Re: June Change Proposal: Personalization

How about if we pull the "No Personalization" subsection out of the "Permitted Uses" section, and clearly specify that the only personalization allowed is with de-identified data?

I agree that this is probably an uncontroversial change.  That said, depending on how one reads the June Draft, it could be interpreted as a substantive shift.  I'd suggest we leave it as an issue / wiki item with the expectation that we'll only have to briefly address it.

Jonathan


On Wednesday, June 26, 2013 at 8:05 PM, Nicholas Doty wrote:

> If, as it sounds, this is actually an editorial question of whether the "No Personalization" subsection of "General Principles for Permitted Uses" is redundant or clarifying, I won't create a separate issue.
>  
> I'm not aware of any arguments that DNT should prohibit personalization, for example, without data collection / from de-identified data, which would be a large change from current text and group agreements that I think you're not arguing for.
>  
> Thanks,
> Nick
>  
> On Jun 26, 2013, at 5:45 AM, Jonathan Mayer <jmayer@stanford.edu (mailto:jmayer@stanford.edu)> wrote:
> > On Wednesday, June 26, 2013 at 4:35 AM, Justin Brookman wrote:
> > > This change is unnecessary.  The standard already states that deidentified data is out of scope.
> >  
> > On re-read, it looks like the No Personalization section might only apply to permitted use data.  In which case maybe it's redundant with the No Secondary Uses section?
> > > This language (or at least concept) has long been stable in the document
> >  
> > I don't believe we ever reached consensus on whether DNT allows privacy-preserving personalization.
> > > , and we shouldn't be challenging every uncontroversial wording decision at the 11th hour.
> > > > From: Jonathan Mayer [mailto:jmayer@stanford.edu]
> > > > To: public-tracking@w3.org (mailto:public-tracking@w3.org) Group WG [mailto:public-tracking@w3.org]
> > > > Sent: Wed, 26 Jun 2013 02:57:02 -0500
> > > > Subject: June Change Proposal: Personalization
> > > >  
> > > > I would propose dropping section 5.1.3 ("No Personalization").  I would be comfortable allowing third parties to personalize a user's web experience… so long as it is done in a privacy-preserving way, with rigorously de-identified data.  
> > > >  
> >  
>