RE: Deciding Exceptions (ISSUE-23, ISSUE-24, ISSUE-25, ISSUE-31, ISSUE-34, ISSUE-49)

Rigo,

EU law requires each company retain data only for as long as it's needed for the intended purpose for which it was acquired.  This in essence is the minimization principle - and is tied to each company's unique and individual business models.  Attempting to assess all current business practices and predict all future appropriate business models is nearly impossible and therefore attempting to declare an arbitrary time bound (1 min, 24 hours, 3 months, etc.) is impractical and runs the real-risk of harming legitimate business practices and barring future innovation that attempts to honor DNT.

- Shane

-----Original Message-----
From: Rigo Wenning [mailto:rigo@w3.org] 
Sent: Wednesday, February 08, 2012 12:22 PM
To: Shane Wiley
Cc: public-tracking@w3.org; Roy T. Fielding; Jonathan Mayer
Subject: Re: Deciding Exceptions (ISSUE-23, ISSUE-24, ISSUE-25, ISSUE-31, ISSUE-34, ISSUE-49)

On Wednesday 08 February 2012 09:29:59 Shane Wiley wrote:
>  I would expect each company to minimize their data retention for this
> business purpose to come in-line with their demonstrated need for the
> data.  AKA - minimization standard which is unique and specific to each
> company.

I have trouble understanding where the standard is when each company has their 
own. 

Given the absence of a standard the current incentives have let us to a 
situation where those who collect the most data will earn the most money. And 
given this fact, having no standard and no limits, the only thing left is a 
use limitation on collected tracking data against all incentives.

Everything can be expressed by a mere use limitation and I know that some 
defend that model in the US. I do not have the feeling that relying solely on 
use limitations is a possible consensus. But I think even the industry can 
gain from some smart data minimization strategies. Data still has a cost, 
especially if it has to be maintained at a certain quality and is not just a 
mere data-graveyard where you don't know if something is actually correct or 
not. 

The suggestion is to get some more intelligence in frequency capping with 
DNT=1 and not just lay back and do use limitation while tracking as usual. 

I agree that this is a hard nut to crack

Rigo

Received on Wednesday, 8 February 2012 20:34:29 UTC