time.
> 
> Ok, cool. Just wanted to make sure. It seems we are pretty much aligned in our thinking then.