Private Click Measurement

[ehsan]

Request for Mozilla Position on an Emerging Web Specification

• Specification Title: Privacy Preserving Ad Click Attribution
• Specification or proposal URL: https://wicg.github.io/ad-click-attribution/index.html
• Caniuse.com URL (optional):
• Bugzilla URL (optional):
• Mozillians who can provide input (optional): @ehsan, @englehardt, @TanviHacks, @bakulf

Other information

We're interested in this specification in order to support a way for a click source to measure conversions as a result of the user clicking on an ad without sharing user-identifying information with the click source in the third-party context.

[marcoscaceres]

This serves as a fairly good explainer:
https://webkit.org/blog/8943/privacy-preserving-ad-click-attribution-for-the-web/

[dbaron]

Do you think this is worth prototyping, important, or something else?

[marcoscaceres]

This seems "worth prototyping", specially if Apple is doing it in parallel. The privacy preserving aspects make it pretty aligned with Mozilla's mission, but it will be highly disruptive to adtech so it might be worth seeing what the large-scale impacts will be on the web and sites ability to monetize through ads.

[hober]

A couple of notes:

1. We've renamed it Private Click Measurement (due to acronym collision: PP-ACA is the Affordable Care Act)
2. We'd like to move it from @WICG to the @privacycg, but need multi-implementer interest to do so. If Mozilla is interested, please say so in Adopt Private Click Measurement as a deliverable privacycg/proposals#1

[hober]

On the fraud question, one possible route would be to use something like Privacy Pass #261 / the Trust Token API #262 to prevent / mitigate fraud.

[annevk]

It's not entirely clear to me we should endorse this quite yet as it fundamentally depends on some kind of fraud prevention protocol that isn't specified. And the ultimate nature of that fraud prevention technique has itself privacy and security properties that we need to consider.

I believe from the latest Privacy CG meeting that steps are being taken to get more concrete on that and I would suggest we wait on that.

[dbaron]

Does that seem likely to happen relatively soon (i.e., we should just wait for it), or do you think it might take a while and we should mark this as defer in the meantime?

[ekr]

I tend to think defer

…

On Mon, Jun 1, 2020 at 11:12 AM L. David Baron ***@***.***> wrote: Does that seem likely to happen relatively soon (i.e., we should just wait for it), or do you think it might take a while and we should mark this as defer in the meantime? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#161 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAIPLIODMUYIVKNOB5BJDTDRUPVRHANCNFSM4HOYMBNA> .

[TanviHacks]

We are trying to get some more specification on the proposed fraud preventions, and hence need some more time to sort this out.