RE: Issue for discussion on Wed

Agree with both you and hopefully we all agree the focus should be on unique identifiers - whether they be from cookies, digital fingerprints, or some future form of unique IDs.  

-----Original Message-----
From: Alan Chapell [mailto:achapell@chapellassociates.com] 
Sent: Tuesday, July 09, 2013 8:58 AM
To: Walter van Holst
Cc: public-tracking@w3.org
Subject: Re: Issue for discussion on Wed

I agree that being technology specific may create unintended consequences.
However, if an entity that has access to the URL string engages in tracking as defined by the WG, we should ensure that the spec addresses this - regardless of the technology utilized to access the URL string.


Thanks Walter.

Alan

On 7/9/13 11:39 AM, "Walter van Holst" <walter.van.holst@xs4all.nl> wrote:

>On 2013-07-09 17:02, Alan Chapell wrote:
>> Hi Walter - I offered two links to articles that might be helpful.
>> 
>> 
>>http://download.cnet.com/8301-2007_4-20123464-12/amazons-silk-browser-
>>now
>>-e
>> ff-approved-really/
>> http://www.theregister.co.uk/2003/11/07/help_my_belkin_router/
>> 
>> I'm making a point that most UA's have access to URL history and 
>> other information that could easily be used for tracking as defined 
>> by the WG.
>> It would seem inconsistent to have DNT block other forms of tracking 
>> if we're not also going to have DNT block UA tracking.
>
>Thank you for the links and reminding me of them.
>
>Neither the multi-tiered Silk browser-architecture nor the Belkin 
>router are user agents in the strict sense of the word. The Belkin 
>router is the easiest case: it does not initiate http requests. While I 
>would agree with anyone who would consider Belkin's behaviour highly 
>worrisome from multiple perspectives, it is a Belkin problem, not a UA problem.
>
>Silk is more complicated (and no less worrisome), but again, anything 
>that happens in Amazon's EC2 caching part of that multi-tiered browser 
>appears more akin to a proxy than an UA-extension. Admittedly, this is 
>a pretty grey area.
>
>As a result, I am much more favourable to a functional definition of 
>tracking such as the one proposed by Rob van Eijk today than of trying 
>to cover both ends of a HTTP interaction.
>
>Even if we did, this is an issue we can't easily solve through this 
>standard simply because both in the case of Silk and of Belkin one 
>could argue that both Amazon and Belkin have acquired 
>out-of-band-permission through their terms of usage/general terms and 
>conditions. As a European privacy advocate I could easily argue that 
>these are insufficient mechanisms for acquiring meaningful consent, 
>reality tells me that this doesn't mesh well with the sanctity of the 
>freedom of contract in some other parts of the world.
>
>One can only hope that the EU DPAs will have a chat with both Amazon 
>and Belkin in the near future.
>
>Regards,
>
>Walter
>

Received on Tuesday, 9 July 2013 16:23:35 UTC