Feature request: include Chrome-facilitated testing label in trusted server requests

[rahulkooverjee-google]

There's been prior discussion about the need for coordinated experiments between all parts of the ad stack - including contextual, on-device, and trusted server (e.g. issue 191).

To make it easy for the ecosystem to achieve such coordinated experimentation while using the Chrome facilitated testing labels for diversion, it'd be helpful if Chrome could share the label as a parameter in requests to buyer and seller trusted servers.

Would that be possible?

[dmdabbs]

Agree, and it would have been nice from the start for the label header to have been sent on all requests without new APIs or magic cookies. But no soup for us from the explainer...

Accessing labels involves accessing information stored on the user's device. In some jurisdictions (e.g., the EU and UK), we understand that this activity is analogous to the use of cookies and thus accessing labels likely requires end user consent. Before you begin requesting labels, we recommend that you seek legal advice as to whether this consent obligation applies to you.

[dmdabbs]

Perhaps Chrome would consider for the duration of the testing period to pass along a special reserved key placed in the TL auction by the RAA() invoker (obtained with adequate consent, of course).

[rahulkooverjee-google]

Hmm...I'm certainly no expert on EU privacy law, but my understanding from these FAQ is that using Privacy Sandbox APIs also involves accessing data stored on the user device and therefore would also require the same consent as accessing the testing labels, meaning that a trusted server request (potentially including a traffic label) as part of the PA API would only happen if there is user consent to access information stored on the device.

[michaelkleber]

Hi folks, thanks for the request and the discussion. We certainly want to encourage the use of the Chrome-facilitated testing labels, and I see how this request is important to everyone being able to make good use of them.

Of course we will need to talk to lawyers and the folks who worked out the Chrome testing label story. But my initial reaction is that this sounds very reasonable — and that Rahul is right, the requests to the Key-Value servers are indeed only made in the case where a party is already storing and retrieving Interest Group data.

[dmdabbs]

Et voilà: https://bugs.chromium.org/p/chromium/issues/detail?id=1512917

[lbdvt]

That's a feature we would also be interested in at Criteo.

[ajvelasquezgoog]

Hi everyone, the Chrome team is targeting to implement this FR for the M123 release

[dmdabbs]

Appreciate the update. Too bad it was reverted, but approvals are approvals, eh? 🤷

[piwanczak]

Would it be possible to also send the labels in other types of reports: reportWin/result, possibly also ARA reports? This addition would potentially help with testing

[dmdabbs]

+1 to @piwanczak

If this if greenlighted for trusted signals as posted yesterday and the justification follows Michael Kleber's rationale,

...the requests to the Key-Value servers are indeed only made in the case where a party is already storing and retrieving Interest Group data.

then the other requests should be considered to also receive labels.

[dmdabbs]

Hello @ajvelasquezgoog. What are your thoughts?

[brusshamilton]

@JensenPaul I think we can close this issue now.