RE: [geolocation] accuracy attribute

That is correct.

The Android values could (and probably should) be scaled up to meet the
confidence level.
On Sep 18, 2015 8:51 AM, "Mandyam, Giridhar" <mandyam@qti.qualcomm.com>
wrote:

> I did not participate in the early days of development of the geoloc. API
> - I did however participate at the time in the development of the old BONDI
> specification and looked into what platforms were offering at that time
> (2007 - 2008).  A popular API at that time was the JavaME JSR-179
> specification, which defined accuracy as a radius (meters) with 1-sigma
> confidence.
>
> It looks like it was a prescient decision to make the confidence level a
> "should" requirement.  For instance, Android LocationManager seems to only
> require 68% confidence level:  see
> http://developer.android.com/reference/android/location/Location.html#getAccuracy%28%29
> .
>
> My conclusion is that your interpretation is correct, with the caveat that
> the confidence level of 95% may not be met by a given implementation.
>
> Corrections/modifications/comments welcome, of course.
>
> -Giri
>
> -----Original Message-----
> From: Tobie Langel [mailto:tobie@sensors.codespeaks.com]
> Sent: Friday, September 18, 2015 8:31 AM
> To: public-geolocation@w3.org
> Subject: [geolocation] accuracy attribute
>
> Hi,
>
> Looking at the accuracy attribute[1] in the Geolocation spec which
> reads:
>
> "The accuracy attribute denotes the accuracy level of the latitude and
> longitude coordinates and is specified in meters. [...] The accuracy and
> altitudeAccuracy values returned by an implementation should correspond to
> a 95% confidence level."
>
> I assume this means that the user agent is actually positioned within a
> circle with radius equal to the accuracy in meters. Is this a correct
> interpretation?
>
> Thanks,
>
> --tobie
>
> ---
> [1]: http://dev.w3.org/geo/api/spec-source-v2.html#accuracy
>
>
>