Tracking Protection Working Group Teleconference -- 30 Oct 2013

<trackbot> Date: 30 October 2013

<WaltervH> Good afternoon Carl

<wseltzer> scribenick: wseltzer

<Chris_IAB> just joined the call

-> http://lists.w3.org/Archives/Public/public-tracking/2013Oct/0349.html Agenda

<JC> Let me do first 30 mins

<npdoty> scribenick: JC

Carl: we are now moving to agenda item 3

Update on evolution of TPWG plan

<FPFJoeN> 202-587 is FPFJoeN

<moneill2> who let the dogs out

Alan: I have a question
... It seems like we have moved from working on TPE and more like options 1 & 2

<fielding> the other terms are used in the definition of tracking and in the TSR

Alan: can you explain the shift?

Carl: We had issues 1 & 2 and 3 &4

<WaltervH> you mean options?

<Brooks> not sure that was 3 &4?!

<jeff> I believe Carl means Options

Carl: the chairs and staff felt that we want to get something out that is useful

<jeff> Option 3 was TPE first

<jeff> Option 4 was TPE only

Carl: we also want to include the sense of the group that included some parts of compliance
... we want to have a TPE that is implementable and not just a vacuous spec

Alan: Is it the chairs position that working on the TPE with a tracking definition was not workable?

Carl: That was not discussed in the fianal options
... we thought about which definitions would be necessary for the TPE to be useful
... unfortnately we felt we would know it when we got there

Matthias: We felt it would be weird to talk about tracking all the time without knowing what it is
... we decided to keep some defnitions from compliance to make the TPE self-contained
... there are several places in the TPE that rely on compliance definitions and we want to remove those dependencies

Heather: Ian was expressing that we should make sure we are creating something simple enough to implement as a 1.0

Chris: I see what you are trying to do, but I don't see where you draw the line
... if you add the definitions into the TPE then you are really merging it with the compliance spec

<Chapell> This is the decision of the chairs and w3C staff, not necessarily the intent of the working group.

Chris: I feel that will unnecessarily hamstring orgs that want to do compliance in different ways

<moneill2> or the server

Chris: I would prefer that we point to a compliance spec and the browser can point to it

Browser or website?

<fielding> Can we please phrase these comments as suggestions to the editor on what to include in a first attempt, rather than stop action based on unknown concerns?

Carl: In general, that is what we are trying to do

<WaltervH> fielding: +1

Carl: Roy point of view is that we are putting content in that will make the TPE a viable spec

Matthias: Our goal is not to create a TPE that will work with arbitrary compliance specs

<npdoty> I think it may actually be a separate question about whether the documents themselves are combined, and more prioritizing issues for Last Call for implementation and review of TPE first

Matthias: we plan to continue work on Compliance spec after completion of TPE

<vinay> How will companies test the TPE without a compliance spec / compliance regime?

Chris: I'm concerned about the breadth of the TPE

<WileyS> Matthias - I disagree as then we end up not separating the issues with the compliance spec. By drawing in the defintions into the TPE you're completely missing the point of Option 3

Brooks: I am going to agree with others

<moneill2> +1

Brooks: if service provider becomes a definition in the TPE then it becomes more like the compliance spec
... I understand that we need to define some things, but it may make it impossible without defining the compliance doc

<dwainberg> i got disconnected

Alan: What I am hearing is that the TPE cannot be completed without X number of terms.

<Brooks> +1

<WileyS> The point of Option 3 in the Poll was to completely separate the TPE from the Compliance Spec and allow the TPE moving forward without those issues in contention.

Alan: Is that was the assumption why wasn't that an option on the poll

<dwainberg> thanks, Wendy

<Chris_IAB> dwainberg is back

<Brooks> If TPE only wasn't a viable option, it shouldn'

Carl: That sort of developed after a two-hour discussion. It didn't seem like a poll able question at the time.

<Brooks> t have been listed in the poll

<WileyS> Carl - you're hearing the Working Group disagrees with your "consensus" between co-chairs

Carl: we are trying to adopt what the group has said and we feel we need to move some definitions over to have a viable spec
... we don't want to have a meaningless TPE

<WileyS> Carl - Who determines "valid", "neutered", or "incomplete" in this context?

Carl: at some level we are winging it. We want to find a way forward that works.
... we are trying to get through this as quickly as we can

<npdoty> the options for the poll itself were briefly described, but comments in poll responses and in our call dedicated to discussing the poll, the group discussed issues to prioritize for TPE to go to Last Call

David: Firstly, I want to echo Ian's comments from the list
... removing parts of TPE that are not useful would be good
... the sensible approach would be to remove parts of TPE that don't stand on their own
... then we should look at which definitions that we need
... We are continuing on the same path to complete the compliance doc and we are going to debate defnitions that we have for years
... I don't think that will work and produce an unaccpetable outcome
... I believe the will of the group is to focous on the TPE, removing unneeded items
... At the end we can look at definintions needed

Roy: We have gone down the path of not defining things and that can't continue

<ninja> dwainberg, how can the TPE spec be an entirely new context?

Roy: we can't continue functioning without them
... I need to know what I am writing. So give me a chance to include things to define the protocol
... After the first draft we can remove unneded items.
... we don't need to decide today. Let me finish the document first

Walter: I support the idea of completing the TPE first, but we don't have to have all definintions
... industry hammered on the need for a tracking definition
... now that we won't have a compliance spec then we must have a tracking definition
... I don't agree with supporting a number of compliance specs
... I can see an industry spec having more permitted uses
... I don't think you can have a sensible TPE if there are different expectations of what tracking means

Matthias: I want to emphasize that we don't want to create a Compliance spec through the back door
... we can't create a TPE with empty spaces

<WaltervH> JC: I actually support having multiple compliance specs

Matthias: we can build a protocal with all the pieces well defined.

Walter: I actually support multiple compliance specs

Mike: I see David's points.
... We have had the expectation that we will have a TPE and Compliance spec
... I think we should start by defining tracking
... if that is going to be in the TPE then we should recognize that this is a new process
... what we have now from today forward is potentially a different document from the original TPE
... my discomformt comes from truncated deadlines.
... the change in process and scope should be recognized

I need to drop off

!!!!!!!

<npdoty> scribenick: npdoty

<susanisrael> npdoty, do you want me to scribe for a bit

<dwainberg> but that's compliance

<WaltervH> ninja: +1!

<Chris_IAB> we have that today and people are saying it works fine (I agree that assertion that it works is questionable)

<dwainberg> what I heard there is that we need a mini compliance spec in the standalone TPE

<dwainberg> I disagree

<ninja> npdoty, sure I write it down myself

<susanisrael> npdoty, i will scribe after i speak

<jeff> Mike, fwiw, I'm not sure what the alternative is to "truncated deadlines". The WG has spent weeks debating Issue 5? what approach would you take to resolve it?

<Chris_IAB> browsers are implementing today without any agreed upon definitions

<wseltzer> scribenick: wseltzer

<Chris_IAB> so the notion that we can't test without definitions seems silly

<rvaneijk> +1 Nick

npdoty: poll option 3, do TPE first, is roughly what the chairs are suggesting

<WaltervH> Chris_IAB: that is not the notion

<Chris_IAB> WaltervH, that's one thing I have heard today

npdoty: input from the group seemed to be TPE-first, so going through necessary subset of issues

<Chris_IAB> but please, correct me if I am wrong

susanisrael: if we incorporate definition of tracking in order to set user expectations, we have to be clear that there will always be some of what a user might consider "tracking"

<WileyS> If we feel definitions follow a compliance regime and we offer up the ability to map to different compliance regimes in the TPE then the definitions are tied to what the Server replies with that its compliance regime is.

susanisrael: e.g. security and fraud. Need to be clear to refer to permitted uses so as not to mislead

<susanisrael> npdoty: I can scribe

<WileyS> +q

<npdoty> scribenick: susanisrael

<WileyS> Apologies - thought I was already in queue

cargill: not sure i can summarize all i have heard...

<ninja> I am surprised about the counter arguments on including definitions in the TPE spec. In my view the TPE spec without any basic definitions or semantics on what the DNT:0 and DNT:1 signals shall mean in regards to tracking is pretty useless. We should not put even more confusion upon the user to choose their compliance regime to define what tracking in their signal does mean.

<Chris_IAB> my point is, we have a test of the TPE today, in market, without definitions in a TPE... so why can't we study that, finish up the TPE, and THEN work on the compliance spec (based on learnings)

wileys: trying to reconcile roy's view that definitions are required, and the idea that bringing in definitions are a back door to bring in compliance
... propose variable definitions, that would support multiple compliance regimes

<npdoty> we often have early implementations even before we have a draft; Last Call is a milestone to allow more implementation testing

wileys: this is a path forward to meet in the middle

<WileyS> Ninja, confusion already exists for users.

jeff: this notion of being open to alternate compliance regimes is already part of the chairs' plan. notion is to discuss/debate that in the next few weeks. W3C believes we need a unified standard....

chairs agree with this but we also point out that we could include a field that would indicate compliance with an alternative regime. Need to consider that kind of flexibility

<WileyS> HL7 (Health Level 7) - the core medical standards system allows for variable definitions. Too bad the W3C couldn't figure it out by many others have

<ninja> WileyS, if this way forward leads to competing compliance regimes, at least the scope of what the users are signaling should be clear. And this relies on a clear definition of tracking.

cargill: story based on standards experienced. look at Open Systems Interconnect (OSI).[ didn't hear] example of a complex spec that wasn't adopted. we don't want to waste time and money on something that doesn't work
... this is either a radical departure or more of same....

<Chris_IAB> WileyS, I think your idea is worth pursuing further to see if it could work

<WaltervH> WileyS: most health professionals I know start foaming at the mouth as soon as you bring up HL7

cargill: I think matthias has first item in TPE, issue 5

Issue-5

cargill: I told people we now have 8 or more proposals, highlight one to go to consensus with....

<Chris_IAB> WAIT, so that's it? We are moving forward as stated in Carl's email yesterday? Done, period, that's it?

<Brooks> It is difficult to know how meaningful debate over alternatives will be when one of those alternatives has already been definitely closed by comments by the chairs

<WileyS> Walter - it is admittedly a beast but that's where I cut my standards teeth nearly 2 decades ago and while its riddled with issues its an awesome standard that allows 100s of thousands of systems to interoperate.

so we had one, but now 2.....

<JackHobaugh> Matthias, please also add proposal 2 to the list for Issue - 5

<WaltervH> WileyS: well, that bit of info explains a lot ;-)

<wseltzer> ISSUE-5?http://www.w3.org/wiki/Privacy/TPWG/Change_Proposal_Tracking_Definition

scribe: from my perspective, I asked Dsinger if we could make roy's defintion good enough to satisfy his requirements. david is traveling, but if he sees way to reconcile, I will send around his proposal to reconcile...

<WileyS> All large, complex standards have issues - any lack of understanding of this explains alot as well

<JackHobaugh> +q

<npdoty> JackHobaugh, to confirm, you'd like to consider the No definition option for the definition of tracking issue?

<jeff> Chris, I would say to first order you are correct, that we are proceeding with the plan; but Carl also said that the Chairs would review the input at the Chairs call today.

<WaltervH> WileyS: let's aim for this standard not to be large and complex, shall we?

scribe: if david and roy are unable to merge their definitions we will go to CFO with 2 definitions......will tell announce list what we do

<WileyS> +1 to Jack

jackhobaugh: I would like to see proposal 2, no definition of tracking kept as a proposal

<WileyS> Matthias - we did say that on the list!!!

matthias: why didn't you say so on list?

jackhobaugh: may have missed it apologies...

<johnsimpson> I thought all proposals on the wiki were under consideration

matthias: i think jack is right, but want to caution everyone to look at email, especially mine.....
... will go to CFO with 2 or 3 definitions.

cargill: cfo due when?

matthias: 2 weeks. npdoty pls edit wiki

<npdoty> note to nick: add No Definition to proposals list

<fielding> npdoty, and candidate B should have a different heading title

npdoty: i want to check something that is implicit....there were a few definitions in wiki that seem similar--3,4, and 7....

matthias: i asked if they insist on 5 different flavors of same thing, so I now want to have 3 options, no definitions, a, and b...and they did not object.

<wseltzer> so those have been consolidated to Candidates A and B

rvaneijk: i was in a side conversation but followed conversation between david and mike and understood that my text of non-normative language made sense. Can we include in proposal b?

<npdoty> I think using the single option (of 3, 4, 7) is good, just wanted to check while some of us were on the phone that we're set on that

matthias: would like to focus on normative text first, then can turn to non-normative text to explain and illustrate

<fielding> no, that non-normative text has very little to do with tracking

rvaneijk: defintion a, last lines, includes an assumption that could be read in a non-normative way. Maybe that should be removed.

matthias: good point that there is non normative language in a, I will look at it and come back

<fielding> That is a list of dependencies to explain to folks here -- the definition is what we are proposing.

matthias: i think i also got a minor edit from moneill

moneill: i said off-list that we say assume, instead of recognize

<npdoty> if dsinger and moneill are agreed, I'm changing recognizes/assumes while I have the wiki edit page open

moneill: there has been a bit of discussion between david me and rob that might be handled in non-normative text

<fielding> The dependencies are dependencies on understanding the definition, which means that if the assumption becomes false then it is new information to reconsider the definition.

cargill: did you just roll your proposal into option b?

matthias: i think mike does not have a separate proposal, just an edit
... any problem with augmenting option b with rob's non-normative text

dwainberg: it's impossible to keep track of conversations. so is non-normative text part of definition?

matthias: will discuss with david and rob

<robsherman> +q

dwainberg: i object to the additional text. It's a mini-statement of compliance. It's not relevant to the definition of tracking.

cargill: can you bring that up in cfo

<Chris_IAB> Can we let David finish?

<moneill2> +q

<Chapell> +1 to dwainberg - I might support that definition, minus the additional text

dwainberg: but we're narrowing choices, and none may be satisfactory. I don't know how this happened. I thought we were discussing a list of definitions to be negotiated. Not sure how we got down to 2.

matthias: so a week or so ago i asked people which definitions of the 8 or 9 should be nominated for the CFO, then I created a short list based on nominations by the group

<fielding> http://www.w3.org/wiki/Privacy/TPWG/Change_Proposal_Tracking_Definition

<Chapell> If we want to add RVE's language to that definition as a SEPERATE proposal, that makes more sense. As the additional text materially changes the definition

matthias: david, roy and now Jack have nominated definitions.

<npdoty> I think we currently have 3 candidates, which represent the major options in the longer list of proposals (with a few merged)

dwainberg: are we editing these proposals as we move them into the short list?

fielding: yes, if you don't like them propose one of your own

<Chris_IAB> +1 to deainberg 's point

matthias: yes, or edit.....
... would like to find a good definition that covers a large segment of the group
... don't want to have to consider defintions that might cover only 10 % of defintions

dwainberg: i can't keep up with the process changes. We keep moving goalposts. And my company is one that will be really affected by the changes....

<Chris_IAB> you can fix it by slowing down, and being clear about process

<jeff> [Jeff suggests to Matthias that we be willing to go to CfO with more options if that helps. Perhaps we cannot achieve narrowing in this case.]

cargill: can't fix that. you are asking us to fix a problem that you have, common across all standards groups. no one has unlimited time.

dwainberg: i am trying to contribute and be productive. Could we focus on one thing at a time, therre are multiple threads going on at once. Could we stop everything else if we are trying to define tracking:?

<fielding> http://lists.w3.org/Archives/Public/public-tracking/2013Oct/

<Chapell> +1 to Jeff's approach. otherwise, we run the risk of having a last minute amendment that completely changes an otherwise acceptable definition

<Chris_IAB> agree with dwainberg that we should dedicate time to core topics and not try to shove too much down the pipe just to finish

dwainberg: let's have a thorough discussion and make sure we get a good result. I can't follow 12 threads of process going on at the same time.

cargill: all have that problem. Trying to do that right now.

<Chris_IAB> is Carl not on IRC?

dwainberg: but we are rushing it...

<wseltzer> Chris_IAB, I believe not

<Brooks> More importantly, if we thought we were choosing between 5 paths then don't go forward with plan #6. No one has time to make those adjustments.

jeff: i think part of the problem and confusion is that we have been attempting to get cfo to small number of proposals but it sounds like on this issue there are nuances that are important to people....

<Chris_IAB> wseltzer, Jeff, how is it acceptable that we have a Chair who is consistently not on the IRC and following the discussion there??

<WaltervH> Chris_IAB: my understanding is that Carl is indeed not on IRC. I don't really understand why not.

so on this if the working group wants to have 7 or 8 proposals, let's do that even if it's more work for the group. People are not satisfied with pseudo-consensus....

<fielding> I agree with Jeff -- a call for objections does not limit anyone to a single choice

<jeff> Chris, that is a fair point that I will raise offline with the Chairs.

<Chris_IAB> WaltervH, that seems absolutely nutso to me

matthias: will discuss in chair's call and I will come back to issue 10. Queue first.

<WaltervH> Chris_IAB: my preferred term would be 'undesirable'

robsherman: I think part of challenge is that there are many side discussions, and people are having trouble understanding the impact of proposals, so besides jeff's objectiions, let's make sure...

<Chris_IAB> Jeff, Carl is acting the roll of "enforcer" on these calls, while we all choke it down, and he's only informed by PART of the entire conversation (by not being on IRC)

<Chris_IAB> WaltervH, fair enough-- but I think it's nuts

robsherman: people know what's on the table and what will be subject of call for objections....i see new things in the 2 proposals that I have not yet been able to fully appreciate the impact of. Need to give people time (a week?) to think through it.

<Mike_Zaneis> +1 to Rob about process

<npdoty> that Roy's definition uses "contexts" has been in discussion for the past month, right?

matthias: good point. But I think i said ...

sisrael: +1 to rob

matthias: my challenge is to know when to call "stop" for objections

<WaltervH> npdoty: I do recall the term contexts being used over a year ago during various conversations

correction for mathias--need to know when to call stop for discussion.....

<npdoty> WaltervH, yes, "contexts" has been a part of discussion for years, I was just referring to this definition and text in particular

<rvaneijk> +1 Mike

moneill: i think it's a good exercise to have a small number of things to focus on. I don't think we should open it up and have loads of choices. Need to make sure it gets right attention.

matthias: so you are saying don't kill this process by going straight to cfo

<ninja> +1 to Mike

moneill: yes, let's have a few more days because we're getting somewhere

matthias: so having a few worked out alternatives to choose from?

<wseltzer> http://www.w3.org/wiki/Privacy/TPWG/Change_Proposal_Party_Definitions

issue-10

matthias: next issue is issue 10. with issue 10, there were a bunch of maintenance changes that I would like to serve as a starting point for this discussion.
... would like to have comments on the text

<JackHobaugh> Regarding Process: Everything that was done over the last three weeks was done without the TPWG knowing what path it was on. And from the poll results, the path appeared to be confined to Options 3 or 4. It is reasonable to expect that during the last three weeks, the conversation was biased by a focus on options 3 and 4. It was only yesterday that Carl announced the path forward. And the path forward now is neither Option 3 or 4 but instead appears to be a

<JackHobaugh> mixture of Options 1 and 3 with perhaps even option 2 mixed in.

dwainberg: why are we defining party?

<Chris_IAB> +1 to JackHobaugh

dwainberg: may not be a dependency for the TPE.

<jeff_> Jack, in my view, the Chairs' path is closest to Option3. Once they decided to do TPE first, they decided that some items needed to be added to TPE to make it self-contained.

<Chris_IAB> Jeff, no way is that Option 3.

matthias: we can indicate where things are used in first and third party context, so still need idea of first and third parties.

<jeff_> Chris, well it is certainly (a variation of) TPE first.

dwainberg: may need to define those contexts, but to ian's point, there are things in here that are specific to a particular compliance approach.

<Chapell> jeff, you may see it as option 3, but please understand that many wg members believe that we're significantly closer to options 1 & 2

<Chris_IAB> Jeff, it's something that wasn't on the poll, and we are just being told, this is how we will move now... I don't think you have consensus on this NEW "plan forward"-- sorry

matthias: we have to define third party for compliance coming later.

dwainberg: but we don't need it for the tpe. and we may not need it for tpe. May instead focus on context and not need to define parties, saving some time.

matthias: other inputs

<npdoty> TPE refers to parties -- communicating a tracking status has "1" and "3", for example

<Chris_IAB> Jeff, it is not a variation of TPE first... it is a MERGER of TPE and some parts of the Compliance spec, which is not one, then the other.

<jeff_> Chris, it is clear that no single proposal will have full agreement. The Chairs' view is that this is closest to a consensus of all the input they received.

waltervH: i can see david's reasoning.

<Chapell> is there a link to the text that we're currently discussing? Issue 10?

<npdoty> issue-10?

<trackbot> issue-10 -- What is a first party? -- open

<trackbot> http://www.w3.org/2011/tracking-protection/track/issues/10

<npdoty> http://www.w3.org/wiki/Privacy/TPWG/Change_Proposal_Party_Definitions

<dwainberg> alan, Matthias sent an email this morning

<npdoty> http://www.w3.org/mid/5271032A.4080906@schunter.org

waltervH: actually there are [?] mechanisms for party to share in TPE . May still need to retain references for parties for both tpe and compliance spec to have a record of which parties/entities have data.
... and what can be expected from them.

matthias: any comments on the text I sent

<moneill2> +q

matthias: heard from david we should postpone to compliance spec but is there actually any comment on the text?

<JackHobaugh> Respectfully Jeff, it is not TPE first, it is "prioritize getting the TPE out to last call." This language gives the chairs room to continue, as they have, with issues against the compliance spec.

<jeff_> Jack, I agree that TPE first (Option 3) means getting TPE to Last Call, first.

ninja: "Party...." [reads schunter def]...i think we had a discussion about whether we need to have a common privacy regime as well as common owner

<WaltervH> ninja: you broke up a lot

i don't think we had an ongoing discussion about a common privacy regime. that was merely part of an alternative proposed by dwainberg

ninja: i would require this from all parties

<npdoty> JackHobaugh, jeff, Option 3 in the poll specified that each specification could go to Last Call at its own pace, and there was support for TPE moving to Last Call earlier

matthias: so if 2 affiliates do not have common privacy regime they would be 2 parties?

ninja: yes

moneill: are we talking about email about service provider? or what's on wiki?

<Chris_IAB> npdoty, but nothing in the poll called for moving certain compliance bits into the TPE

moneill: need to define service provider if it becomes part of definition of parties.

<npdoty> ninja, is there a text proposal for that? there might be some similarities in Alan's proposal (about multiple companies acting as a single party if they have consistent privacy language)

matthias: what i sent around by email. I think it was part of third party definitions

<johnsimpson> Very difficult to discuss text that was received at 6 am PT...

<ninja> Sorry, phone bridge is bad today. Wanted to bring up again the issue whether a common party needs to have a common owner PLUS a common privacy regime.

<WileyS> Ninja - we support common ownership and common (exactly the same) privacy regime

moneill: problem is that this is too vague unless there is contract between user and service provider

<Brooks> again scope creep. we need more and more definitions in the TPE

<Chris_IAB> npdoty, but I suppose we understand why, per Carl's submission earlier today, "We are just winging it" (no kidding!)

moneill, I don't think it's reasonable to suggest that user would have to have contract with service provider. That would NEVER happen but there are service providers, who have a distinct status

matthias: roy do you have an opinion on use of service provider in third party defintion

<schunter> Third Party (from (5):) For any data collected as a result of one or more network interactions resulting from a user's action, a third party is any party other than that user, a first party for that user action, or a service provider acting on behalf of either that user or that first party

fielding: we have an issue for service provider definition, not sure what we're discussing or why

matthias: just put in irc. mike raises point that we need definition, and you are right that we have an issue, so if we address that it should be settled.

fielding: clearly the intent is that this should be a defined term.

matthias: so i am noting that some terms still need defining, like service provider.
... will see if people are ok with this definition or no definition is preferable, but so far no one has done so.

<johnsimpson> How does this to be a candidate when we've had three hours to look at before the call??

matthias: ending discussion on issue 10. Next is issue 16.

Issue 16

<WaltervH> rigo: I think W3C shoult look into using AT4AM for processes like this

<JackHobaugh> To be clear the title for option three is: "Finalise TPE first and then finalise Compliance" and the clarification wording is: "While each specification could go to Last Call at its own pace, Candidate Recommendations for both documents would be published jointly, and the two would proceed further together." It is clear the intent of option 3 was to complete the TPE first and then the Compliance doc. It is only once we get to CR that both would continue in

<fielding> issue-16?

<trackbot> issue-16 -- What does it mean to collect, retain, use and share data? -- open

<trackbot> http://www.w3.org/2011/tracking-protection/track/issues/16

<JackHobaugh> parallel.

<schunter> You will have more time. Note that the text was a consolidation of texts on the wiki for issue 10 that have been there for a while.

cargill: let me take 16 quickly, since there has been comment. this is one where we are close to closure. I would like to take time to introduce parts of it. we are dealing with what it means to collect....

<schunter> I did not create new text... (at least not conciously)

<fielding> http://www.w3.org/wiki/Privacy/TPWG/Change_Proposal_Transience_Collection

cargill: retain and share data. there has been a lot on list, and I'd like to introduce that. Vinay can you discuss?

<Chris_IAB> Jeff_, Chairs, I think dwainberg raised a very fair point today: there are some folks on these calls from industry that have limited time, due to their day jobs (running their business), to focus full time on this spec, and yet, those people are arguably most affected by these specs. The group should adjust for that reality, so it is not construed as being one-sided and railroading the affected.

vinay: david, lee and i have discussed and are near merging, 3 change proposals for definition of collect. Not sure whether still wants to keep his for retain. I think lee has one for uses.
... my question from chairs is whether we still need to worry about old change proposal from jmayer.

cargill: is it incorporated in any of these other proposals?

<npdoty> is jmayer's text significantly different anyway? it looks very similar to Lee's and Vinay's.

<kulick> looks like Lee Tien's proposal captures it

cargill: has anyone looked at it seriously?

<WaltervH> kulick: the editor's draft is pretty good too

<Chris_IAB> Jeff_, Chairs, it should also be noted that for others, this is their full-time job, and thus they have the time necessary to influence the spec work to satisfy their positions and constituencies.

vinay: i personally think it is included in the proposal we are considering.

<WaltervH> kulick: there's some parts of Lee's definition that are better and the other way around

lee: yes, I agree with vinay about where we are. I do not recall what jmayer proposal vinay is talking about. I can't be sure.

cargill: so could you, vinay and david incorporate what we think was the intent of jmayer's proposal.

lee: sounds do-able.

<Chris_IAB> Vinay, which Chris?

<fielding> We need to update the wiki so that the proposals represent complete replacements for the existing definitions -- it is too hard to read them as is

vinay: yes, we can do that. some work still needs to be done re: definition of share. I owe chris a response. will also work with amy who has a defintion of share. ....

<kulick> WalterH: I think that the editor's draft is a bit different than Mayer's proposal in a meaningful way

<vinay> Pedigo

<vinay> sorry for not clarifying which chris

vinay: i think there are 2 schools of thought here, will try to address.

<WaltervH> kulick: I like Lee's definition of collection better, I like the editor's draft on sharing better, although Vinay's may be best actually

<Chris_IAB> Great, can folks please say "Chris P" or "Chris M" so we know which Chris you are referring to? :)

cargill: will be moving into consensus mode on this. Please monitor the message traffic

<npdoty> vinay, to fielding's point, we could use the consolidated updates on the wiki

johnsimpson: that message traffic, has that been on list ?

lee: no, among us

<jeff_> Chris, I'm trying to figure out how to balance industry's critique that we are moving too slowly with your critique that we are moving too quickly.

vinay: i plan to send out once i am sure i am speaking for lee and david

<jeff_> Chris, can you help me push back on those in industry who say that they will go elsewhere because we are moving too slowly?

cargill: initial request last week to vinay was to work with lee and david on this. Now ask them to put back on list.
... trying to be sensitive to number of messages.

<Chris_IAB> Jeff_, what critique from industry are you referring to, that we are moving too slowly? That hasn't come from us...

cargill: at the same time i would like to present something more complete.

<Chapell> jeff, the issue isn't that w3c is moving quickly (or not). Rather, the issue is whether you are moving logically, methodically, and in a way that limits co-chairs and w3c staff from imposing their own consensus upon the working group

<vinay> Sure npdoty. I may need some help doing that, but I'll see if I can figure out how to update the Wiki

cargill: please do monitor

<npdoty> vinay, yes, I can definitely help

vinay: I will send out definition of collect once off call, then will try to work with david singer, chris p and amy to try to find consensus.

<Chris_IAB> Jeff_, you probably shouldn't conflate "moving too slowly" with "moving in the wrong directions" or "moving without consistent direction" -- I have led spec work for industry that's taken 2-years, but it was always moving in a positive direction -- it just took time, because everyone has other "day jobs" that they report to first

<jeff_> Chris http://adage.com/article/privacy-and-regulation/ad-industry-ditches-track-group/244200/

<jeff_> Chris, "If you measure it by progress it is dead"

matthias: by next week we are likely to come back with finalized plan

<Chris_IAB> Jeff_, wrong context

- DRAFT -

Tracking Protection Working Group Teleconference

30 Oct 2013

Attendees

Contents

Update on evolution of TPWG plan

Issue-5

issue-10

Issue 16

Summary of Action Items

Scribe.perl diagnostic output