On Wed, Mar 6, 2024 at 12:07 PM Yoav Weiss (@Shopify) <
yoavwe...@chromium.org> wrote:

>
>
> On Mon, Mar 4, 2024 at 7:00 PM Paul Jensen <pauljen...@chromium.org>
> wrote:
>
>>
>>
>> On Wed, Feb 28, 2024 at 12:24 PM Yoav Weiss (@Shopify) <
>> yoavwe...@chromium.org> wrote:
>>
>>>
>>>
>>> On Thu, Feb 22, 2024 at 6:07 PM 'Orr Bernstein' via blink-dev <
>>> blink-dev@chromium.org> wrote:
>>>
>>>> Contact emails
>>>>
>>>> pauljen...@chromium.org
>>>>
>>>>
>>>> Explainer
>>>>
>>>> Downsampled forDebugOnly: https://github.com/WICG/turtledove/pull/1020
>>>>
>>>> Increase number of component ads:
>>>> https://github.com/WICG/turtledove/pull/1003
>>>>
>>>>
>>>> Specification
>>>>
>>>> Downsampled forDebugOnly: https://github.com/WICG/turtledove/pull/1023
>>>>
>>>> Increase number of component ads:
>>>> https://github.com/WICG/turtledove/pull/1004
>>>>
>>>>
>>>> Summary
>>>>
>>>> Downsampled forDebugOnly:
>>>>
>>>> The forDebuggingOnly.reportAdAuctionWin() and
>>>> forDebuggingOnly.reportAdAuctionLoss() (fDO) APIs were added to the
>>>> Protected Audience (PA) API to allow debugging from within bidding and
>>>> scoring scripts in PA auctions.  We originally intended to remove these
>>>> APIs at third-party cookie deprecation (3PCD) time, but received feedback
>>>> that they were essential for doing root cause analysis in escalation
>>>> situations (i.e. critical crash).  Instead of removing debug reports
>>>> entirely, we now plan to heavily downsample them at 3PCD time as follows:
>>>> they will only send a report with a 1/1000 probability.  Furthermore, if a
>>>> report is sent, the browser will not send another for 3 years (“lockout”),
>>>> and if a report is not sent (999/1000 of the time), future calls to the fDO
>>>> API from the calling origin will be ignored for 2 weeks 90% of the time and
>>>> 1 year 10% of the time (“cooldown”).  To avoid biasing towards new browser
>>>> installs (which aren’t in lockout or cooldown), we may place new browser
>>>> installs initially in a shorter lockout period. For now, until 3PCD, we’ll
>>>> expose whether a fDO caller is in the “lockout” or “cooldown” periods.
>>>>
>>>> Increase number of component ads:
>>>>
>>>> Today, PA allows selection of ads containing 20 component ads (aka
>>>> product level ads).  We plan to increase this number from 20 to 40 as we
>>>> received feedback that ads with higher numbers of components were critical
>>>> (e.g. for ads that rotate through 3 sets of 12 products).
>>>>
>>>
>>> Can you expand on the implications of increasing that number? What's the
>>> tradeoff involved?
>>>
>>
>> As discussed on github here
>> <https://github.com/WICG/turtledove/issues/810> and in person here
>> <https://github.com/WICG/turtledove/blob/main/meetings/2023-10-04-FLEDGE-call-minutes.md#leo---discussion-about-increasing-the-number-of-ad-components-entering-the-bid-from-20-to-40-httpsgithubcomwicgturtledoveissues810>,
>> we've heard from adtechs that large portions of their ad inventory cannot
>> be displayed without allowing higher numbers of component ads, so
>> increasing this number restores more publisher site revenue.  The downsides
>> to increasing this number are fairly minor:  There's a negligible privacy
>> impact as the component ads are not shared with PA reporting scripts, are
>> required to be k-anonymous, and when displayed, each component ad can be
>> isolated from each other in separate Fenced Frames.
>>
>
> I understand the benefits of increasing the number of ads, but are there
> any pointers to past discussion/analysis RE the privacy impact? I
> understand it's not huge (and it's not my role to judge the privacy risk -
> that's what the privacy review is for). I'd just love to better understand
> this :D
>

WRT the privacy impact, as it's negligible, there isn't much to discuss so
there isn't much prior discussion/analysis other than in this email thread
and in the two links I posted before.  If there's some aspect you'd like to
discuss further or dig into more, I'm happy to engage.


>
>
>>
>>
>>>
>>>
>>>>
>>>> Blink component
>>>>
>>>> Blink>InterestGroups
>>>> <https://bugs.chromium.org/p/chromium/issues/list?q=component:Blink%3EInterestGroups>
>>>>
>>>>
>>>> TAG review
>>>>
>>>> The parent proposal, Protected Audience, is still pending:
>>>> https://github.com/w3ctag/design-reviews/issues/723
>>>>
>>>>
>>>> TAG review status
>>>>
>>>> Pending
>>>>
>>>>
>>>> Risks
>>>>
>>>>
>>>> Interoperability and Compatibility
>>>>
>>>> Downsampled forDebugOnly: No expected breakage before 3PCD as the
>>>> downsampling will not be performed until then.  For now only the status of
>>>> whether a fDO caller is in the “lockout” or “cooldown” periods is exposed.
>>>> After 3PCD, the downsampling will surely disrupt some potential uses of the
>>>> reporting channel, but this is an essential privacy requirement of the
>>>> Protected Audience API.
>>>>
>>>> Increase number of component ads: This is an increase to the limit, so
>>>> no breakage is expected.
>>>>
>>>>
>>>> Gecko & WebKit: No signal on parent proposal, Protected Audience.
>>>> Asked in the Mozilla forum here
>>>> <https://github.com/mozilla/standards-positions/issues/770>, and in
>>>> the Webkit forum here
>>>> <https://github.com/WebKit/standards-positions/issues/158>.
>>>>
>>>>
>>>> Web developers:
>>>>
>>>> Downsampled forDebugOnly: Discussed here
>>>> <https://github.com/WICG/turtledove/issues/632> and in person here
>>>> <https://github.com/WICG/turtledove/blob/main/meetings/2024-01-17-FLEDGE-call-minutes.md#:~:text=on%20the%20URL.-,forDebuggingOnly%20availability,-%2D%20%23632>
>>>> .
>>>>
>>>> Increase number of component ads: Discussed here
>>>> <https://github.com/WICG/turtledove/issues/810> and in person here
>>>> <https://github.com/WICG/turtledove/blob/main/meetings/2023-10-04-FLEDGE-call-minutes.md#leo---discussion-about-increasing-the-number-of-ad-components-entering-the-bid-from-20-to-40-httpsgithubcomwicgturtledoveissues810>
>>>> .
>>>>
>>>>
>>>> Debuggability
>>>>
>>>> Use of both APIs is debuggable via DevTools debugging of Protected
>>>> Audience bidding and scoring scripts.
>>>>
>>>>
>>>> Will this feature be supported on all six Blink platforms (Windows,
>>>> Mac, Linux, ChromeOS, Android, and Android WebView)?
>>>>
>>>> It will be supported on all platforms that support Protected Audience,
>>>> so all but WebView.
>>>>
>>>>
>>>> Is this feature fully tested by web-platform-tests
>>>> <https://chromium.googlesource.com/chromium/src/+/main/docs/testing/web_platform_tests.md>
>>>> ?
>>>>
>>>> We plan to land web-platform-tests for both features shortly.
>>>>
>>>>
>>>> Flag name on chrome://flags
>>>>
>>>> None
>>>>
>>>>
>>>> Finch feature name
>>>>
>>>> FledgeEnableFilteringDebugReportStartingFrom, FledgeAdComponentLimit
>>>>
>>>>
>>>> Requires code in //chrome?
>>>>
>>>> False
>>>>
>>>>
>>>> Estimated milestones
>>>>
>>>> Shipping on desktop and Android in M122.
>>>>
>>>>
>>>> Anticipated spec changes
>>>>
>>>> None related to these features.
>>>>
>>>>
>>>> Link to entry on the Chrome Platform Status
>>>>
>>>> https://chromestatus.com/feature/5099305180594176
>>>>
>>>>
>>>> This intent message was generated by Chrome Platform Status
>>>> <https://chromestatus.com/>.
>>>>
>>>> --
>>>> 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/d5ba8ace-8a65-461f-9124-9fcd7a232c26n%40chromium.org
>>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/d5ba8ace-8a65-461f-9124-9fcd7a232c26n%40chromium.org?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/CABQTWrkwm9J3HG6d-_9KewhW5QQSKKPinQeZRnBqJSgNKO3veg%40mail.gmail.com.

Reply via email to