Re: Review Comments: Light Events

On 1 Aug 2012, at 17:25, Doug Turner <dougt@mozilla.com> wrote:

> This is great.  Please let me know more about the 3 failures when you have time.

They are WebIDL conformance check errors (I.e., is type conversion being performed as per WedIDL?). According to my reading of WebIDL, the implantation is not conforming - hence 3 errors.

Ms2ger or someone should check the tests to make sure they are ok. 

When I spoke with Ms2ger, Ms2ger suggested that I should not test WebIDL conformance as a better general WebIDL is in the works (though I don't have any details about that project). 

Hopefully the tests are ok and I can remove them once we have a better WebIDL conformance tool. They will do for now.

> 
> Thanks!
> 
> ----- Original Message -----
> From: "Dzung D Tran" <dzung.d.tran@intel.com>
> To: "Doug Turner" <doug.turner@gmail.com>, "nv balaji" <nv.balaji@samsung.com>
> Cc: "Robin Berjon" <robin@berjon.com>, "public-device-apis@w3.org public-device-apis@w3.org" <public-device-apis@w3.org>
> Sent: Wednesday, August 1, 2012 6:43:53 AM
> Subject: RE: Review Comments: Light Events
> 
> 
> 
> 
> Hello Doug, 
> 
> 
> 
> I finally ran the tests on my Nexus with FF Beta and for: 
> 
> 
> 
> Light: 42 pass and 3 Fail. 3 Fail are with value of undefined resolves to NaN, expected type error. I haven’t dig into these errors yet. 
> 
> 
> 
> Proximity: 81 pass and 2 Not Run. 2 Not Run are with User Proximity. 
> 
> 
> 
> Thanks 
> Tran 
> 
> 
> 
> From: Doug Turner [mailto:doug.turner@gmail.com] 
> Sent: Tuesday, July 31, 2012 2:32 PM 
> To: nv.balaji@samsung.com 
> Cc: Robin Berjon; public-device-apis@w3.org public-device-apis@w3.org 
> Subject: Re: Review Comments: Light Events 
> 
> 
> 
> Thanks for the feedback! 
> 
> 
> On Sun, Jul 29, 2012 at 10:32 PM, BALAJI NERELLA VENKATARAMANA < nv.balaji@samsung.com > wrote: 
> 
> 
> Please find my comments: 
> 
> 
> 
> 1. I would suggest to change the title to "Ambient Light Events" 
> 
> 
> 
> Happy to change the title of the spec. The event names will remain the same. 
> 
> 
> 
> 
> 2. Section 5.2: 
> 
> - Computing "lighting condition" (such as warm day, pitch dark etc...) from light value is non-trivial. Ligting condition corresponds to logarithm of Luminance. Therefore, it is better "lighting condition" parameter is added to the Light event. 
> 
> - If it is not acceptable (as this information is redundant), as an alternate, a note could be appended at the end explaining how the lighting condition could be arrived from the light value 
> 
> 
> 
> What language would you use as a note to describe how the lighting condition can be derived? 
> 
> 
> 
> 
> 
> 
> 3. Section 5.2.1 
> 
> - the unit of the value should be specified 
> 
> 
> 
> Fixed. 
> 
> 
> 
> 
> 
> 
> 4. Section 5.2.2 
> 
> - For all practical purposes, I beleive, the developer would be interested in events only when the lighting condition changes. Therefore, light event should be fired only when "lighting condition" changes. I have not tested the latest Firefox implementation on mobile. Could someone please let me know how often does it fire the event? 
> 
> 
> We fire when there is a change. The UA determines when there is a significant change.
>

Received on Thursday, 2 August 2012 01:50:47 UTC