W3C home > Mailing lists > Public > public-geolocation@w3.org > March 2010

RE: enableHighAccuracy as a privacy feature

From: Thomson, Martin <Martin.Thomson@andrew.com>
Date: Thu, 25 Mar 2010 02:21:34 +0800
To: Matt Womer <mdw@w3.org>, Doug Turner <dougt@dougt.org>
CC: Dominique Hazael-Massieux <dom@w3.org>, public-geolocation <public-geolocation@w3.org>
Message-ID: <8B0A9FCBB9832F43971E38010638454F03E24ED196@SISPE7MB1.commscope.com>
Matt Womer wrote:
> This assumes the only "bad guy" is the one developing the web site.  The bad guy could be a third-party snooping on the wire.  As it stands, the spec allows for the scenario where a web site that doesn't require high accuracy to get high accuracy regardless.  Why send more information than the developer indicated they required?

This is the best approach that I've seen.

Assuming a sliding scale, site picks the value that they would find acceptable, or might like.  If the user is concerned, they provide only that minimum.  It's a negotiation, so either party might have to compromise.

Site wants 10m, user says 50m; site is disappointed
Site says 1000m, user is generous and says 200m; site gets more than they need

The problem with enableHighAccuracy is that it essentially has no semantics.

Received on Wednesday, 24 March 2010 18:21:06 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 22 March 2012 18:13:46 GMT