- From: Andrei Popescu <andreip@google.com>
- Date: Tue, 1 Nov 2011 19:57:52 -0700
- To: Charles McCathieNevile <chaals@opera.com>
- Cc: Dominique Hazael-Massieux <dom@w3.org>, Wojciech Maslowski <wmaslowski@opera.com>, Anne van Kesteren <annevk@opera.com>, Steve Block <steveblock@google.com>, Doug Turner <doug.turner@gmail.com>, Lars Erik Bolstad <lbolstad@opera.com>, "public-geolocation@w3.org" <public-geolocation@w3.org>
Hi, After discussing this topic during the Geolocation WG f2f meeting on September 7 2011, we decided to not pursue the proximity alarm interface for now. We may revisit this decision later if there is more demand for such an API. The minutes of the discussion can be found at http://www.w3.org/2011/09/07-geolocation-minutes.html#item03 Thanks, Andrei On Fri, Jul 1, 2011 at 7:41 AM, Charles McCathieNevile <chaals@opera.com> wrote: > On Thu, 30 Jun 2011 13:01:16 +0200, Andrei Popescu <andreip@google.com> > wrote: > >> On Thu, Jun 30, 2011 at 10:19 AM, Dominique Hazael-Massieux <dom@w3.org> >> wrote: >>> >>> Le jeudi 30 juin 2011 à 09:23 +0100, Andrei Popescu a écrit : >>>> >>>> Yes, you can probably work out roughly where the user is from IP, then >>>> set up a set of proximity alerts that will track the user with almost >>>> the same accuracy as watchPosition(). It's slightly more involved but >>>> perfectly possible, so I agree the privacy gains aren't that great. >>> >>> They might not be great in the worse case scenario (a site asking for >>> proximity alerts for many locations), but they're pretty good in better >>> scenarios (e.g. a single location). >>> >>> Also, a user agent could better give guidance to the user when a site >>> requests alert for too many locations, or refuse giving alerts when too >>> many are requested, or when they are to close to one another, etc. >> >> But what exactly is "too many" or "too close"? > > A good question for the market to decide. As Dom says, there isn't a > perfect answer, but a continuum where at the good end you get lots of cool > things, and at the bad end you just have what you already do. Given the > interface, we'd all be in a position to try and move towards the good end, > instead of the usual pressures that make so much of the Web a race to the > bottom... > >> Pick too small a number and you'll break many use-cases. Pick a bigger >> number and the privacy benefit quickly evaporates. > > Indeed. And one gorilla in the market might accustom developers to a > particular number, which would probably be a shame. But it turns out there > are more axes than the raw number - density of points is also important, > and more so where users can choose the accuracy of the location they > offer. This complicates the task of the UX people, and the price we'll pay > if we get it really wrong is not small. On the other hand, there are > plenty of potential upsides... > > cheers > > -- > Charles McCathieNevile Opera Software, Standards Group > je parle français -- hablo español -- jeg lærer norsk > http://my.opera.com/chaals Try Opera: http://www.opera.com >
Received on Wednesday, 2 November 2011 02:58:31 UTC