Re: Status of non-normative text sections

On Feb 2, 2015, at 9:25 AM, David Singer <singer@apple.com> wrote:
> 
>> On Jan 12, 2015, at 15:28 , Nick Doty <npdoty@w3.org> wrote:
>> 
>> Explaining party distinction:
>> The October 2012 draft includes text explaining why party is relevant for a particular interaction, because it can change. I believe we’ve generally improved that text subsequently, as we now always refer to first- and third-party as related to a given user action (and have defined terminology on user action) rather than any implicit connotation about the status of an organization. I don’t believe we need separate non-normative text to explain that motivation in the actual text of the recommendation.
> 
> This may be opening a can of worms, but…I am not sure I understand the need for a party distinction, given our current definition of ‘tracking’.  When I proposed ‘tunnel vision’ it was partly to remove dependency on something that cannot be easily determined (which is the web site the user ‘intended’ to interact with?), and since then we have, I think agreed that ‘no cross-site tracking’ and tunnel vision are the same.

We aren't currently referring to "tunnel vision" in any of our documents in normative or non-normative text. I don't believe your earlier "tunnel vision" text completely captures all of the definition of "tracking" or agreements in Compliance. For example, when a user navigates to a page from a different site, or clicks on an advertisement, compliance with a DNT preference doesn't require not seeing where a user came from or what page they were on when they clicked on an ad. A party may be a first party *to a given user action* (that's the key part) where users are likely to have different expectations.

As I've said above, I don't believe there are any places in the document where text refers to first and third parties where it doesn't describe that status with respect to a particular user action, but I would welcome review to see that I've done that correctly.

Thanks,
Nick