Tracking Protection Working Group Teleconference -- 12 Mar 2014

<trackbot> Date: 12 March 2014

<Chris_IAB> just joined the conf bridge from NYC

<walter> ninja: have a flu, can't call in today

<Chris_IAB> npdoty, super

Confirmation of scribe. Volunteers welcome!

<mecallahan> i have to come in and out, i dont thinki can scribe

<npdoty> scribenick: mecallahan

ISSUE-240: Do we need to define context?

justin: thanks.
... chairs meeting face to face next week.
... working to get TPE to last call, adn start to think about compliance document.

brookman: re: last call for TPE.

<ninja> ISSUE-240: https://www.w3.org/2002/09/wbs/49311/tpwg-context-240/results

<trackbot> Notes added to ISSUE-240 Do we need to define context?.

brookman: brookman posting document.

<npdoty> https://www.w3.org/2002/09/wbs/49311/tpwg-context-240/results
....brookman: in the end, Roy mdoified his definition substantially, close to definition to party. there were mulitple options. adopted Roy's definition of tracking.

<vinay> is anyone else having problems dialing-in?
....brookman: with regard to context definition, the objections remained strong on the objections on PARTY.

<fielding> I think Justin meant Roy's definition of tracking

<vinay> i'll call from my cell phone

brookman felt the Option A and Option C were similar, but they decided to define context in the same way.

<ninja> Option C (with a small editorial change): "A context is a set of resources that are controlled by the same party or jointly controlled by a set of parties."

scribe: Chairs will issue written opinion shortly.

wainberg asked clarification, brookman described process. ninja spoke on behalf of matthias, matthias's editorial change shown above at 12;07.

matthias had run this editorial change past rob sherman and chris pedigo, and both were ok with it.

<fielding> That is a reasonable editorial change.

brookman to send context editorial change to the list.

<justin> https://www.w3.org/2002/09/wbs/49311/tpwg-qualifiers-241/results

ISSUE-241: Distinguish elements for site-internal use and elements that can be re-used by others (1/3)

wainberg asked about written opinions on 3 calls for objectiion; brookman said that was forthcoming.

wainberg asked about optionD, whether that was considered. Brookman said that would be addressed in the written opinion.

<WileyS> If the editorial change is met with objections - does that delay the Last Call while the CfO is processed?

<fielding> WileyS, in general, objections do not delay a last call -- they merely have to be noted as such.

Issue 241: written opinion forthcoming. objection to including this probably stronger. down the road, may make sense to put back in.

<WileyS> Thank you Roy

Proposed editorial changes to the TPE before Last Call

brookman: TIMING -- communicated results of hte call for objections to the editors earlier.

<JackHobaugh> So which option was chosen for Issue-241?

brookman: editors want a few weeks to implement the changes, and implement the changes.

<fielding> action on fielding to incorporate changes for defn of context

<trackbot> Error finding 'on'. You can review and register nicknames at <http://www.w3.org/2011/tracking-protection/track/users>.

brookman: expect March 26 date for the editors.

<npdoty> JackHobaugh, justin indicated less strong objection to Option B: no change

<fielding> action fielding to incorporate changes for defn of context

<trackbot> Created ACTION-437 - Incorporate changes for defn of context [on Roy Fielding - due 2014-03-19].

chris_IAB: what option for 241? Brookman: no change.

<fielding> ACTION-437 due 2014-03-26

<trackbot> Set ACTION-437 Incorporate changes for defn of context due date to 2014-03-26.

on march 26, chairs will share document with the group and then vote to bring to last call.

brookman: anything else on the timing?
... a few editorial changes have been suggested.

<justin> http://lists.w3.org/Archives/Public/public-tracking/2014Mar/0017.html

<rvaneijk> Ninja, could W3C please confirm the TPE planning on the mailinglist?

npdoty: editorial changes: track status.

<fielding> action fielding to remove issue box for qualifiers

<trackbot> Created ACTION-438 - Remove issue box for qualifiers [on Roy Fielding - due 2014-03-19].

<ninja> rvaneijk, good suggestion. We will do so. Although it's not set in stone.

<fielding> action fielding to find a media type for the tracking status representation

<trackbot> Created ACTION-439 - Find a media type for the tracking status representation [on Roy Fielding - due 2014-03-19].

<fielding> mine

<npdoty> I'm sure we won't object, but if fielding can share his conclusion with the mailing list, that'd be great

npdoty: editorial change. suggestion that we describe exceptions API, user provided a special permission, change from DNT0 to DNT1. may want to give permission for particular sites.

editorial change cont: suggestion to call it "permission" for that function names and section title names.

<moneill2> +1

<moneill2> +q

<npdoty> Mike O'Neill raised this during Compliance discussion: http://www.w3.org/2011/tracking-protection/track/issues/212

<WileyS> No concern with UGE -> UGP

brookman: dsinger noted permission is used differently in the compliance document, need to reconcile the definitions.

<justin> thanks wileys!

<npdoty> +1 to dsinger that we would want to change the language in compliance to match up

moneill notes there already is a permission API on the working group. suggested using "tracking permissions" as a phrase.

brookman: yes, need to revise compliance document.
... last written opinion: R flag.

<npdoty> MSFT has implemented it early/already, so we should check with them in case there's been a lot of uptake already

ninja question: since matthias did most of the dicussion, maybe wait for matthias?

brookman: yes, lets wait on this issue for matthias.

jack: offers to take questions, thinks the proposal is fairly clear.

brookman to jack: what is it supposed to do?

jack: takes out ambiguitiy of T flag.
... doesnt say what is really happening at the server, since that is the complianhce document/adjaceent format.

brookman: but that functionality already in the response.

<npdoty> was just going to repeat that point, that the compliance field in the same tracking status resource does indicate that

npdoty -- FYI mecallahan has to leave in 5 minutes.

jack: more precise to have a document that refers to the response. need to be more precise/definitive.

jack this R flag is more precise.

brookman: how does R flag help?

<fielding> T is defined with more preciseness than R, so I don't understand that.

jack: since we havent implemented yet, jack is anticipating how to comply.
... doesnt think T flag will slow things down, and would unaccpetable for a user.

<npdoty> loading the tracking status resource will be uncommon, and won't ever block the loading of the page we expect

brookman: lets park this discussion until matthias joins.

npdoty -- you are up!

<npdoty> scribenick: npdoty

ISSUE-181: Finalize language regarding multiple first parties

justin: take up work on compliance, have done a lot of work on compliance in the past few years

<ninja> https://www.w3.org/2011/tracking-protection/track/issues/181

justin: have heard from group members that we do want to take these up, some issues we might be quite close on
... get started with a couple issues that might not be particularly difficult
... issue-181 on finalizing multiple first parties

<ninja> text in the party definition: “In some cases, a resource on the Web will be jointly controlled by two or more distinct parties. Each of those parties is considered a first party if a user would reasonably expect to communicate with all of them when accessing that resource. For example, prominent co-branding on the resource might lead a user to expect that multiple parties are responsible for the content or functionality.”

justin: I think we might be done with this issue now. would call on robsherman to see if this is resolved

robsherman: agree that that was a legacy issue, fine with closing the issue now

justin: +1
... send to the list and make sure there's no objections to it
... any questions/comments on the teleconference?
... have some issues that are pending review/open that might be stale and can be merged or closed
... and there might be some more serious issues that will take more time (deidentification, say)

<justin> https://www.w3.org/2011/tracking-protection/track/issues/209

ISSUE-209: Description of scope of specification

justin: keeping calls closer to an hour would be great for everyone
... scope discussion would be a good way to start discussing Compliance again

<justin> http://www.w3.org/2011/tracking-protection/drafts/tracking-compliance.html#scope-and-goals

wiki page on proposals: http://www.w3.org/wiki/Privacy/TPWG/Change_Proposal_Scope

<fielding> action fielding finish editorial changes for defn of tracking now that we have a defn of context

<trackbot> Created ACTION-440 - Finish editorial changes for defn of tracking now that we have a defn of context [on Roy Fielding - due 2014-03-19].

justin: amyc has a proposal without the additional paragraph [and with a third-party qualifier]
... suggestion from Brooks about an additional sentence on laying the groundwork
... in the past we've had a much longer scope section

Brooks: logical consistency problem regarding scope, if the TPE doesn't force the allow tracking function

justin: we've generally worked on the functionality in the TPE for indicating allowing, so it makes sense

fielding: haven't updated the Compliance document since past CfOs, suggest that the editors go through and update the draft

justin: fair point, will check with chairs, Heather (editor on Compliance) to bring compliance into accord

<fielding> I mean the editorial bits are now inconsistent with the decisions on definitions

<fielding> …, but easily corrected

npdoty: can Brooks check with Amy?

Brooks: has been a while, I can check with Amy. Amy: yes, would need to refresh, but happy to talk
... if there isn't a choice between 0 and 1, then the scope can't refer to a choice

<fielding> I don't see the point -- we don't have any requirements on black or white

justin: but the TPE does mean that you're allowed to choose DNT

Brooks: but the scope should only talk about MUST requirements

justin: the TPE is designed to allow you to do that

Brooks: misleading to say we give you a choice between black and white, when it's not a MUST to provide both options
... we've been inconsistent about what no choice means

<fielding> TPE defines what is communicated. Compliance defines what is conforming behavior after that communication has been received.

schunter: don't recall colors in TPE

Brooks: about choosing a preference with respect to tracking, only provides the option to not track

schunter: in tracking status resource, have a T & N (previously 0 and 1), we now detail compliance

Brooks: I'm still talking about the user agent side

schunter: the UA can send 0 and 1 that are defined in the TPE

<fielding> We have no need for a section called "Scope"

<fielding> We should have a section called "Introduction"

Brooks: scope is really important, has to be things that are enforced

justin: do you want to propose something for scope that matches past decisions?

Brooks: leave it to someone else to define scope now that you've decided that you can only have one option

justin: up to you all, whether we even need a Scope section or Introduction
... need to log off, but if we want to take up the R discussion again

schunter: better to take it to the list

justin: next week we'll work on narrowing issues that are before us
... chairs will be meeting to talk about how to handle the harder issues going forward
... any general questions?
... will try to keep these calls shorter going forward

[adjourned]

- DRAFT -

Tracking Protection Working Group Teleconference

12 Mar 2014

Attendees

Contents