- From: Valentino Volonghi <dialtone@nextroll.com>
- Date: Thu, 25 Jun 2020 10:21:58 -0700
- To: public-web-adv@w3.org
Hey, I'd love to see if there's any interest in discussing Apple's announcement from WWDC around IDFA and how that could relate to our discussions. Context around the why is: it seems like having the user agent (in this case the phone) asking the user for permission on advertising tracking, and providing an identifier post-consent, is an acceptable solution for Apple in the mobile space, so I think it would be interesting to understand why that wouldn't work on the web from Apple, Google and Microsoft. Overall it would potentially solve multiple issues with the current proposals, from ability to maintain accountability, to proper monetization of platforms and it would allow the user to control tracking easily. Since an identifier would be provided by the browser, the vast majority of HTTP requests happening on a page from advertising companies would disappear as they were mostly for cookie matching purposes, and these can move server side like they do in the mobile space, saving power, bandwidth and data. The browser could have controls, with the website, around what pages to not send the identifier through, and we could come up with specs that allow the browser to require the tracking to provide transparency in order to continue to provide the identifier (APIs for example similar to those available in turtledove to add users to interest groups). The browser can also rotate this identifier periodically to avoid too much data aggregation, and provided that this exists, we could introduce all of the IP blindness and privacy budgets and standardization of other entropy in the browser to prevent fingerprinting. Is it worthwhile to talk about the reasoning behind not taking this approach, that apparently works for mobile? -- Valentino Volonghi CTO, Founding Team dialtone@nextroll.com
Received on Thursday, 25 June 2020 17:22:23 UTC