W3C home > Mailing lists > Public > public-tracking@w3.org > October 2012

Re: Defining Issue-5: What is tracking?

From: John Simpson <john@consumerwatchdog.org>
Date: Fri, 5 Oct 2012 14:53:39 -0700
Message-Id: <5C51C72D-1536-495F-9E5B-76DB4B294900@consumerwatchdog.org>
Cc: <public-tracking@w3.org>
To: Alan Chapell <achapell@chapellassociates.com>
Alan,

I agree that first parties should NOT be able to use data gathered while acting as a first party to target users when they are operating as third parties.  I said that on the telephone during the Amsterdam F2F meeting.  I specifically suggested that I would be opposed to Google or Yahoo! using data gathered on their 1st party sites on their ad networks. (That was probably bad form.  I think I'm supposed to cite use cases without naming companies.) Perhaps you were not in the room?

I've always believed that DNT should put more obligations on 1st parties than is the case in the current draft standard and have always spoken up to remind those who have said it places none that it does in fact require they don't share data with a 3rd party.  The consensus around few first-party requirements had already emerged when I joined the working group about a year ago and unfortunately (in my view) the FTC in its privacy report takes the same view.

Finally I'm puzzled by your preference for Roy's definition of tracking.  He said:

"Tracking is any non-consensual collection, correlation, or transfer of
data about the Internet activities of a particular user, user agent, or
device beyond the (first party) context in which that activity occurred."

Cheers,
....Roy"

As I read this, his definition does not include first-party data collecting as tracking.  My definition calls it tracking, but then I would spell out in the spec what first-party tracking would be allowed.  Given your expressed concern about what 1st parties can do that 3rd parties cannot, I'd have thought you would want 1st party data collection to count as tracking.

I would find this modification of Roy's definition acceptable:

"Tracking is any non-consensual collection, correlation, or transfer of
data about the Internet activities of a particular user, user agent, or
device."

Finally, I did not intend to make accusations.  I was explaining what conclusion I would reach IF the people who had been calling for definitions failed to offer their own.  I am not there yet. Please note that I said I would sadly draw that conclusion and I would be sad, for I am genuinely trying to move toward a consensus that is good for business and consumers.

So in that spirit, I have offered two definitions.  One original;  one a modification of Roy's who engaged in the discussion.  Again, my preferred position is silence on a specific definition of tracking with the understanding that tracking would be whatever the specification prohibits when a site receives a DNT:1 message. However, if a specific definition helps get us consensus, let's consider some.

I ask again, those of you who are most concerned by the failure to define tracking specifically and who were calling for definitions, could you please offer yours?

I hope everyone in Amsterdam has a good trip back home.

73s,
John



----------
John M. Simpson
Consumer Advocate
Consumer Watchdog
2701 Ocean Park Blvd., Suite 112
Santa Monica, CA,90405
Tel: 310-392-7041
Cell: 310-292-1902
www.ConsumerWatchdog.org
john@consumerwatchdog.org

On Oct 4, 2012, at 4:54 PM, Alan Chapell wrote:

>  John  the definition of tracking means very little if first parties are able to take their own data and use it for ad targeting across the internet. While I recognize that you may not be a big fan of third parties, I can't believe that you think that's a good outcome for privacy or fostering competition. And as someone who  in your own words  made the big concession to "industry" that has now created this mess, it would be helpful for you to communicate your displeasure about this rather than make accusations about others stalling the process. If you do  in fact think this is a good outcome for privacy or competition, I would invite you to communicate this on the record as well.
> 
> That said  I liked Roy's working definition.
> 
> 
> 
> From: John Simpson <john@consumerwatchdog.org>
> Date: Thursday, October 4, 2012 7:10 PM
> To: "public-tracking@w3.org (public-tracking@w3.org)" <public-tracking@w3.org>
> Subject: Defining Issue-5: What is tracking?
> Resent-From: <public-tracking@w3.org>
> Resent-Date: Thu, 04 Oct 2012 23:10:28 +0000
> 
>> Colleagues,
>> 
>> As many of you may have seen, I've been up all night for the past two days following the "discussions" in Amsterdam making a nuisance of myself on IRC asking people to remember to speak into the microphone.  I've repeatedly heard calls for definitions, particularly that of "tracking."
>> 
>> I have long held the view that one is not necessary for the spec; you need only spell out how to send the DNT header and what your obligations are when you get one. However, in the spirit of a genuine desire to move toward consensus I recently offered a definition on the list.  So far only Roy Fielding, another one of us participating from afar, responded. Here is what I proposed:
>> 
>> "Tracking is the collection and correlation of data about the Internet activities of a particular user, computer, or device, over time and across a website or websites."
>> 
>> Here is a link to the email chain: http://www.w3.org/2011/tracking-protection/track/issues/5
>> 
>> May I suggest that those who have repeatedly been calling for a definition now offer one?  Otherwise I would sadly reach the conclusion that such calls are nothing but deliberate attempts to stall the process and undermine the Working Group's efforts.
>> 
>> As ever,
>> John
>> 
>> ----------
>> John M. Simpson
>> Consumer Advocate
>> Consumer Watchdog
>> 2701 Ocean Park Blvd., Suite 112
>> Santa Monica, CA,90405
>> Tel: 310-392-7041
>> Cell: 310-292-1902
>> www.ConsumerWatchdog.org
>> john@consumerwatchdog.org
>> 
Received on Friday, 5 October 2012 21:53:52 UTC

This archive was generated by hypermail 2.3.1 : Friday, 21 June 2013 10:11:36 UTC