Tracking Protection Working Group Teleconference

05 Jun 2017

See also: IRC log


fielding, schunter, mikeoneill, aleecia


anyone scribing?

<fielding> scribe: aleecia


schunter: call for objections
... holiday in some parts of EU but going through text proposals. Mike proposed

<schunter> https://github.com/w3c/dnt/milestone/2

schunter: in this we have two issues basically done, one is 22 is under call for objections

mikeoneill: edited the draft, in github. promises are in there.

<schunter> https://github.com/w3c/dnt/blob/master/drafts/tracking-dnt.html

mikeoneill: amended to fix a bug to return array for site-wide consent
... browser can decide to only support site-wide consent, return it in an object

schunter: can we look at it?

mikeoneill: up on the list,
... also support Aleecia’s view on “should” for policy link, that’s the thread, and continue to site-wide exceptions

<fielding> https://github.com/w3c/dnt/commit/803e40b370dfc7e7b986f8cf8936a6a346744d2b

(scribe plugs in laptop while others hunt for text to discuss)

(thanks, Roy!)

<fielding> https://github.com/w3c/dnt/commit/445d5a9a98f2808a0d72e978713e8f33630b00f5

mikeoneill: needs to return a value, resolves to an object, trackingprotectionresults.
... also did same thing for confirm to make it the same :-)
... confirm & restore return: bools site wide, exists
... before was just the promise with a bool, now both are an object so it’s the same and you don’t have to implement two things
... resolves to an object or returns a syntax error.
... site-wide set to true if the exception is site-wide

fielding: normally would require API name if you change the API
... still using the same API names for the published prior — change the names

mikeoneill: good idea
... shorter names would be good

fielding: just need to decide to change them

mikeoneill: sure, will come up with short names by next week

schunter: agrees on shorter new names

mikeoneill: used to be derived from common object, now sep
... some disadvantages put a link up, like David Singer suggested consent in batches.
... wouldn’t have to match them with the confirm if you don’t have the same objects
... there’s no explaination string and site name

fielding: saw the thread, wasn’t sure
... there was one where the store tracking exception was responding with -

mikeoneill: the dictionary, the property bag, they’re just sep but before, on the main list, it’s derived from a common property bag with common properties. before they all had a site name, property string, detail URI and all that
... just pointing that out. not wrong to do it this way, but affects what David Singer was saying on batches of exceptions
... might be better to have it the old way

aleecia: David’s point could be really important with dynamic partnerships on ads

mikeoneill: could also say “these are analytics” and batch by purpose
... user could revoke by purpose
... better granularity
... if we don’t have detail URI & explaination string, can’t match them up

fielding: understand you could theoretically do that but there’s nothing in the spec that mentions this

mikeoneill: don’t have the fields any more

fielding: right, because it didn’t make any sense. if we want them in, we need to add text to say what they’re for, and use them
... fine to have it but needs explaination

mikeoneill: could use Dave Singer in this discussion

schunter: please send email to David

mikeoneill: will do

schunter: new topic
... 2nd topic, must have a TSR

mikeoneill: did that, done

schunter: everyone’s fine with the update and we’re done?

fielding: change granted to requested?

mikeoneill: yes

<fielding> https://github.com/w3c/dnt/commit/803e40b370dfc7e7b986f8cf8936a6a346744d2b


(appears to literally change the word “granted” to “requested”)

schunter: seems done, now we’re waiting for comments on call for objection. once done, ready to publish CR
... any other steps we need?

fielding: we’re changing the API, so have to go to working draft or last call, then request a new CR

schunter: believe that has changed now

fielding: ok, process is changed, not sure

schunter: anything else?
... how about a diff or wait until the end
... will be on holiday in june, so want to push it out so we don’t lose a month

fielding: last diff is basically it

schunter: appreciate after the API changes, then another diff

mikeoneill: site-specific tracking isn’t as pretty so something’s not right, will work on formatting too

fielding: will look at it, but going to London & Paris. might have time in transit

schunter: thanks a lot!
... all from me, will ask W3C staff for any final actions, will wait for Roy’s link for next version
... will look for call for objections comments

<fielding> resent+ mikeoneill, aleecia

<fielding> trackbot, end meeting

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.152 (CVS log)
$Date: 2017/06/05 16:33:15 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.152  of Date: 2017/02/06 11:04:15  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Default Present: fielding, schunter, mikeoneill, aleecia
Present: fielding schunter mikeoneill aleecia
Found Scribe: aleecia
Inferring ScribeNick: aleecia

WARNING: No "Topic:" lines found.

Found Date: 05 Jun 2017
Guessing minutes URL: http://www.w3.org/2017/06/05-dnt-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.

WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report

[End of scribe.perl diagnostic output]