Improving Web Advertising BG

14 September 2021


AramZS, blassey, bmay, Brendan_eyeo_GmbH, dinesh, dmarti, FredBastello, gendler, imeyers, jrosewell___, kleber, kris_chapman, lbasdevant, nlesko, piwanczak, shigeki, wseltzer
Karen, Karen Myers

Meeting minutes

Agenda-curation, introductions

TAG Privacy Principles TF [Lionel Basdevant] https://github.com/w3ctag/privacy-principles

<wseltzer> lbasdevant: great initiative to define what privacy means for the web

<wseltzer> ... curious about current state of work, what its next steps

<wseltzer> https://github.com/w3ctag/privacy-principles

<charlieharrison> +present

Wendy: TAG, the W3C architecture group; and PING
… small group working in public here
… and looking to publish a note that they might recommend to be put forward as a TAG finding
… a draft that hasn't reached consensus is current state
… various input documents and conversation
… that's where it is
… you can see separate open issues and pull requests that are outstanding
… We'll share drafts for wider public review

Lionel: thank you
… do you think these documents should be used when published as an input for this group?

Wendy: it can help inform work, provide common definitions as reference points
… as another input document for work in consideration
… I hope it would be useful

James: similar question in reverse
… we spent last 18 months producing documents
… to represent a wide spectrum of stakeholders
… are this group's document being considered by stakeholders in the other groups?

Wendy: They started with a couple of existing documents, the privacy threat model in PING
… and privacy principles document

<wseltzer> https://github.com/w3ctag/privacy-principles#input-documents

Wendy: and the mode of work has primarily been working through the draft produced
… looking to add/edit there
… I think there are other related documents listed and other related documents could be proposed as input

James: I guess some specific ones from this group
… two documents under "success criteria"
… would represent some of this group's thinking
… and IAB did work on global privacy platform which also contains a lot of good work
… and @ for Addressable Media also published work

<jrosewell___> https://iabtechlab.com/wp-content/uploads/2021/03/iabtechlab_global_privacy_platform_rfc_2021_march.pdf

James: May want to have their work considered as well

Wendy: Thank you for those pointers
… I don't speak for the TAG task force
… we have some participants in this group who are participating there
… I think any of them or us can bring those documents to the task force's attention
… pointing out where in the document as it's structured they might fit, or suggest additions

<jrosewell___> https://www.responsibleaddressablemedia.com/

Lionel: I was wondering to what extent that the work we are doing here might have to be revised or updated
… once we have definition of privacy for the web
… for example...some resources involved
… building test frameworks around APIs
… wonder if @ can happen once privacy definition is agreed upon by all vendors

Wendy: let me go back a step
… the TAG TF work is also preliinary
… looking to produce an opinion
… which would first have to be shared by TF with TAG for review and approval
… and then if they want to make a W3C statement, for approval by the entire membership
… would be presented to entire W3C membership for discussion, consideration and approval
… At this point it is a collection of opinions that people hope will be useful
… but not a definition requiring everyone to change their architectures
… people writing documents and hope they are persuasive
… but it may be at odds with architectures you are developing
… and that might prompt you to consider which things you would change
… and what you might do to bring them closer together
… Will be a lot of opportunity for input before it becomes a W3C position

Moshe: Is there a pointer to the current document?

<wseltzer> https://w3ctag.github.io/privacy-principles/

Wendy: yes, linked in the repository
… this is rendered version; but this draft does not reflect work of the TF yet, but that is the working copy

Joshua: hello
… First, I think it's a great idea to write down the definition of the topics we have wrestled with
… very grateful this document is being worked on
… Question, I was slightly confused by your last answer
… Sounds like document is meant to be persuasive
… but then I heard something around consensus
… wonder if that includes regulators
… given the number of statutes coming out globally
… How is group looking to harmonize with various government positions with work being done inside W3C

Wendy: Thanks, Joshua
… let me clarify
… speaking casual with "persuasive"
… they hope when they write something it means something to listeners
… the repository of the TF

<wseltzer> https://github.com/w3ctag/privacy-principles#input-documents

Wendy: describes the pathway that
… they propose for the document to get consensus among the TF, support from TAG and then possibly as a W3C finding
… the report which can be published as a TAG finding and can potentially be elevated to a W3C statements
… at each stage, input would be sought to reach that level
… I think as a general matter, W3C invites participation from a wide range of stakeholders
… government regulators are among those stakeholders to be invited into conversation
… if we can reach a globally shared understanding of privacy principles, that would be fantastic
… It's a tough challenge to bring all those perspectives together
… especially with different modes of decision making
… we can seek consensus among those who participate here
… Hope that gave some answers to your questions

Joshua: The consensus across W3C will be important
… have terms be defined and less opinion

James: points on procedure and meaning of these things
… we have TAG note for privacy questionnaire
… WGs are listing in their charter
… security and privacy questionnaire is problemmatic because it descriminates against third parties
… if other groups are obliged to follow
… if PING is involved, not sure
… if PING will be represented
… important that this is recognized in practice and not just a note
… given how Security & Privacy note is being used and it's discriminatory

<AramZS> Can someone link the questionnaire James is talking about?

<wseltzer> https://www.w3.org/TR/security-privacy-questionnaire/

<AramZS> Thanks!

James: If someone can pass the notes along, I would be grateful

Wendy: thanks
… if someone has security & privacy questionnaire; here it is
… developed jointly by TAG and PING as a Note
… I take it you are referring to the differentiation between first and third party
… as being discriminatory
… based on basis...

James: yes, a large first party delivering all the services, rather than a smaller first party that relies on a supply chain of vendors
… the domain names, sec & priv boundary
… nice to follow minutes over last couple of months
… but direction of thought is that the fundamental building block is not being addressed
… in order to move on, we need to look at registerable domain names, privacy law and contracts

Wendy: Thank you

<jrosewell___> https://github.com/w3ctag/security-questionnaire/issues/83 - Issue on supply chain choices

Wendy: that is a principle that we have seen/heard debated in a few different places
… Privacy CG, discussion of first party sets raise questions around that as well
… Any further questions or comments?

Lionel: I do have a question, on a different topic

Wendy: thanks for raising this
… I wanted to address the question as you raised
… it is preliminary because the TF is still early in its development
… What I would expect, when they do publish a first draft, even with lots of note about it still being under debate in the text
… so there is an agreed-upon text for discussion
… invite chairs, participants to present here
… and wonder about plans for moving it forward
… I hear the interest here
… and will schedule that conversation when it is ripe
… I don't see anyone else on the queue, so we can move on to another topic

Lionel: Thank you

FLEDGE Origin Trial?

Lionel: about FLEDGE origin trial
… the website says...interoperable in Q4
… is there any additional information on timeline?
… would fenced frames be part of trial or not?

Wendy: don't know if we have Google?

Michael_Kleber: answer is you will know
… we will know when origin trial starts; timeline is forward looking
… best guess for future; we are updating timeline monthly
… not making promises about what will happen
… just let everybody have most up to date info about what we expect to happen in future
… Fenced Frames have not yet shipped in Chrome
… not ready to turn on today
… we don't know answer to first trials of FLEDGE
… and whether will include FF, or just rendering ad in iFrame
… with intention to switch to FF once available
… one of those two will happen, but cannot promise which of those two it will be

Lionel: thanks

Wendy: thanks, Michael

AramZS: hey, Michael
… that is good to hear
… any additional technical progress on FF and how it is going to work
… I understand it is not yet launching
… but wonder if there is technical detail on how it is intended to work?

<wseltzer> https://github.com/shivanigithub/fenced-frame

Kleber: asking specific questions in repo or on the every other week FLEDGE call would be fine way to get answers on what we have done so far
… I think Paul gave an update on which features we have had so far and which things are still in progress

<piwanczak> @Aram - have a look at public design doc for FF: https://docs.google.com/document/d/17rtX55WkxMcfh6ipuhP4mNULIVxUApvYt4ZYXfX2x-s/edit#heading=h.5vbbrxo3hzye

Kleber: We have some features we have already implemented in Chrome and we are working on others; it's a mixed bag and will change over time

AramZS: I see there is a public document, so that is useful

Kleber: about FF or FLEDGE more generally?

<wseltzer> FLEDGE

AramZS: what is schedule for the FLEDGE meetings again?

Kleber: let me post link

<kleber> https://github.com/WICG/turtledove/issues/88

Kleber: they happen during this same hour of the week, on every second Wednesday, which includes tomorrow
… posted issue on Turtledove 88
… where we announce issue and where we talk about it

<kleber> https://github.com/WICG/turtledove/tree/main/meetings

Kleber: let me post link to the directory on the Github repo
… with minutes of all the previous FLEDGE call meetings if you want to review the history

AramZS: tahnks

@: where are links being posted?

Wendy: We are in irc.w3.org, channel web-adv
… that link is sent out with the agenda each time; you can get to irc via the W3C web client or another client

Angelina: hi there
… wanted to ask about aggregated reporting API of origin trial
… wasn't fully aware that origin trial was open; like to get an update on where that stands

Charlie_Harrison: Do you mean attribution reporting for conversion measurement?

Angelina: under attribution API

<wseltzer> https://privacysandbox.com/timeline/

<Katherine_Wei> Is a log of the chats here posted anywhere?

Charlie_Harrison: best guess, it's up to date
… we have begun prototyping in Chrome

Angelina: so everyone is aware, I am kicking off a working group for publishers, advertisers and agencies
… who may not be aware, especially those on tech side
… I am creating a working group to help those constituents in marketplace to better understand those proposals and to do testing frameworks
… I don't think they are aware of the testing OT
… and help them understand on business side, how to get involved

Charlie: we also have meetings every two weeks for attribution reporting; great way to get involved with that API
… I can post details into irc; you are welcome to attend

<charlieharrison> https://github.com/WICG/conversion-measurement-api/issues/80

<kleber> web-adv issue 104 has collected together various links to other w3c meetings: https://github.com/w3c/web-advertising/issues/104

Angelina: part of it is to help companies develop a testing framework more than results, and to educate
… thank you

Charlie: I think that work will be really useful; please come to those meetings
… we would all benefit from any work you are doing in a test suite

Angelina: appreciate it, thank you

Wendy: Thanks for the info sharing
… for those who are newer to this group, we share the minutes of all the meetings and all the links are available in the minutes once posted
… Lots of good information sharing and exchange
… Appreciate that there are lots of conversations happening and have the pointers collected

Dashboard highlights and agenda+? https://w3c.github.io/web-advertising/dashboard/ https://github.com/w3c/web-advertising/labels/agenda%2B

Wendy: Seeing no one on the queue, I would like to remind people of another place where info is collected
… the dashboard offers a wealth of pointers to the topics we cover
… including Github issues opened up on those proposals
… for a scan of activity across a wide range of advertising related proposals, that dashboard is available
… it si an overwhelming amount of information
… So the various calls that the incubations are holding
… and places where overviews are being collected on current state are quite helpful
… any other business for today?
… it is starting to feel like the end of summer in Boston
… that may mean that the pace of proposals and ideas for this group will pick up
… Let's keep to the every other week schedule for a bit longer
… we will come back on 28 Sept.
… invite people to send agenda requests and ideas

Wendy: Thank you, Karen, to scribe during geek week; she invites us to think us about TPAC agenda planning
… and continue to encourage Github discussion to highlight topics
… a good subject for us to come back to on the 28th [Sept]
… anything else?

AramZS: so, a couple of notes
… I added additional document, already been merged with some privacy advocate position; please add
… as we talk about definitions about privacy
… check in with this group
… we discussed idea of a detailed doc to sort through use cases in a single dashboard
… Ben from FB was interested
… but it will take some time and investment to do it if people think it makes sense for me to spend the time

<wseltzer> https://github.com/w3c/web-advertising/pull/117

AramZS: I find the use cases very useful
… and building more of them is work of this group
… I use my own tools to sort through them
… that is a whole other framework
… curious if anyone else has feedback on how they would like it to work...should I invest the time to build it out?

Wendy: thanks, Aram
… I see that Dom HM from W3C team has offered help on tooling
… I can invite you both to a conversation on what type of support W3C could offer
… I find the work useful

Wendell: I'll go +1 on the utility of such a list
… I was calling it "the safe harbor API"
… if there is such a list, and you keep your list within those confines, you can support it and get your business done
… should fit together with ethical principles, etc.
… impressed by what Ben did back in the day

Aram: that is good to hear
… if you don't want to respond, email me
… next step is to build a prototype and run it by Dom
… I did see his feedback
… so next step is putting together the HTML prototype
… and then turning it into...
… jekyll collections
… If I am building it, I don't have to learn any new data standards
… think that is the way to go

Wendy: yes, and +1 to your mention of Ben's work
… and taking off the use cases document
… reminds me that Ben has moved time zones
… I was going to start another poll as to whether this group wants to consider some alternating times, or other time availability
… so we can get more of the globe into conversations
… I don't have any 'flatten the globe' skills
… but I offer the group some time zone polls

+1 more globally friendly for Asia/Australia

Wendy: seeing nobody else on the queue, thank you for lively conversations
… and look forward to hearing input for upcoming meetings and TPAC
… see you all online and back here on the 28th [Sept]

<wseltzer> [adjourned]

Minutes manually created (not a transcript), formatted by scribe.perl version 136 (Thu May 27 13:50:24 2021 UTC).


Succeeded: s/@/Moshe/

Succeeded: s/an/and

Succeeded: s/@/fenced frames/

Succeeded: s/@@/FLEDGE Origin Trial?/

Succeeded: s/@:/Charlie_Harrison:/

Succeeded: s/@:/Charlie_Harrison:/

Succeeded: s/@/jekyll/

No scribenick or scribe found. Guessed: Karen

Maybe present: @, Angelina, Aram, Charlie, Charlie_Harrison, James, Joshua, Lionel, Michael_Kleber, Moshe, Wendell, Wendy