W3C home > Mailing lists > Public > public-tracking@w3.org > June 2013

Re: June Change Proposal, tracking

From: Roy T. Fielding <fielding@gbiv.com>
Date: Wed, 26 Jun 2013 18:09:52 -0700
Cc: "public-tracking@w3.org public-tracking@w3.org" <public-tracking@w3.org>
Message-Id: <17D1E66F-36D3-427B-83F6-94CB2C5EE030@gbiv.com>
To: Nicholas Doty <npdoty@w3.org>
On Jun 26, 2013, at 4:57 PM, Nicholas Doty wrote:

> Hi Roy,
> I've moved ISSUE-5 to the Compliance June product; I believe that existing issue closely tracks the topic of this change. 
> I've set up a wiki page for this proposal: http://www.w3.org/wiki/Privacy/TPWG/Change_Proposal_Tracking_Definition
> The wiki page also has the text from the editors' draft, for comparison.
> I'm not sure I caught your explanation during the call today, but one interpretation of this text would also suggest changes to the party, first party and third party definitions to instead use "contexts", and perhaps a re-writing of the first and third-party compliance sections. Is that your intention with this proposal?

It can go either way.  It is primarily meant as a middle
ground between the folks who like the simplicity of first
party being defined by ownership, and the other folks who
believe conglomerates that happen to own thousands of
brands should not be given a free pass to track across
sites that don't even share common branding.

The definition of first party could be rewritten in terms of
the user's current browsing context, thereby fixing several
problems with the notion that a user interacts with only one
site owner per interaction.  [E.g., the existing text about the
first party being the domain name of the page is just plain
wrong, as the first party in any given set of network
interactions often starts with one site (e.g., search) and
ends with another site, and each one is considered the first
party depending on where you are in the set of network
interactions that the UA is processing in order to build
the next page.]

Or, the definition of context could be more specifically
defined (for the sake of this spec) to be the first party,
and we could spend more time trying to explain why there are
multiple first parties in a network interaction.

Or, network interaction could be redefined in some way.

Or, if nobody like this notion of context, we can go back to
the "across unaffiliated websites" in Shane's proposal,
which is an antiquated way to think about the Web and HTTP use,
but would at least correspond to the current requirements.

Regardless, this wording allows the notion of "what is tracking"
to be discussed separately from the scope of what the group
wants to allow as a common context. I have removed the
implication of common ownership from the scope.  If we can't
agree on what tracking is, regardless of allowed scope, then
there really is no point in continuing further.

Received on Thursday, 27 June 2013 01:10:16 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:39:42 UTC