RE: [sensor] Feedback to Geolocation WG on Device Orientation?

Agreed, even the sensor may be accessed over a network connection of any type... the controls we are talking about affect only the behavior of the first hop (the webapp API implementation) in delivering the events. How that affects what happens under the hood can leverage/affect any number of functions in a particular implementation.

Thanks,
Bryan Sullivan

-----Original Message-----
From: Marcos Caceres [mailto:w3c@marcosc.com] 
Sent: Thursday, March 15, 2012 3:33 PM
To: SULLIVAN, BRYAN L
Cc: Tran, Dzung D; Frederick.Hirsch@nokia.com; public-device-apis@w3.org
Subject: Re: [sensor] Feedback to Geolocation WG on Device Orientation?



On Thursday, 15 March 2012 at 16:39, SULLIVAN, BRYAN L wrote:

> Thanks Dzung, that captures my general understanding. Having written signaling software dealing with low-level signal sensor integration, I can relate to the value of both handing that function off to a OS-level API while give the app the ability to set the criteria that makes sense for it.

There are multiple layers in between the sensor and the application. We should make no assumption that the there is a direct link between the web application, it's timing model, the OS, it's timing model, and the timing model of the sensor (which could be controlled by both hardware and software). Each will have it's own timing and could be controlled independently by separate parts of the software and hardware (i.e., completely separate from the Web app, which could (and will likely) just be talking to a proxy of the sensor and not the sensor itself). 

-- 
Marcos Caceres
http://datadriven.com.au

Received on Friday, 16 March 2012 02:54:11 UTC