W3C

- DRAFT -

Tracking Protection Working Group Teleconference

16 Apr 2014

See also: IRC log

Attendees

Present
Ninja, Wendy, ChrisIAB, hefferjr, dsinger, Jack_Hobaugh, RichardWeaver, Peder_Magee, rvaneijk, dwainberg, npdoty, justin, WileyS, +1.323.253.aaaa, Chris_Pedigo, Ari, schunter, MattHayes, [FTC], Brooks, Chris_IAB, [Microsoft], +1.202.370.aabb, robsherman, vinay, WaltMichel, +1.201.530.aacc, hwest, moneill2, Fielding, +1.312.923.aadd, mecallahan, vincent, adrianba
Regrets
Chair
justin, schunter
Scribe
ninja

Contents


<trackbot> Date: 16 April 2014

<Chris_IAB> I just joined

<Chris_IAB> maybe it's me?

<Chris_IAB> btw, am I the only one hearing some static on the line?

<sidstamm> hey all, sorry I can't dial in again this week, but will monitor irc

<Chris_IAB> I'm back - no static this time :)

thanks, ChrisIAB, much better, no echo

<Chris_IAB> I am muted

<scribe> scribenick: ninja

<WileyS> Thank you!

<Chris_IAB> seltzer, npdoty, that's odd... as I'm physically muted on my side

<Chris_IAB> let me call back in again

justin: On the plan how to move forward. We will try to do a better job about announcing the agenda earlier.

<justin> https://www.w3.org/2011/tracking-protection/track/issues/207

<Chris_IAB> dialed back in now

justin: We will take up compliance issues again. We will follow the usual procedure, asking for text proposals and discussing them and if we do not find consensus get to a Call for objection.
... Today talk about ISSUE-207 the disregard signal

ISSUE-207: Conditions for dis-regarding (or not) DNT signals

<justin> https://www.w3.org/wiki/Privacy/TPWG/Change_Proposal_Disregarding

justin: Jonathan Mayer had put up a text proposal on servers must not disregard a technically valid signal.

<Chris_IAB> did I miss something? are we talking about a compliance provision in the TPE?

<Chris_IAB> oh, did we shift now to working on the TCS?

justin: We are talking about how to address the disregard signal in the Compliance and Scope specification.

<WileyS> We of course strongly disagree with that perspective. A Server should be able to send the Disregard response anytime they feel there is a material issue with the signal or the sender of the signal.

justin: There are a lot of people in the group that disagree with Jonathan's proposal. Currently there are no provisions on when to send this signal.
... When I looked at the document this morning I thought that Jonathan's proposal may not be necessary. Based on the current text for the D signal in the TPE.

<wseltzer> zakim mute WaltMichel

<Chris_IAB> I can't here anything

<Chris_IAB> better now

<WileyS> It was WaltMichel

<dwainberg> Why is this the first TCS question we're taking up now? And why are we considering a proposal from someone who's withdrawn from the group many months ago.

dwainberg: What is the reasoning for taking up this TCS issue first. And why taking up Jonathan's proposal when he left the group months ago.

<dsinger> my perception is that we have reached an uneasy peace and maybe opening this is going to open old wounds to no useful effect

justin: fair question, no magic to it. We thought it may be easier to tackle since closely related to our recent work and may not be overly controversial.

<npdoty> " A third party to a given user action that disregards a DNT signal MUST indicate so to the user agent, using the response mechanism defined in the [TRACKING-DNT] recommendation. "

<Zakim> npdoty, you wanted to comment on one sentence mention of disregard in compliance

<Chris_IAB> many, if not most of the provisions in the draft TCS, are out of date

justin: I heard previously from Walter and John Simpson support for Jonathan's proposal. We need to check if that's still the case.

<ChrisPedigoOPA> +q

<JackHobaugh> Which TCS draft are we working on?

npdoty: Referncing the section of the TCS spec on the D signal: " A third party to a given user action that disregards a DNT signal MUST indicate so to the user agent, using the response mechanism defined in the [TRACKING-DNT] recommendation. "

<npdoty> I copied/pasted from the editors' draft: http://www.w3.org/2011/tracking-protection/drafts/tracking-compliance.html

<WileyS> The two points we agreed upon earlier: (1) if you're going to disregard then you must share that with the user and (2) if you send Disregard you must also provide a resource link to explain why that position is being taken.

<WileyS> They changed the draft on us - so we need to fix it

<vinay_> Agreed with Shane. I thought that's where the group landed.

justin: We made these decisions for TPE, we may need to port these over to TCS. TCS may require additional transparency.

<Chris_IAB> when did this agenda item go out? I'm not feeling prepared to answer these questions today

<WileyS> Hearing no one - please close the issue.

justin: You may be right that no one wants to take up Jonathan's old proposal that servers must honor all DNT signals and not second guess them.

<WileyS> Justin, no one is arguing that a Server must always honor all signals no matter what. That's the whole purpose of having the D signal in the first place.

<WileyS> Can we please close this ticket? This is a waste of time.

<WileyS> One that we were given literally an hour to prepare for.

ChrisPedigoOPA: Circular argument to say that you must honor all signals. If the company wants to comply it will take the DNT signals serious. And transparency is always key in the communication with the user.

<dsinger> Quick question: does anyone WANT to discuss this (ever) again? Or can we close it as belonging to Mr Mayer?

<WileyS> Please close

justin: taking up dwainberg's question. Does someone wants to keep this issue open?

<npdoty> I'm fine with leaving the text as it is. (I was just trying to help us understand what the language in the document was.)

justin: I will send the question to the mailing list and if no one takes this up. Let's close it. Just want to make sure, that we do not close relevant issues

dwainberg: Will the chairs provide some kind of roadmap how to take on the compliance work. There is confusion about this particular document. Which version to use etc.

<mecallahan> aadd is mecallahan

<mecallahan> zakim aadd is mecallahan

<wseltzer> We have an editors' draft, with open issues

<npdoty> I believe this is the list of open/raised issues: https://www.w3.org/2011/tracking-protection/track/products/5

justin: Last summer the chairs decided to work on the previous June draft and collected issues up to October. We will take up this list of issues in the next weeks.

dwainberg: Will we walk at some point through the document and raise new issues?

justin: Think we have a pretty comprehensive list of open issues. We have identified 15ish areas where we will ask the group for new text proposals. New issues are currently on the side.

rvaneijk: Getting back to the D signal - Would a server in a chain be able to transfer the D signal down the chair. Asking about the transitivity.

justin: Not sure I get this completely.

<WileyS> In an Exchange model, if the platform itself sends a "D" then I could see a transitive structure issue but if individual bidders in the ad placement offer different responses (some "D", some not) I'm not sure transitive value is necessary.

rvaneijk: News website would answer a DNT signal with D and also transfer the D signal to all ad networks on the site.

<fielding> I can't think of any use case where D would be transitive.

<WileyS> The Publisher / Direct 3rd Party example is not a good use case for transitivity as each party responds on its own.

justin: I think in this example the publisher would answer N for Not tracking and the ad networks will answer in their own capacity.

<WileyS> Roy, there is a possible case in a Client-Server to Server-Server call chain - as exists in many Ad Exchanges.

<WileyS> Nick, to your point, they'd likely not forward the DNT signal in the server-to-server call

<npdoty> WileyS, right, and the first party in that case would respond 'D' directly to the user

<npdoty> npdoty: in the client-server case (301 redirects, etc.), each party can respond directly to the end user (with a D or not)

<npdoty> ... in the first party server to other servers case, the first party might choose to disregard, reply with D and then not comply with first party compliance restrictions (like sharing with a third party)

<dsinger> I don’t think ‘passing a D’ makes sense. Each response contains what that server thinks it is doing. If you want to *instruct* your thirtd parties to ignore something, they will need to respond ‘D’ and explain ‘Fred made me do it’ in their policy.

<Zakim> npdoty, you wanted to comment on transitivity

ninja: We will take up dwainbergs roadmap suggestion. Team will send out an overview of what the areas we have identified for compliance and will take up one by one.

<WileyS> Nick, agreed - we're on the same page

justin: I invite folks to take a look at TCS and the disregard signal. I found the language a little confusing. But if the WG is happy with it as it is, let's not make it more complicated.

<npdoty> yeah, if everyone is fine with the document, let's send out an email to confirm and then close it?

justin: I will send out an email to the group and point to this issue.

charter

Charter extension

<WileyS> What happens if a majority of the group doesn't want to extend?

wseltzer: We noticed that our current charter will end on 30 April this year. Our current work plan based on the poll and the discussion in the group
... goes up to June 2015 as a revised schedule.
... The director suggested to extend the charter not only up to this date but up to December 2015 to stay around longer to cover errata etc.

justin: What would happen if we do not extend the charter.

wseltzer: It is the director's decision. If the charter would not be extended the group would be out of charter.

justin: Could the WG formally object?

<WileyS> Could we change the Charter to simply complete the TPE and leave it at that?

<dwainberg> What is critical mass?

wseltzer: Working Group members are not able to tell other WG members to stop their work. If we have still interest in the work the WG will continue.

<WileyS> Its been 6 months since then

<WileyS> Many participants are no longer with the working group

<WileyS> If we took a poll now I believe the outcome would be very different

<WileyS> 3 or 4?

<WileyS> How else can new facts come out if we don't take a poll?

justin: Answering Shane's question. Decision on plan to move forward was to advance TPE first and then continue on compliance. Not inclined on do another poll. If there is no new information.
... Currently W3C and the chairs and at least part of the WG feel it is worthwhile to continue the discussion on compliance.

<Brooks> It seems like there is interest in such a poll. Am I wrong?

<dsinger> TTWG got down to 4 members at one point (maybe 5)

dwainberg: Wendy said that there needs to be critical mass. How many people are required?

<npdoty> the group can decide on its own work at any time; the charter is just a formal matter of whether the Consortium allows the group to work on things in its charter

wseltzer: We need different stakeholders in the process. Although it is not the volume from the beginning. Most important are the actual implementers.

<npdoty> dsinger, I think we prefer more than 5 :) although I think the preference is to have 10 to 15 people

dwainberg: Are you saying the advertisers keeping this alive?
... Would our leaving have the impact of terminating the WG?

<WileyS> If less than 10 people on today's participants raise their hand to suggest they want to continue work on the TCS, does that give the co-chairs a meaningful signal?

<WileyS> As none of the co-chairs represent industry, we'll continue to see this disconnect.

justin: The intention from the beginning was to work on a standard for a DNT signal. Has this intention changed?

dwainberg: Wendy said one of the factors keeping this alive is our participation. It's smaller than ever, mostly just me. It may make sense to me to drop off the group

<dsinger> both participation and implementation are voluntary...

<Chris_IAB> npdoty, the advertising industry has been railroaded at every step in the development of these specs

npdoty: Sorry to hear that people feel that they have no choice but to be here. It is a voluntary standard. We had and need a variety of stakeholders. I was optimistic that we have sufficient interest from the industry to do the work.

<Brooks> I think we disagree on the meaning of "compelled to be here"

justin: q?

<WileyS> Not a poll - a "hum"

<Chris_IAB> let's not conflate interest with agreement with the direction the chairs and staff have taken this initiative.

<WileyS> Many things have changed in the past 6-7 months

justin: We will not ask for a new poll on motivation. We did this in October. We will still follow this decision.

<WileyS> Seeing that the co-chairs are not taking working group feedback, its difficult to see a fair path forward.

<dwainberg> All in favor of not continuing work on the TCS?

<dwainberg> +1

<WileyS> +1

<Chris_IAB> I'm hearing from Justin, that it's not about critical mass... "we are moving forward" (period)

justin: We will do a better job to put up the agenda earlier next week. I will send out a note to the group about the D signal.

<Chris_IAB> +1

<Brooks> +1

<Chris_IAB> so to be clear, the call was cut short

<vincent> thx

<Chris_IAB> by the Chair

justin: If you want to keep discussing going forward, please do so on the mailing list. Will keep the call focused.
... bye

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2014-04-16 16:46:37 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.138  of Date: 2013-04-25 13:59:11  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/designers/dwainberg's/
Succeeded: s/like/likely/
Found ScribeNick: ninja
Inferring Scribes: ninja
Default Present: Ninja, Wendy, ChrisIAB, hefferjr, dsinger, Jack_Hobaugh, RichardWeaver, Peder_Magee, rvaneijk, dwainberg, npdoty, justin, WileyS, +1.323.253.aaaa, Chris_Pedigo, Ari, schunter, MattHayes, [FTC], Brooks, Chris_IAB, [Microsoft], +1.202.370.aabb, robsherman, vinay, WaltMichel, +1.201.530.aacc, hwest, moneill2, Fielding, +1.312.923.aadd, mecallahan, vincent, adrianba
Present: Ninja Wendy ChrisIAB hefferjr dsinger Jack_Hobaugh RichardWeaver Peder_Magee rvaneijk dwainberg npdoty justin WileyS +1.323.253.aaaa Chris_Pedigo Ari schunter MattHayes [FTC] Brooks Chris_IAB [Microsoft] +1.202.370.aabb robsherman vinay WaltMichel +1.201.530.aacc hwest moneill2 Fielding +1.312.923.aadd mecallahan vincent adrianba
Found Date: 16 Apr 2014
Guessing minutes URL: http://www.w3.org/2014/04/16-dnt-minutes.html
People with action items: 

[End of scribe.perl diagnostic output]