Closed Gunni closed 2 years ago
The purpose of this repository is to explore ways in which we can support interest based advertising in a way that's significantly more private than today's practices. Please keep comments on-topic.
@gunni General discussions of the overall advantages/disadvantages of a particular proposal can be brought to the Private Advertising Technology Community Group (PATCG).
PATCG is a community group that anyone may join, free of charge. As a W3C Community Group, all work and communication within PATCG is covered by the W3C Code of Ethics and Professional Conduct. (The group could use more representation from greater variety of web stakeholders. and more participation from people who make and use independent web sites would be welcome.)
The purpose of this repository is to explore ways in which we can support interest based advertising in a way that's significantly more private than today's practices. Please keep comments on-topic.
All I am doing is suggesting not implementing such an API, or any API that specifically facilitates interest based advertising.
I consider "interests" private and being targeted to be a violation of privacy, with the exception being context based ads, which are only dependent on the content of the currently loaded webpage.
I only now saw your charter "to to incubate web features and APIs that support advertising while acting in the interests of users, in particular providing strong privacy assurances" and now see that my suggestion goes directly against your charter, so of course you are against it, one of you being from Google, and the other from an ad management company.
Unfortunately, the best information we have is that most web sites lose somewhere between 50% and 70% of their ad if the only advertising is contextual. See e.g. https://twitter.com/garjoh_canuck/status/1318989360407236609. Moreover, news sites are hit much harder than other types of websites. So this proposal, and this CG, are working on alternatives to the Thanos Snap outcome for the web.
Interest group based advertising has two main effects on the web advertising market.
Additional incremental ad revenue for publisher sites because of increased advertiser knowledge of audience composition, similar to the effect shown by Johnson et al. for the third-party cookie. (mentioned by @michaelkleber in previous comment)
Leakage of ad revenue from high-engagement, high-reputation sites to lower-reputation or brand-unsafe sites where the same audience appears to be available. (some recent examples: Ad-funded Disinformation on the Ukraine Conflict, Google is forcing everyone to fund Kremlin propaganda right now )
The big opportunity for this CG is in maximizing the first effect while limiting the second. I can create a separate issue to cover the brand safety and revenue leakage problem.
Have you considered just ... not supporting that use case.
Pretty much all users are against tracking. Those that are not, are simply unaware of how it works, and when explained to them they are against it.
There is a lot of laws in recent years because tracking is too ubiquitous (GDPR, CCPA, et al).
Anyone asking for any kind of tracking does so only because they profit from it (directly or indirectly).