Re: June Change Proposal, Geolocation

Hi David,

I've created ISSUE-202 on the Compliance June product. This may also be relevant to our existing decision on closed ISSUE-39.

I've set up a wiki page for this proposal: http://www.w3.org/wiki/Privacy/TPWG/Change_Proposal_Geolocation

I think it would be great to clarify the alternative you suggest immediately below; I think the original intent (and our group's earlier consensus) was that this was a specific additional requirement rather than a replacement. Can you provide specific text for that option? We can add it to a new section on the same wiki page.

Thanks,
Nick

On Jun 21, 2013, at 11:43 AM, David Singer <singer@apple.com> wrote:

> It's been pointed out to me that an alternative is to make it clear that the Geolocation section is a specific addition to the general requirements on non-tracking, rather than a replacement or relaxation.  That would be fine.
> 
> On Jun 20, 2013, at 16:48 , David Singer <singer@apple.com> wrote:
> 
>> This web site might provide some idea why I fear that 'allowing' postal-code (zip-code) granularity might not be OK
>> 
>> http://localistica.com/usa/zipcodes/least-populated-zipcodes/
>> 
>> 
>> On Jun 20, 2013, at 15:29 , David Singer <singer@apple.com> wrote:
>> 
>>> Problem
>>> 
>>> Section 5.3: 
>>> "If a third party is part of a network interaction with a DNT: 1 signal, then geolocation data must not be used in that interaction at any level more granular than postal code, unless specific consent has been granted for the use of more granular location data."
>>> 
>>> 
>>> 1) This suggests that Geolocation data can be treated differently. Though it deserves careful handling, I am not sure it's qualitatively different.
>>> 2) This is actually weaker than the definition of 'tracking' for the cases of postal codes with few inhabitants.
>>> 
>>> 
>>> Proposal
>>> 
>>> 1) Delete 5.3
>>> 2) Insert after the definition of 'tracking':
>>> Note:  Geolocation data requires careful handling, as in some cases it can be combined with other readily available data to identify a specific user, user-agent, or device.
>>> 3) In the accompanying non-normative text, expand on the explanations of why geolocation data needs treating with kid gloves.
>>> 
>>> David Singer
>>> Multimedia and Software Standards, Apple Inc.
>>> 
>>> 
>> 
>> David Singer
>> Multimedia and Software Standards, Apple Inc.
>> 
>> 
> 
> David Singer
> Multimedia and Software Standards, Apple Inc.
> 
>

Received on Saturday, 22 June 2013 00:43:40 UTC