LGTM3 I agree with Alex that we have a bad track record of "temporary" APIs on the web, but I'm comfortable with this given the particulars. A feature which is only a dictionary member is detectable (through a getter) but you have to do extra work and I don't recall ever seeing a problem from removing a no-op dictionary member. Given this, going through OT is extra work to reduce a risk that is already very low. In the very worst case we'd be left with a no-op dictionary member that generates some bindings but does nothing, so it's a small risk of a small problem.
On Mon, Mar 11, 2024 at 11:28 AM Yoav Weiss (@Shopify) < [email protected]> wrote: > > > On Thu, Mar 7, 2024 at 5:03 PM Alex Turner <[email protected]> wrote: > >> Alex and Yoav, >> >> This particular addition is exposed only as an optional parameter on the >> existing `AuctionAdConfig` dictionary passed to `navigator.runAdAuction()`. >> So, I'm not aware of any web visibility from removing the new >> `auctionReportBuyerDebugModeConfig` field (if it was already having no >> effect). But please let me know if I'm missing something there. >> >> Are you referring instead to the existing >> `privateAggregation.enableDebugMode()` function? I agree that removing that >> would be web visible and have this risk. However, that function is already >> fully launched (as part of the I2S here >> <https://groups.google.com/a/chromium.org/g/blink-dev/c/8cKaLstq2QQ/m/Udrbb4ifDQAJ>) >> and isn't affected by this change. >> >> As for why this shouldn't be an OT, alignment with the other way to >> enable debug mode is a big reason. For other usages of Private Aggregation, >> debug mode is enabled with a call to >> `privateAggregation.enableDebugMode()`. This is tied to third-party cookie >> eligibility in the same way (with the exception for Mode B). >> > > That makes perfect sense! I think the benefits of tying this to an OT > (better assurances that removal will be smooth) are outweighed by the > friction it'd introduce for adoption, which is especially stark when > considering the already shipped and much more visible enableDebugMode. > My LGTM stands. > > >> The other major reason is that we intend for this API to be available to >> all callers in order to facilitate testing and adoption of the API, which >> seems at odds with OTs being intended to be run only on a limited subset of >> Chrome traffic. As this API is primarily used by third-parties on a page, >> there is also a potential barrier in the need to edit the top-level page. >> >> Best, >> Alex >> >> On Thu, Mar 7, 2024 at 4:26 AM Yoav Weiss (@Shopify) < >> [email protected]> wrote: >> >>> Alex T - can you expand on the other debug modes that are already >>> shipped? Are they also tied to 3P cookie eligibility? >>> >>> If so, I agree that having all these mechanisms activated together and >>> stop working together makes sense. >>> >>> On Wed, Mar 6, 2024 at 11:04 PM Alex Russell <[email protected]> >>> wrote: >>> >>>> Removal being guaranteed to be backwards compatible depends, at a >>>> minimum, on nullability. As this API would, instead, be globally exposed >>>> (but would not "work" when 3p cookie ineligible), addition and removal >>>> present the usual crop of addition and deprecation issues. >>>> >>>> 3p cookie turndown is now into it's second or third year of delay, so >>>> optimism about the window for growth of use feels hard to justify. >>>> >>>> Why shouldn't this be an OT? >>>> >>>> Best, >>>> >>>> Alex >>>> >>>> On Wed, Mar 6, 2024, 1:54 PM Alex Turner <[email protected]> wrote: >>>> >>>>> Hey Alex, >>>>> >>>>> Totally understand the concern given how difficult removing web APIs >>>>> typically is. However, I think for this case removal should be much >>>>> simpler >>>>> as the functionality (and debug mode more generally) is already tied to >>>>> third-party cookie eligibility*. In other words, after third-party cookie >>>>> deprecation is fully launched, this new feature will (automatically) have >>>>> no effect. At that point, removing the new field from the IDL definitions >>>>> should also be fully backwards compatible as it would simply be ignored. >>>>> >>>>> We leaned away from an OT given that other uses of Private >>>>> Aggregation's debug mode do not require an OT token and we wanted to keep >>>>> the workflow across debug mode consistent. Other considerations were the >>>>> increased difficulty for usage of the API in third-party iframes and the >>>>> large number of expected pages impacted (i.e. likely much higher than >>>>> 0.5%). We're definitely willing to commit to formally removing this >>>>> feature >>>>> shortly after third-party cookie deprecation is fully launched. >>>>> >>>>> I hope this addresses your concerns, but very happy to discuss further. >>>>> >>>>> (*I do want to acknowledge the exception for Mode B traffic softens >>>>> this claim, but we have communicated the temporary nature of this >>>>> exception >>>>> in the developer documentation >>>>> <https://developers.google.com/privacy-sandbox/relevance/private-aggregation#enabledebugmode>. >>>>> Also, note that trying to enable debug mode already has no effect for >>>>> users >>>>> who have explicitly disabled third-party cookies.) >>>>> >>>>> Best, >>>>> Alex >>>>> >>>>> On Wed, Mar 6, 2024 at 12:02 PM Alex Russell <[email protected]> >>>>> wrote: >>>>> >>>>>> Hey all, >>>>>> >>>>>> This may be overfitting against my personal priors, but I'm *intensely >>>>>> *skeptical of any web platform API addition that claims to be >>>>>> "temporary". If we want a temporary mechanism, we can use OTs and set a >>>>>> date-certain for removal and prevent over-use that would back us into a >>>>>> corner. >>>>>> >>>>>> Without some sort of guardrail, I'm not optimistic that we'll be able >>>>>> to actually get rid of a useful debugging aid that is web-exposed. >>>>>> >>>>>> Can we reformulate this as an OT? Or is the team proposing this >>>>>> willing to document a more thorough plan for deprecation and removal, >>>>>> perhaps including reporting of use (via UMA histograms)? >>>>>> >>>>>> Best, >>>>>> >>>>>> Alex >>>>>> >>>>>> On Wednesday, March 6, 2024 at 7:49:16 AM UTC-8 Mike Taylor wrote: >>>>>> >>>>>>> LGTM2 >>>>>>> On 3/6/24 5:39 AM, Yoav Weiss (@Shopify) wrote: >>>>>>> >>>>>>> LGTM1 >>>>>>> >>>>>>> On Thu, Feb 29, 2024 at 7:29 PM Alex Turner <[email protected]> >>>>>>> wrote: >>>>>>> >>>>>>>> Contact emails [email protected] >>>>>>>> >>>>>>>> Explainer >>>>>>>> https://github.com/WICG/turtledove/blob/main/FLEDGE_extended_PA_reporting.md#temporary-debugging-mechanism >>>>>>>> >>>>>>>> Specification >>>>>>>> https://github.com/patcg-individual-drafts/private-aggregation-api/pull/118 >>>>>>>> >>>>>>>> Summary >>>>>>>> >>>>>>>> Adds support for Private Aggregation’s debug mode for the per-buyer >>>>>>>> extended Private Aggregation reporting to Protected Audience sellers >>>>>>>> (aka >>>>>>>> auctionReportBuyers reporting). This is done with a new, optional >>>>>>>> auctionReportBuyerDebugModeConfig field passed to runAdAuction(). >>>>>>>> auctionReportBuyers reporting is currently the only use of Private >>>>>>>> Aggregation reports that does not have a way to enable its debug mode >>>>>>>> (a >>>>>>>> temporary mechanism tied to third-party cookie eligibility that relaxes >>>>>>>> some of the API’s privacy constraints to allow for easier debugging and >>>>>>>> integration). >>>>>>>> >>>>>>>> >>>>>>>> Blink component Blink>PrivateAggregation >>>>>>>> <https://bugs.chromium.org/p/chromium/issues/list?q=component:Blink%3EPrivateAggregation> >>>>>>>> >>>>>>>> TAG review https://github.com/w3ctag/design-reviews/issues/846 (We >>>>>>>> have not requested a signal for these changes specifically.) >>>>>>>> >>>>>>>> TAG review status Pending >>>>>>>> >>>>>>>> Risks >>>>>>>> >>>>>>>> >>>>>>>> Interoperability and Compatibility >>>>>>>> >>>>>>>> This feature is optional and backwards compatible. >>>>>>>> >>>>>>>> >>>>>>>> *Gecko*: No signal ( >>>>>>>> https://github.com/mozilla/standards-positions/issues/805) We have >>>>>>>> not requested a signal for this change specifically. >>>>>>>> >>>>>>>> *WebKit*: No signal ( >>>>>>>> https://github.com/WebKit/standards-positions/issues/189) We have >>>>>>>> not requested a signal for this change specifically. >>>>>>>> >>>>>>>> *Web developers*: Positive ( >>>>>>>> https://github.com/WICG/turtledove/issues/709) This has been >>>>>>>> requested by developers. >>>>>>>> >>>>>>>> *Other signals*: >>>>>>>> >>>>>>>> WebView application risks >>>>>>>> >>>>>>>> Does this intent deprecate or change behavior of existing APIs, >>>>>>>> such that it has potentially high risk for Android WebView-based >>>>>>>> applications? >>>>>>>> >>>>>>>> None >>>>>>>> >>>>>>>> >>>>>>>> Debuggability >>>>>>>> >>>>>>>> This is primarily a debugging feature -- allowing the existing >>>>>>>> debug mode for this kind of Private Aggregation reporting. >>>>>>>> >>>>>>>> >>>>>>>> Will this feature be supported on all six Blink platforms (Windows, >>>>>>>> Mac, Linux, ChromeOS, Android, and Android WebView)? >>>>>>>> >>>>>>>> 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> >>>>>>>> ? Yes >>>>>>>> >>>>>>>> Flag name on chrome://flags None >>>>>>>> >>>>>>>> Finch feature name >>>>>>>> PrivateAggregationAuctionReportBuyerDebugModeConfig >>>>>>>> >>>>>>>> Requires code in //chrome? False >>>>>>>> >>>>>>>> Tracking bug https://crbug.com/1513013 >>>>>>>> >>>>>>>> Launch bug https://launch.corp.google.com/launch/4302460 >>>>>>>> >>>>>>>> Estimated milestones >>>>>>>> >>>>>>>> We’re aiming to ship in M123. >>>>>>>> >>>>>>>> Anticipated spec changes >>>>>>>> >>>>>>>> None >>>>>>>> >>>>>>>> Link to entry on the Chrome Platform Status >>>>>>>> https://chromestatus.com/feature/5088927692357632 >>>>>>>> >>>>>>>> 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 [email protected]. >>>>>>>> To view this discussion on the web visit >>>>>>>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAA%2BBiFkYb%3DWswXSKM%2B9ESFdgTBUdbjoDq21Q5sWJm%2BPWqqE_nw%40mail.gmail.com >>>>>>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAA%2BBiFkYb%3DWswXSKM%2B9ESFdgTBUdbjoDq21Q5sWJm%2BPWqqE_nw%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 [email protected]. >>>>>>> >>>>>>> To view this discussion on the web visit >>>>>>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOmohS%2BJt8bhfgO5iLVa7j3byo%2Bov8%3D_kh-Jt6hLfBEbWyJHPA%40mail.gmail.com >>>>>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOmohS%2BJt8bhfgO5iLVa7j3byo%2Bov8%3D_kh-Jt6hLfBEbWyJHPA%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 [email protected]. >> To view this discussion on the web visit >> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAA%2BBiFkLqQ6DwUqtg1z_1jFStOTjLT_mma6%3Dx_c47C_ZKOTnEA%40mail.gmail.com >> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAA%2BBiFkLqQ6DwUqtg1z_1jFStOTjLT_mma6%3Dx_c47C_ZKOTnEA%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 [email protected]. > To view this discussion on the web visit > https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOmohSLJkpiDk6tMmfj1d3-oiA896GLv9O5tz_XvnQuh2XPEUQ%40mail.gmail.com > <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOmohSLJkpiDk6tMmfj1d3-oiA896GLv9O5tz_XvnQuh2XPEUQ%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 [email protected]. To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAARdPYepC8EpgvzBwPtcm5-vjt55iXGv5UO5GUpBkjWav80A4g%40mail.gmail.com.
