Tracking Protection Working Group Teleconference -- 03 Dec 2014

<trackbot> Date: 03 December 2014

volunteers to scribe? shouldn't be very strenuous

<scribe> scribenick: npdoty

justin: only a few things left open to discuss, talk about process at the end of the call
... first checkin with dsinger
... resolving javascript issues. add Mike's expiry parameters?

dsinger: sorry, not yet
... know what to do in all except one case, about origin/effective-script-origin
... if anyone else wants to weigh in, that would be useful

justin: anyone with an opinion?

dsinger: whether document.domain should affect the origin or not

justin: send an email to the group about planning to go with existing unless anyone else objects

<WileyS> +1 to staying with Roy's approach

<moneill2_> +q

<justin> npdoty: What's the functional difference between the two?

moneill2_: when you load the script from a library, doesn't matter which origin it's loaded from
... the origin of the site you're on (scheme, host, port)
... depends on how you access subdomains. the javascript library could set document.domain to remove a subdomain
... can give a different domain off of the main domain
... anne's objection was less about the definition of the term, but a problem about using cookie-like rules with subdomains
... going out of favor because some subdomains are controlled by different parties, etc.
... want to get the standard out. <lost connection>

<justin> npdoty: Thanks to Mike for the explanation.

<moneill2_> my phone is playing up,

<justin> npdoty: Good reasons for following cookie-like rules.

npdoty: +1, keep as is to keep cookie-like setting

<WileyS> +1 to David - let's not increase scope

dsinger: right, keep current practice, not effect new practice

justin: sounds like we're all in agreement on that. dsinger, please send a note about that

issue-262?

<trackbot> issue-262 -- guidance regarding server responses and timing -- pending review

<trackbot> http://www.w3.org/2011/tracking-protection/track/issues/262

<dsinger> issue-262?

<trackbot> issue-262 -- guidance regarding server responses and timing -- pending review

<trackbot> http://www.w3.org/2011/tracking-protection/track/issues/262

justin: we were expecting a submission from fielding on this

fielding: total slacker ;) same plan as last time, but haven't had a chance to send email

<vincent> yes

<rvaneijk> yep, but vincent (and I) prepared something in the mean time

<vincent> sent a first version but I'll adapt to Roy's proposal

justin: know vincent and rvaneijk were looking at a proposal as well
... did you review nick's changes regarding issue-203?

fielding: yes, don't recall issues with that, but will remind myself and send issues to the mailing list

justin: nick, can you explain again?

http://lists.w3.org/Archives/Public/public-tracking/2014Nov/0034.html

<vincent> npdoty, here is the link: http://lists.w3.org/Archives/Public/public-tracking/2014Dec/0004.html

<justin> npdoty: I tried to address fielding's concern that a party might not always know what it is.

<justin> npdoty: Tried to adapt fielding's langauge about resources that are intended to be used in a first or third party capacity.

<justin> npdoty: separate section of "unknowing" collection.

<justin> npdoty: I had hoped that would address fielding's concerns about sites' not always knowing. And I still want to merge fielding's indication about indicating tracking status.

justin: hope fielding can review later today. sounds like we're fairly close conceptually. hopefully we can hammer out the language on email
... one more question for roy, proposed response about validation or indication of who set the dnt signal

fielding: sometime today

thanks

justin: auditability language that walter had proposed

<WileyS> Could someone please send the link to the CfO?

justin: Nick had announced a Call for Objections on that a couple of weeks ago

https://www.w3.org/2002/09/wbs/49311/tpwg-auditing-235/

<WileyS> I would like to simply cut/paste our email arguments into the CfO.

(also included in the agenda)

<WileyS> Thank you Nick

<WileyS> I do

justin: if people need a couple more days?

<fielding> I can put something in today

<WileyS> That works

justin: extend til Friday?

may ping walter about that timing as well

<vincent> ok

<rvaneijk> could you please repeat the question in irc?

http://lists.w3.org/Archives/Public/public-tracking/2014Dec/0004.html

<justin> ack vincent on issue-262

npdoty: we heard a proposal from vincent on 262, do we want to explain it?

vincent: if a server is a gateway, it should send a G response. are there other use cases besides real-time bidding where it will apply?
... if a server knows that it won't transmit data related to the current transaction, then it can send an N response
... G would stand for "gateway"

justin: requirements regarding winning/losing bidders?

<fielding> if adopted, I would rephrase it as a requirement not to send G if the server is not forwarding tracking data

vincent: we assume that the user won't receive a direct response from the losing bidders, only the winning bidder.

justin: how would this differ from fielding?

<rvaneijk> the proposal may change, depending on roy's proposal

vincent: not sure that it will (that's why I waited until today), but will adapt based on what I hear from roy

<fielding> I think that is a reasonable clarification, though my presumption would be that a gateway doesn't know what might become tracking data

justin: let's try to iterate on the mailing list to close these substantive issues on the mailing list
... so that we don't need to have regular calls next year, except as necessary

Process

justin: chairs need to send out response about comments that don't raise new issues

<justin> npdoty: To transition out of LC, we need to reach out to everyone who wrote comments in some public way, take all the issue resolutions and then take to the director.

justin: chairs can send email to individuals about resolutions of issues

npdoty: need the group's agreement to advance to Candidate Recommendation, and address any outstanding formal objections

justin: regarding Compliance to Last Call, did that process before, anything different?

npdoty: yes, similar steps. we got quite a few comments, so we seem to be getting that outreach reasonably well

justin: any other comments regarding process?

<fielding> do we need more use cases describedin TPE?

rvaneijk: regarding stakeholders from European E DAA, if we get new requirements from them

justin: what might there issues be?

rvaneijk: they haven't been involved at all. they might be interested in learning whether Do Not Track would work for them or not.

justin: did receive Last Call comments from trade associations. was there overlap with IAB EU people?

WileyS: IAB UK overlap with EDAA (Kimon, who had been participating, not involved)

thanks for the explanation

<fielding> we can only address comments after they have been sent to us, so there is nothing we can do about people who wait until it is too late. There are plenty of later stages in which comments might be addressed, particurlarly if they are about technical limitations.

justin: have had opportunities to weigh in, a Last Call on TCS would also give that opportunity
... Candidate Recommendation wouldn't be finished either, even if we believe it's ready for implementation
... regarding use cases?

fielding: in general, I think lots of people would like more examples and use cases, never got around to writing into the document
... is it okay for me to add more to that non-normative section without disrupting Last Call?

justin: sounds okay to me
... as long as people have the opportunity to see
... useful for explaining what we're tying to accomplish

any other issues for this call?

<WileyS> Thank you!

three issues, I'll try to push people on the mailing list

justin: a couple more calls in December, but probably not talk on Christmas Eve

thanks all

trackbot, end meeting

- DRAFT -

Tracking Protection Working Group Teleconference

03 Dec 2014

Attendees

Contents

Process

Summary of Action Items

Scribe.perl diagnostic output