On 7/1/23 3:09 AM, Paul Jensen wrote:
On Wed, Jun 28, 2023 at 5:33 AM Yoav Weiss <yoavwe...@chromium.org> wrote:
On Tue, Jun 27, 2023 at 9:54 PM Paul Jensen
<pauljen...@chromium.org> wrote:
Yoav,
Protected Audiences has been fortunate to have a ton of design
contributions and feedback, but consequently has a lot of
issues filed. We try to respond to all issues, as you can see
by the discussion comments on nearly all issues. I went
through and triaged all the issues recently. I closed many of
them, created some labels and labeled many of them. Here’s
where I think the open issues stand:
*
65
<https://github.com/WICG/turtledove/labels/Non-breaking%20Feature%20Request>I
labeled “Non-breaking Feature Request”, meaning they’re
requesting new functionality that is unlikely to cause
backwards compatibility issues.
*
29 <https://github.com/WICG/turtledove/labels/spec>are
spec related. As Dominic said above, most of these
changes are unlikely to break web content.
*
8
<https://github.com/WICG/turtledove/labels/Looking%20for%20feedback>are
seeking feedback rather than pointing to a problem.
*
4
<https://github.com/WICG/turtledove/labels/compat%20concern>could
potentially break compatibility. I think for all of these
we’ve decided to not adopt the proposed changes or we’ve
decided to adopt the proposed changes but as part of our
longer-term plans in the future. I should note that
recently we adopted many breaking changes to our API, but
did so in a way that supports backwards compatibility, so
we can wean developers off of the old APIs without causing
immediate significant breakage. If we chose to adopt some
of these changes, I imagine we could do so in a similar
non-breaking way.
*
86
<https://github.com/WICG/turtledove/issues?q=is%3Aopen+-label%3A%22Non-breaking+Feature+Request%22++-label%3Aspec+-label%3A%22Looking+for+feedback%22+-label%3A%22compat+concern%22>didn’t
fit well into a particular category:
o
Some were questions seeking to clarify details of our
timeline or the explainer or design.
o
Some were discussions that are mostly addressed but
left open so we don’t forget about remaining pieces.
o
Some are open discussions or examples.
I think it’s worth noting that our usage of the issue system
differs from those of many other folks who ship features: We
tend to use the issues as open forums as opposed to only
leaving open issues that need to have decisions made. Many of
the issues predate the FLEDGE explainer and represent design
discussions that culminated in FLEDGE’s design.
Thanks for going over the issues!! To be clear, the number of
issues is not a concern in itself, and is indeed an indication of
the level of engagement this had.
This list of compat-related issues
<https://github.com/WICG/turtledove/labels/compat%20concern> is
the only relevant bit for this intent IMO. At the same time, it'd
be good to settle these issues, or at least have a clear path
towards future-compat around them, before shipping. WDYT?
I think we’ve settled on paths to addressing each of the compat issues:
#444 <https://github.com/WICG/turtledove/issues/444>and #586
<https://github.com/WICG/turtledove/issues/586>I think we’ve settled
on not pursuing for reasons expressed in the issues.
#522 <https://github.com/WICG/turtledove/issues/522>has been our long
term plan but we've heard feedback that it blocks adoption and
usability at this stage, especially in the long-tail of advertisers.
Providing a solution to audience stealing is an important goal of
Protected Audience. Our current implementation offers opt-in
protection via our Permission-Policy, and we're going to continue to
look for an ergonomic solution that facilitates adoption sufficiently
to offer the protection by default.
#554 <https://github.com/WICG/turtledove/issues/554>is something we
might do, and could do in the future while offering a temporary
backward-compatible period. It doesn’t have significant developer
benefits, other than making it potentially more web-like, so I’m
reluctant to adopt it.
Thanks Paul. Could you close out 586 and leave comments on 522 and 554
with your current thinking?
Re: 554, do you have plans to update the spec to match Chromium's
implementation of setBid(), setPriority(), and
setPrioritySignalsOverride()? Or do something else?
I hope the labels I added make it clearer which are future
enhancements and not likely to break backwards compatibility.
I honestly think over the years before our Origin Trial and
over the course of our lengthy Origin Trial we’ve addressed
all the feedback for core functionality in Protected Audience
and don’t anticipate breaking backwards compatibility in
significant ways.
On Thu, Jun 22, 2023 at 3:56 AM Yoav Weiss
<yoavwe...@chromium.org> wrote:
Glancing at the open issues, I see 291 of them
<https://github.com/WICG/turtledove/issues?page=2&q=is%3Aopen>..
Would it be possible to go over the issues and label them
so that it's clearer which are about future enhancements,
which are editorial and which may have an impact on the
processing model or API shape in ways that can impact
future compatibility?
On Wed, Jun 21, 2023 at 8:00 PM Dominic Farolino
<d...@chromium.org> wrote:
As the spec mentor for this feature I'll offer a spec
maturity summary
<https://www.chromium.org/blink/launching-features/#:~:text=If%20your%20specification%20isn%27t%20a%20modification%20of%20an%20existing%20specification%2C%20include%20a%20one%2Dline%20spec%20maturity%20summary%20from%20someone%20outside%20your%20team%20(like%20your%20spec%20mentor)%20who%20has%20done%20a%20review.>.
@Jeffrey Yasskin <mailto:jyass...@chromium.org> and I
reviewed the spec in detail recently and were pleased
with the improvements that the team worked with us to
make recently, especially with regards to:
*
Formalizing the interaction with times and dates
*
Adding rigor to the in parallel work (and its
interaction with the main thread and the Script
Runner realms)
*
Fetch integration
*
Specifying the conversions from internal spec data
to JS objects when calling into the Script Runners
<https://wicg.github.io/turtledove/#script-runners>,
mostly by increasing the use of WebIDL
In a few of these points there is still work to be
done, and we've been filing bugs
<https://github.com/WICG/turtledove/labels/spec>against
the specification for individual tasks that the team
has committed to making progress on in the very near
future. The spec overall is not yet very readable
<https://github.com/WICG/turtledove/issues/646>, which
means external reviewers will have to spend time to
understand the flow before they can give substantive
feedback. From a completenessperspective, the spec
still has over a dozen "TODOs" (I expect that they’ll
be finished soon given how many have recently closed),
including the bulk of the integration with Fenced
Frames <https://github.com/WICG/turtledove/pull/616>,
whose completion might help other browser engines
notice new interoperability issues. The team is
completing these at a good pace, but this implies that
in addition to finishing pieces of the spec that
document the current implementation, there will
probably be minor web-visible changes after shipping
in M115. However, most of these changes are unlikely
to break web content, and if anything bigger comes up,
the Privacy Sandbox's general tools for migrating
their users should be effective.
On Tue, Jun 20, 2023 at 4:06 PM Paul Jensen
<pauljen...@chromium.org> wrote:
*Contact emails*
pauljen...@chromium.org
<mailto:pauljen...@chromium.org>,
kle...@google.com
<mailto:kle...@google.com>,ajvelasq...@google.com
<mailto:ajvelasq...@google.com>
Explainer
https://github.com/WICG/turtledove/blob/master/FLEDGE.m
<https://github.com/WICG/turtledove/blob/master/FLEDGE.md>d
Specification
https://wicg.github.io/turtledove
<https://wicg.github.io/turtledove>
Summary
The Protected Audience API (formerly known as
FLEDGE) provides a method of interest-group
advertising without having to track individual
users’ detailed browsing history as is done today
with third-party cookies. Additional advantages
over cookies include time limits on group
membership, better user controls, and more user
transparency.
Blink component
Blink>InterestGroups
<https://bugs.chromium.org/p/chromium/issues/list?q=component%3ABlink%3EInterestGroups&can=2>
TAG review
https://github.com/w3ctag/design-reviews/issues/723
<https://github.com/w3ctag/design-reviews/issues/723>
TAG review status
Pending since March 2022
Risks
Compatibility
This is not a breaking change. To use it, sites
will need to call the Protected Audience API.
There is no change to existing behavior for sites
not calling the API. It’s worth noting that the
spec uses WebIDL to describe the script runners
<https://wicg.github.io/turtledove/#script-runners>but
the implementation does not. There may be minor
compat issues as we align the implementation with
the WebIDL semantics over time.
Interoperability
Gecko: No signal, requested March 2023
<https://github.com/mozilla/standards-positions/issues/770>
WebKit: No signal, requested March 2023
<https://github.com/WebKit/standards-positions/issues/158>
Edge: Edge explored interest group based
advertising, namely withthe PARAKEET proposal
<https://github.com/WICG/privacy-preserving-ads/blob/main/Parakeet.md>.
PARAKEET shares much of its API with Protected
Audience but as discussed in TPAC 2022
<https://docs.google.com/presentation/d/1QQgrm4oaRRRBr1gfvKj7D8rS2EW8kRgRUHPscvR8BNo/edit#slide=id.g15545e7b627_0_173>,
involves proxying data to non-trusted servers in
real-time whereas Protected Audience does not have
long term plans to do this.
Web developers: Significant interest from many web
developers. Significant Origin Trial participation
<https://github.com/WICG/turtledove/blob/main/fledge-tester-list.md>.
WICG FLEDGE calls
<https://github.com/WICG/turtledove/issues/88>are
heavily attended. Interest in Protected Audience
is further evidenced by the many related
discussions and proposals that Protected
Audience’s design draws from, most notably:
*
Theoriginal TURTLEDOVE
<https://github.com/WICG/turtledove/blob/main/Original-TURTLEDOVE.md>from
Chrome.
*
SPARROW <https://github.com/WICG/sparrow>from
Criteo.
*
Outcome-based TURTLEDOVE
<https://github.com/WICG/turtledove/blob/main/OUTCOME_BASED.md>andProduct-level
TURTLEDOVE
<https://github.com/WICG/turtledove/blob/main/PRODUCT_LEVEL.md>from
RTB House.
*
Dovekey
<https://github.com/google/ads-privacy/tree/master/proposals/dovekey>from
Google Ads.
*
PARRROT
<https://github.com/prebid/identity-gatekeeper/blob/master/proposals/PARRROT.md>from
Magnite.
*
TERN
<https://github.com/WICG/turtledove/blob/main/TERN.md>from
NextRoll.
Demo link
https://developer.chrome.com/docs/privacy-sandbox/fledge-api/#demo
<https://developer.chrome.com/docs/privacy-sandbox/fledge-api/#demo>
Debuggability
To learn more about debugging Protected Audience
in Chrome please follow these
links:https://developer.chrome.com/blog/fledge-api/#debugging
<https://developer.chrome.com/blog/fledge-api/#debugging>https://developer.chrome.com/blog/fledge-api/#observe-fledge-events
<https://developer.chrome.com/blog/fledge-api/#observe-fledge-events>
Will this feature be supported on all six
Blink platforms (Windows, Mac, Linux,
Chrome OS, Android, and Android WebView)?
All except WebView
Is this feature fully tested by
web-platform-tests
<https://chromium.googlesource.com/chromium/src/+/main/docs/testing/web_platform_tests.md>?
We've tested all of the primary functionality in
WPT. This API has a lot of surface area and so
we're continuing to add platform tests over time.
https://wpt.fyi/results/?q=fledge
<https://wpt.fyi/results/?q=fledge>
Flag name
InterestGroupStorage,AdInterestGroupAPI,Fledge,AllowURNsInIframes,BiddingAndScoringDebugReportingAPI
Requires code in //chrome?
Yes, for settings UI controls and k-anonymity
server communication.
Estimated milestones
Has been in Origin Trial since M101. We intend to
start an incremental ramp to 100% in Stable with
Chrome Release M115.
Anticipated spec changes
*
We’re addressing some remaining TODOs and
specifying some recently added non-breaking
features (e.g. #304
<https://github.com/WICG/turtledove/issues/304>,
#305
<https://github.com/WICG/turtledove/issues/305>,
#310
<https://github.com/WICG/turtledove/issues/310>,
#166
<https://github.com/WICG/turtledove/issues/166>).
*
Moving beyond our core use cases, we
anticipate the need to support new
functionality going forward. We don’t
currently anticipate changes that would break
backwards compatibility.
*
Support for Bidding and Auction services
<https://groups.google.com/a/chromium.org/g/blink-dev/c/bGd_nPuUrUg/m/j39WQ7e2AwAJ>is
in progress. This is a non-breaking
additional feature.
Link to entry on the Chrome Platform Status
https://chromestatus.com/feature/5733583115255808
<https://chromestatus.com/feature/5733583115255808>
Links to previous Intent discussions
Intent to Prototype:
https://groups.google.com/a/chromium.org/g/blink-dev/c/w9hm8eQCmNI
<https://groups.google.com/a/chromium.org/g/blink-dev/c/w9hm8eQCmNI>
Intent to Experiment:
https://groups.google.com/a/chromium.org/g/blink-dev/c/0VmMSsDWsFg/m/_0T5qleqCgAJ
<https://groups.google.com/a/chromium.org/g/blink-dev/c/0VmMSsDWsFg/m/_0T5qleqCgAJ>
Intent to Extend Origin Trial:
https://groups.google.com/a/chromium.org/g/blink-dev/c/SD8Ot2gpz4g/m/A9uA-_cGAwAJ
<https://groups.google.com/a/chromium.org/g/blink-dev/c/SD8Ot2gpz4g/m/A9uA-_cGAwAJ>
https://groups.google.com/a/chromium.org/g/blink-dev/c/gpmaOi3of_w/m/SyMclFhMAAAJ
<https://groups.google.com/a/chromium.org/g/blink-dev/c/gpmaOi3of_w/m/SyMclFhMAAAJ>
https://groups.google.com/a/chromium.org/g/blink-dev/c/CBrV-2DrYFI/m/RTojC6kHAgAJ
<https://groups.google.com/a/chromium.org/g/blink-dev/c/CBrV-2DrYFI/m/RTojC6kHAgAJ>
--
You received this message because you are
subscribed to the Google Groups "blink-dev" group.
To unsubscribe from this group and stop receiving
emails from it, send an email to
blink-dev+unsubscr...@chromium.org.
To view this discussion on the web visit
https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CABQTWrn8eM3wOtUY3RzmDrt7SVxR_y_6Fo02bJ%2BF1bzbwpFfkQ%40mail.gmail.com
<https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CABQTWrn8eM3wOtUY3RzmDrt7SVxR_y_6Fo02bJ%2BF1bzbwpFfkQ%40mail.gmail.com?utm_medium=email&utm_source=footer>.
--
You received this message because you are subscribed
to the Google Groups "blink-dev" group.
To unsubscribe from this group and stop receiving
emails from it, send an email to
blink-dev+unsubscr...@chromium.org.
To view this discussion on the web visit
https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAP-uykDhn3EzgNacgnEExhiLwrdnc%2Bf7ZV6qMf%3DHk1ns1oHdTw%40mail.gmail.com
<https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAP-uykDhn3EzgNacgnEExhiLwrdnc%2Bf7ZV6qMf%3DHk1ns1oHdTw%40mail.gmail.com?utm_medium=email&utm_source=footer>.
--
You received this message because you are subscribed to the Google
Groups "blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to blink-dev+unsubscr...@chromium.org.
To view this discussion on the web visit
https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CABQTWr%3D1ROXvBN-k6trfMvLpnE74avuc4WtmyZRrAuOHdh0zNQ%40mail.gmail.com
<https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CABQTWr%3D1ROXvBN-k6trfMvLpnE74avuc4WtmyZRrAuOHdh0zNQ%40mail.gmail.com?utm_medium=email&utm_source=footer>.
--
You received this message because you are subscribed to the Google Groups
"blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to blink-dev+unsubscr...@chromium.org.
To view this discussion on the web visit
https://groups.google.com/a/chromium.org/d/msgid/blink-dev/a62ad34d-2479-2b5d-3d1c-4caf3b215e59%40chromium.org.
