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

RE: Identity providers as first parties

From: Shane Wiley <wileys@yahoo-inc.com>
Date: Sun, 17 Jun 2012 12:26:39 -0700
To: Tamir Israel <tisrael@cippic.ca>
CC: Rigo Wenning <rigo@w3.org>, "public-tracking@w3.org" <public-tracking@w3.org>, "rob@blaeu.com" <rob@blaeu.com>, Kimon Zorbas <vp@iabeurope.eu>, "ifette@google.com" <ifette@google.com>, "JC Cannon (Microsoft)" <jccannon@microsoft.com>
Message-ID: <63294A1959410048A33AEE161379C8023D18786A7D@SP2-EX07VS02.ds.corp.yahoo.com>
Tamir,

Jonathan's proposal does attempt to address this point but many in the room feel this should be left to local law.  Justin Brookman and I took a pass at this language but it shifted to becoming overly prescriptive (legislating via tech standard) so many in the WG asked for local law to determine.

I would suggest this conversation be extracted from Jonathan's proposal to be handled separately as the rest of proposal has been mostly rejected by those in the WG that are intended to implement DNT in the real-world (on the 1st party/3rd party side).

More to come in Seattle...

- Shane

-----Original Message-----
From: Tamir Israel [mailto:tisrael@cippic.ca] 
Sent: Sunday, June 17, 2012 12:19 PM
To: Shane Wiley
Cc: Rigo Wenning; public-tracking@w3.org; rob@blaeu.com; Kimon Zorbas; ifette@google.com; JC Cannon (Microsoft)
Subject: Re: Identity providers as first parties

Shane -- I am not remotely attempting doing so.

As far back as I can see, the spec was going to put conditions on the 
means by which out of band consent can be sought.

Jonathan et al's proposal is:

1. Actual presentation: The choice mechanism MUST be actually presented 
to the user. It MUST NOT be on a linked page, such as a terms of service 
or privacy policy.
2. Clear terms: The choice mechanism MUST use clear, non-confusing 
terminology.
3. Independent choice: The choice mechanism MUST be presented 
independent of other choices. It MUST NOT be bundled with other user 
preferences.
4. No default permission: The choice mechanism MUST NOT have the user 
permission preference selected by default.

On 6/17/2012 3:16 PM, Shane Wiley wrote:
> Tamir,
>
> That's up to local laws to determine.  Please do not attempt to legislate via W3C tech standard.
>
> - Shane
>
> -----Original Message-----
> From: Tamir Israel [mailto:tisrael@cippic.ca]
> Sent: Sunday, June 17, 2012 12:14 PM
> To: Shane Wiley
> Cc: Rigo Wenning; public-tracking@w3.org; rob@blaeu.com; Kimon Zorbas; ifette@google.com; JC Cannon (Microsoft)
> Subject: Re: Identity providers as first parties
>
> Shane -- Out of band consent *does* trump DNT-1. We are now trying to
> define the parameters by which out of band consent can be sought.
>
> Best,
> Tamir
>
> On 6/17/2012 3:11 PM, Shane Wiley wrote:
>> Tamir,
>>
>> Out-of-band consent trumps DNT.  We've been repeating this mantra for over a year now - becoming repetitive.
>>
>> - Shane
>>
>> -----Original Message-----
>> From: Tamir Israel [mailto:tisrael@cippic.ca]
>> Sent: Saturday, June 16, 2012 5:23 PM
>> To: Shane Wiley
>> Cc: Rigo Wenning; public-tracking@w3.org; rob@blaeu.com; Kimon Zorbas; ifette@google.com; JC Cannon (Microsoft)
>> Subject: Re: Identity providers as first parties
>>
>> Shane --
>>
>> Just so we're really clear: if a user authenticates with Yahoo! on site
>> A and controls preferences on that site, does the out of band consent
>> dialogue Jonathan showed invalidate DNT-1: on site A? in general?
>>
>> Best,
>> Tamir
>>
>> On 6/15/2012 11:29 PM, Tamir Israel wrote:
>>> Ok.
>>>
>>> On 6/15/2012 2:07 PM, Shane Wiley wrote:
>>>> DAA Opt-out and single-sign on are not related.  There are some
>>>> implementations where the ID is needed beyond the authentication
>>>> event and therefore data collection occurs outside of the initial
>>>> authentication event.  Users do NOT need to choose Yahoo! as their ID
>>>> provider if they feel uncomfortable with that outcome.
>>>>
>>>> - Shane
>>>>
>>>> -----Original Message-----
>>>> From: Tamir Israel [mailto:tisrael@cippic.ca]
>>>> Sent: Friday, June 15, 2012 10:56 AM
>>>> To: Shane Wiley
>>>> Cc: Rigo Wenning; public-tracking@w3.org; rob@blaeu.com; Kimon
>>>> Zorbas; ifette@google.com; JC Cannon (Microsoft)
>>>> Subject: Re: Identity providers as first parties
>>>>
>>>> Shane,
>>>>
>>>> Maybe we are getting sidetracked.
>>>>
>>>> Can you please explain the scope of tracking that results from using
>>>> Yahoo!'s IdM mechanism? Does it mean you can track all my activities on
>>>> the specific authenticated site? If so does this carry across multiple
>>>> explicitly authenticated sites? Does it operate in a manner analogous to
>>>> single sign-on? How does it interact with the existing DAA opt-out?
>>>>
>>>> Thanks and best regards,
>>>> Tamir
>>>>
>>>> On 6/15/2012 11:28 AM, Shane Wiley wrote:
>>>>> Tamir,
>>>>>
>>>>> Any service gets to determine its own primary purpose - so if OBA is
>>>>> the payment for the service and this is disclosed as a primary
>>>>> purpose, then that's the bargain the users can choose to consent to
>>>>> or not.
>>>>>
>>>>> - Shane
>>>>>
>>>>> -----Original Message-----
>>>>> From: Tamir Israel [mailto:tisrael@cippic.ca]
>>>>> Sent: Friday, June 15, 2012 8:21 AM
>>>>> To: Shane Wiley
>>>>> Cc: Rigo Wenning; public-tracking@w3.org; rob@blaeu.com; Kimon
>>>>> Zorbas; ifette@google.com; JC Cannon (Microsoft)
>>>>> Subject: Re: Identity providers as first parties
>>>>>
>>>>> Shane --
>>>>>
>>>>> There are 2 questions here. One is whether you can bundle in the
>>>>> obligation to consent to secondary purposes as a condition of
>>>>> authentication in an IdM context. The primary service in an IdM context
>>>>> is authentication, not OBA.
>>>>>
>>>>> The second is to what extent the DNT spec should address this. I took
>>>>> the 'independent choice' out of band consent criteria as an attempt to
>>>>> prevent bundling of choices.
>>>>>
>>>>> Best,
>>>>> Tamir
>>>>>
>>>>> On 6/15/2012 11:06 AM, Shane Wiley wrote:
>>>>>> Tamir,
>>>>>>
>>>>>> But in the use case we're discussing the service being provided is
>>>>>> the primary purpose - a user's online identity.  A service
>>>>>> determines its primary purpose, discloses this to the user, user
>>>>>> consents.  Case closed.
>>>>>>
>>>>>> - Shane
>>>>>>
>>>>>> -----Original Message-----
>>>>>> From: Tamir Israel [mailto:tisrael@cippic.ca]
>>>>>> Sent: Friday, June 15, 2012 8:02 AM
>>>>>> To: Shane Wiley
>>>>>> Cc: Rigo Wenning; public-tracking@w3.org; rob@blaeu.com; Kimon
>>>>>> Zorbas; ifette@google.com; JC Cannon (Microsoft)
>>>>>> Subject: Re: Identity providers as first parties
>>>>>>
>>>>>> Shane, I disagree. Under PIPEDA you should offer users the possibility
>>>>>> of opting out of collection, use or disclosure for purposes
>>>>>> secondary to
>>>>>> the primary service being offered.
>>>>>>
>>>>>> This is the basis of the opt-out consent scheme being applied to
>>>>>> online
>>>>>> tracking.
>>>>>>
>>>>>> Best,
>>>>>> Tamir
>>>>>>
>>>>>> On 6/15/2012 10:58 AM, Shane Wiley wrote:
>>>>>>> Tamir,
>>>>>>>
>>>>>>> I disagree and PIPEDA does as well.  As long as you're clear to a
>>>>>>> user what a service provides and a user expressly consents to
>>>>>>> those practices, the discussion is over.
>>>>>>>
>>>>>>> Please don't try to raise CA regulatory schemes into conversations
>>>>>>> on one hand then completely reverse your stance at whim - this
>>>>>>> seriously undermines your credibility.
>>>>>>>
>>>>>>> - Shane
>>>>>>>
>>>>>>> -----Original Message-----
>>>>>>> From: Tamir Israel [mailto:tisrael@cippic.ca]
>>>>>>> Sent: Friday, June 15, 2012 7:54 AM
>>>>>>> To: Shane Wiley
>>>>>>> Cc: Rigo Wenning; public-tracking@w3.org; rob@blaeu.com; Kimon
>>>>>>> Zorbas; ifette@google.com; JC Cannon (Microsoft)
>>>>>>> Subject: Re: Identity providers as first parties
>>>>>>>
>>>>>>> Shane --
>>>>>>>
>>>>>>> The need for independent choice is critical, I think, to the out
>>>>>>> of band
>>>>>>> consent scheme. You shouldn't be able to force users out of their DNT
>>>>>>> choices as a condition of authentication.
>>>>>>>
>>>>>>> Best,
>>>>>>> Tamir
>>>>>>>
>>>>>>> On 6/15/2012 10:48 AM, Shane Wiley wrote:
>>>>>>>> Rigo,
>>>>>>>>
>>>>>>>> DNT will NEVER trump an out-of-band consent.  The user would
>>>>>>>> simply withdraw from using the service they had provided prior
>>>>>>>> consent to.  If the product would like to offer two levels of
>>>>>>>> service, it can of course do that, but that would be completely
>>>>>>>> outside the scope of DNT.
>>>>>>>>
>>>>>>>> DNT is not the privacy silver bullet and answer to all privacy
>>>>>>>> issues on the Internet - let's stop trying to push it in that
>>>>>>>> direction.
>>>>>>>>
>>>>>>>> Thank you,
>>>>>>>> - Shane
>>>>>>>>
>>>>>>>> -----Original Message-----
>>>>>>>> From: Rigo Wenning [mailto:rigo@w3.org]
>>>>>>>> Sent: Friday, June 15, 2012 1:28 AM
>>>>>>>> To: public-tracking@w3.org
>>>>>>>> Cc: Shane Wiley; rob@blaeu.com; Kimon Zorbas; ifette@google.com;
>>>>>>>> Tamir Israel; JC Cannon (Microsoft)
>>>>>>>> Subject: Re: Identity providers as first parties
>>>>>>>>
>>>>>>>> Shane, Kimon,
>>>>>>>>
>>>>>>>> On Thursday 14 June 2012 16:47:03 Shane Wiley wrote:
>>>>>>>>> I’ve used a few others and they appears to do the same so I’m
>>>>>>>>> confused as to what real-world identity provider scenario someone
>>>>>>>>> is considering where consent wasn’t already obtained?
>>>>>>>> I confirm that we agreed that the out-of-band agreement will trump
>>>>>>>> the DNT:1 signal. We also agreed that the service has to signal this
>>>>>>>> to the client.
>>>>>>>>
>>>>>>>> I guess, what Rob is trying to achieve is to say, even in this
>>>>>>>> context, a service could offer the choice of stopping to track and
>>>>>>>> only use information for the login/authentication purpose. This
>>>>>>>> could be the meaning of DNT:1 if the Service sends ACK in a
>>>>>>>> login/authentication context. If you're looking for medical
>>>>>>>> information in a login context, you don't want your login provider
>>>>>>>> to spawn that to your insurance. I think this is a very legitimate
>>>>>>>> use case. The service could say: "yes, I see your point" and send
>>>>>>>> ACK instead of "out-of-band".
>>>>>>>>
>>>>>>>> We are just defining switches. People will decide whether they
>>>>>>>> switch stuff on or off or provide a switch at all.
>>>>>>>>
>>>>>>>> Rigo
Received on Sunday, 17 June 2012 19:27:23 UTC

This archive was generated by hypermail 2.3.1 : Friday, 21 June 2013 10:11:30 UTC