W3C home > Mailing lists > Public > public-privacy@w3.org > October to December 2010

Re: MAC addresses and privacy...

From: Dave Raggett <dsr@w3.org>
Date: Tue, 05 Oct 2010 12:00:32 +0100
To: David Singer <singer@apple.com>
Cc: Harry Halpin <hhalpin@ibiblio.org>, public-privacy@w3.org
Message-ID: <1286276432.1210.72.camel@ivy>
On Mon, 2010-10-04 at 16:42 -0700, David Singer wrote:
> I think we have the inverse here.
> 
> Skyhook and the like tell *me* where *I* am, based on the Mac addresses
>  that can be seen in the vicinity (as I say, I always assumed it was of
>  infrastructure base stations).
> 
> I think this is miles away from it telling *me* where *you* are, based
>  on sightings of your phone/laptop/netbook etc.'s MAC address near
>  other points.  At the moment, the database 'only' says (when you look
>  up my MAC address) "you're seeing an address I'd normally expect to
>  find in San Francisco", so if you see this in (say) Algiers, you might
>  notice.  It's not far to stretch this to "you're seeing a MAC address
>  usually associated with <something that more closely identifies me>". 

Isn't Google returning the last reported location as determined the last
time a script on your browser asked for its location via the W3C
geolocation API. So if you are an avid user of geolocation-based
services, I can track you (assuming your browser is relying on google's
location service and not a locally connected GPS device).

This seems like a privacy flaw in google's geolocation service, but not
in the W3C geolocation API.

p.s. the Firefox implementation of the W3C geolocation API fails to
return a location if you are using a wired connection and not a wifi
connection. This is true even if the wifi interface is active, i.e. you
can view the nearby access points but aren't connected to any of them.
Firefox 3.6.8 also fails to make use of locally connected GPS devices.

-- 
 Dave Raggett <dsr@w3.org> http://www.w3.org/People/Raggett
Received on Tuesday, 5 October 2010 11:00:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 5 October 2010 11:00:55 GMT