add SPARROW's answers to the list of use-cases in the community propo… #64
Conversation
|
Hi @Pl-Mrcy If there is no straightforward reference at least to have redirect to paragraph where we can understand this from context Same thing would apply to TD proposals and probably we can add this in later edits. @benjaminsavage WDYT? |
|
I'm concerned about some of the claims of support made here — I can't see how they're possible with what I understood of the SPARROW design. I've left a few questions in WICG/sparrow#11. |
support_for_advertising_use_cases.md
Outdated
| | [Cross Browser / Cross Device Measurement](#cross-browser--cross-device-measurement) | No support | No support | Facebook proposal for “[Cross Browser Anonymous Conversion Reporting](https://github.com/w3c/web-advertising/blob/master/cross-browser-anonymous-conversion-reporting.md)” | | ||
| | [Conversion Fraud](#conversion-fraud) | A section of the [Multi-Browser Aggregation Service Explainer](https://github.com/WICG/conversion-measurement-api/blob/master/SERVICE.md#authenticating-inputs) describes using a service that includes an authentication mechanism. This use case is also being explored for the event level API in this [GitHub Issue](https://github.com/WICG/conversion-measurement-api/issues/13). | Confirmation this is a use-case Safari cares about addressing. In the process of collaboratively designing a solution on this [GitHub Issue](https://github.com/WICG/ad-click-attribution/issues/27). | Facebook proposal for “[Private Fraud Prevention](https://github.com/siyengar/private-fraud-prevention)” | | ||
| | [Click Flooding](#click-flooding) | No solutions yet for this problem. However, lift measurement could be a potential way of measuring this problem. This might be supported by the “[Aggregate Reporting API](https://github.com/csharrison/aggregate-reporting-api)”. | No solutions yet for this problem. | Facebook proposal for “[Private Lift Measurement](https://github.com/w3c/web-advertising/blob/master/private-lift-measurement-conceptual-overview.md)” a potential way of measuring this problem | | ||
| | [Conversion Fraud](#conversion-fraud) | A section of the [Multi-Browser Aggregation Service Explainer](https://github.com/WICG/conversion-measurement-api/blob/master/SERVICE.md#authenticating-inputs) describes using a service that includes an authentication mechanism. This use case is also being explored for the event level API in this [GitHub Issue](https://github.com/WICG/conversion-measurement-api/issues/13). | Confirmation this is a use-case Safari cares about addressing. In the process of collaboratively designing a solution on this [GitHub Issue](https://github.com/WICG/ad-click-attribution/issues/27). | Facebook proposal for “[Private Fraud Prevention](https://github.com/siyengar/private-fraud-prevention)” <br><br> SPARROW : Thanks to exact reporting, it will be possible to link conversion to both an interest group and a publisher. It will also be possible to have user-level granularity on the website, helping to spot fraudulent activity. | |
There was a problem hiding this comment.
SPARROW : Thanks to exact reporting, it will be possible to link conversion to both an interest group and a publisher.
I don't see how this is compatible with the privacy goals of TURTLEDOVE, which the SPARROW explainer says are its goals also. Your Privacy Considerations section says "The Gatekeeper receives interests group x publisher data, but cannot link this data to individual users since it has no user-level information." Linking to a specific conversion seems to contradict that.
It will also be possible to have user-level granularity on the website, helping to spot fraudulent activity.
Again this contradicts "The Gatekeeper [...] has no user-level information." And I'm not sure what "the website" refers to here, but if it's the publisher site then I don't see how to reconcile this with "Web sites cannot learn the interest groups of the people who visit them" — one of the two original privacy goals.
There was a problem hiding this comment.
Everything here is advertiser side.
The advertiser is the one getting conversion information, not the gatekeeper. The advertiser can use the click_id in the granular reporting to join to first-party user-level data gathered on his domain.
I will edit the section to make it clearer.
There was a problem hiding this comment.
OK sure. I've opened WICG/sparrow#16 to talk about my privacy worries here.
|
OK, looks good to me. |
…sal column.