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

Re: Multiple First Parties

From: Lauren Gelman <gelman@blurryedge.com>
Date: Fri, 21 Sep 2012 11:31:04 -0700
Cc: Vinay Goel <vigoel@adobe.com>, Rob Sherman <robsherman@fb.com>, Chris Pedigo <CPedigo@online-publishers.org>, Justin Brookman <justin@cdt.org>, "public-tracking@w3.org" <public-tracking@w3.org>
Message-Id: <C92E8A40-82FC-4174-ABAA-EAF61AA8DFE7@blurryedge.com>
To: Jeffrey Chester <jeff@democraticmedia.org>

This is a very interesting example! 

If you click privacy you get links to both ATT and Yahoo's policies with no overt information about how they interact.

"In order to provide you with online access, personalized content, customized advertising and many other valuable services, AT&T and Yahoo! collect and use information about you. We at AT&T and Yahoo! are committed to maintaining high standards of data privacy. AT&T and Yahoo! each maintain separate policies to describe how we treat your information. Please take a moment to read our privacy policies."

In ATT's policy there is a section specifically referring to Yahoo:

Online Activity Tracking and Advertising
We collect information about your activity on AT&T websites for a number of purposes using technologies such as cookies, Flash cookies, Web beacons, widgets and server log files.
We and our non-AT&T advertising partners use that information, as well as other information they have or we may have, to help tailor the ads you see on our sites and to help make decisions about ads you see on other sites.
Opt-out of ad matching by Yahoo!.
Opt-out of targeting advertising from many other ad networks.
Opt out of YP.com's targeted advertising program.

Lauren Gelman
BlurryEdge Strategies

On Sep 21, 2012, at 9:37 AM, Jeffrey Chester wrote:

> That is very interesting.   How would a user know the different data collection practices run by the parties, and its implications?  What does ATT.net do with the data versus Yahoo?  What is shared and used by both parties internally and operationalized?  Or shared with third parties, used by ad exchanges, etc.
> This is a good example to fully flesh out the data practices on co-branded sites to understand what it means for privacy under the DNT frame.  I hope you and colleagues to build on this so we have a living example to consider.
> Thanks,
> Jeff
> Jeffrey Chester
> Center for Digital Democracy
> 1621 Connecticut Ave, NW, Suite 550
> Washington, DC 20009
> www.democraticmedia.org
> www.digitalads.org
> 202-986-2220
> On Sep 21, 2012, at 12:26 PM, Vinay Goel wrote:
>> Hi Jeff,
>> Here's one example: http://att.yahoo.com
>> -Vinay
>> From: Jeffrey Chester <jeff@democraticmedia.org>
>> Date: Friday, September 21, 2012 9:33 AM
>> To: Rob Sherman <robsherman@fb.com>
>> Cc: Chris Pedigo <CPedigo@online-publishers.org>, Justin Brookman <justin@cdt.org>, "public-tracking@w3.org" <public-tracking@w3.org>
>> Subject: Re: Multiple First Parties
>> Resent-From: <public-tracking@w3.org>
>> Resent-Date: Friday, September 21, 2012 9:34 AM
>> Rob:  Thanks for all this.  Can you give us a real world example of a co-run site?   What are the models we can examine to help us better understand the implications for users?
>> Regards,
>> Jeff
>> Jeffrey Chester
>> Center for Digital Democracy
>> 1621 Connecticut Ave, NW, Suite 550
>> Washington, DC 20009
>> www.democraticmedia.org
>> www.digitalads.org
>> 202-986-2220
>> On Sep 21, 2012, at 3:01 AM, Rob Sherman wrote:
>>> Thanks very much for all of this feedback.  As I understand it, the group
>>> generally agrees that the party responsible for a website that a user
>>> visits is a first party on that website.  Text in the existing draft
>>> acknowledges that, in some circumstances, there may be more than one party
>>> responsible.  The point of my proposal is to provide context around that
>>> concept so that parties have some guidance in the spec about how to
>>> determine whether they fall into this category.  Currently, we simply say
>>> that it may sometimes happen and leave it at that.  The Example Sports on
>>> Example Social example - which comes from Jonathan and Tom's text - is an
>>> attempt to illustrate the point, and what I've tried to do is to elaborate
>>> a bit on what it is about Example Sports and Example Social that make them
>>> both first parties in that instance.
>>> I agree with Mike that the meaningful interaction standard doesn't apply
>>> here.  To be clear, we're talking about two distinct situations:  (1) a
>>> basic third party, such as a "share" button, which is a third party but
>>> becomes a first party when the user interacts with it; and (2) a single
>>> website that is operated by two first parties operating together.  In that
>>> second scenario, just as we agree that a user intends to interact with the
>>> entity responsible for a website when he/she browses to that website, it
>>> seems reasonable to draw the same conclusion when there are two entities
>>> responsible.  This should not implicate Jeff's concern about giving
>>> parties a "free pass" on DNT because, although I think branding is an
>>> important way to ensure that consumers understand who is responsible for a
>>> website, nobody is suggesting that putting a logo on a website, without
>>> more, gives a party license to ignore DNT.
>>> My goal here is simply to describe the concept of multiple first parties,
>>> which has been in the draft for some time and is a concept that I think
>>> most people in the TPWG understand, in a way that helps parties who have
>>> not been a part of our discussions implement the spec in a way that is
>>> consistent with what we envision.
>>> Rob Sherman
>>> Facebook | Manager, Privacy and Public Policy
>>> 1155 F Street, NW Suite 475 | Washington, DC 20004
>>> office 202.370.5147 | mobile 202.257.3901
>>> On 9/20/12 7:08 AM, "Chris Pedigo" <CPedigo@online-publishers.org> wrote:
>>>> Rob, thanks for this clarifying language.  I believe it reflects the
>>>> group's previous decisions on first parties and provides some useful
>>>> guidance for implementers.
>>>> Justin, I don't see how this would be an expansion.  Can you clarify?
>>>> -----Original Message-----
>>>> From: Justin Brookman [mailto:justin@cdt.org]
>>>> Sent: Thursday, September 20, 2012 10:01 AM
>>>> To: public-tracking@w3.org
>>>> Subject: Re: Multiple First Parties
>>>> The existing language already allows for multiple first parties despite
>>>> no meaningful interaction.  Rob (Sherman) is arguing for an expansion.
>>>> I have previously argued against multiple first parties, but I do not
>>>> believe many agreed with me.  The Example Sports on Example Social is an
>>>> interesting example that may be consistent with Jonathan's original
>>>> formulation (he and Tom drafted the original language), though I still
>>>> think we need more to be clear that mere branding and disclosure are not
>>>> sufficient.
>>>> Justin Brookman
>>>> Director, Consumer Privacy
>>>> Center for Democracy & Technology
>>>> 1634 I Street NW, Suite 1100
>>>> Washington, DC 20006
>>>> tel 202.407.8812
>>>> fax 202.637.0969
>>>> justin@cdt.org
>>>> http://www.cdt.org
>>>> @CenDemTech
>>>> @JustinBrookman
>>>> On 9/20/2012 9:52 AM, Jeffrey Chester wrote:
>>>>> I also agree that the meaningful interaction standard should apply.
>>>>> Just because a site may have a syndicated presence on a first part page
>>>>> shouldn't give it a free pass.  Sites could engage in co-branding to
>>>>> wipe out DNT safeguards.
>>>>> On Sep 20, 2012, at 9:24 AM, Mike Zaneis wrote:
>>>>>> Rob,
>>>>>> I don't think the meaningful interaction standard covers what is being
>>>>>> presented here. Meaningful interaction contemplates a user action after
>>>>>> they visit the site. What the examples Rob Sherman provides show is a
>>>>>> clear understanding by the user that there are multiple first parties
>>>>>> upon landing on a particular page (am I getting that right Rob
>>>>>> Sherman?).
>>>>>> I think this is a vitally important distinction for us to make since
>>>>>> the Internet is evolving to provide more examples of this dual
>>>>>> content/owner page. It just needs to be clear to the user that there
>>>>>> are multiple first parties and providing some factors of indicia in the
>>>>>> standard would be helpful.
>>>>>> Mike Zaneis
>>>>>> SVP & General Counsel, IAB
>>>>>> (202) 253-1466
>>>>>> On Sep 20, 2012, at 1:42 AM, "Rob van Eijk" <rob@blaeu.com> wrote:
>>>>>>>>> In these instances, a party will be deemed a first party on a
>>>>>>>>> particular website if it concludes that a user would reasonably
>>>>>>>>> expect to communicate with it using the website.
>>>>>>> Hi Rob,
>>>>>>> This would imply a change of the first party definition, which is
>>>>>>> covered elsewhere in the document. Isn't your scenarion already
>>>>>>> covered with the priniple of meaningful interaction?
>>>>>>> tnks::Rob
>>>>>>> Rob Sherman schreef op 2012-09-19 22:34:
>>>>>>>> *
>>>>>>>>   *
>>>>>>>> The editors' draft of the compliance spec raises a question about
>>>>>>>> how to define the circumstances in which more than one entity
>>>>>>>> operates as a first party on a particular website. As drafted, the
>>>>>>>> first option leaves more questions than answers because it says
>>>>>>>> that this may happen in some circumstances but does not provide any
>>>>>>>> concrete guidance on how a party can tell when it is a first party.
>>>>>>>> I've proposed text below that I hope leaves intact the basic intent
>>>>>>>> behind the existing text - including two examples that are already
>>>>>>>> there as options - but that elaborates a bit on the examples and
>>>>>>>> provides some non-normative guidance about factors that an entity
>>>>>>>> might consider in making a judgment whether it qualifies as a first
>>>>>>>> party. The thinking is that, although we can't - and should not try
>>>>>>>> to - anticipate the specifics every situation in which two entities
>>>>>>>> collaborate, it would be helpful to provide some guidance in the
>>>>>>>> text to people who are not in the Working Group and who may not
>>>>>>>> have the context for situations that this section envisions.
>>>>>>>> Feedback on this text would, of course, be appreciated.
>>>>>>>> Rob
>>>>>>>> # # #
>>>>>>>> _<NORMATIVE>_
>>>>>>>> For many websites, there will be only one party that the average
>>>>>>>> user would expect to communicate with: the provider of the website
>>>>>>>> the user has visited. But, for other websites, users may expect to
>>>>>>>> communicate with more than one party. In these instances, a party
>>>>>>>> will be deemed a first party on a particular website if it
>>>>>>>> concludes that a user would reasonably expect to communicate with it
>>>>>>>> using the website.
>>>>>>>> _<NON-NORMATIVE>_
>>>>>>>> URIs, branding, the presence of privacy policies or other
>>>>>>>> disclosures that specifically identify a party, and the extent to
>>>>>>>> which a party provides meaningful content or functionality on the
>>>>>>>> website, may contribute to, but are not necessarily determinative
>>>>>>>> of, user perceptions about whether a website is provided by more
>>>>>>>> than one party.
>>>>>>>> _Example: _Example Sports, a well-known sports league, collaborates
>>>>>>>> with Example Streaming, a well-known streaming video website, to
>>>>>>>> provide content on a sports-themed video streaming website. The
>>>>>>>> website is prominently advertised and branded as being provided by
>>>>>>>> both Example Sports and ExampleStreaming. An ordinary user who
>>>>>>>> visits the website may recognize that it isoperated by both Example
>>>>>>>> Sports and Example Streaming. Both Example Sports and Example
>>>>>>>> Streaming are first parties.
>>>>>>>> _Example:_ Example Sports has a dedicated page on a Example Social,
>>>>>>>> a social networking website. The page is branded with both Example
>>>>>>>> Sports' name and logo and Example Social's name and logo. Both
>>>>>>>> Example Sports' name and Example Social's names appear in the URI
>>>>>>>> for the page. When a user visits this dedicated page, both Example
>>>>>>>> Sports and Example Social are first parties.
>>>>>>>> Rob Sherman
>>>>>>>> 1155 F Street, NW Suite 475 | Washington, DC 20004
>>>>>>>> office 202.370.5147 | mobile 202.257.3901

Received on Friday, 21 September 2012 18:31:36 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:39:00 UTC