Get to Know the New Topics API for Privacy Sandbox

I think it's safe to assume that we are trending towards: people caring more about privacy, browser fingerprinting getting harder, VPNs being standard (e.g. thanks to Apple), and third party cookies for tracking are either prevented by major browsers or governments have legislated against them.

In that situation, at the end of the day, ads will still be here, and targeted ads are necessary for ads to be useful for both user and advertiser.

These measures, where your user agent says what ads would be useful to you, seem perfect to me. For the average user, they give just enough info to ad providers to not be invasive, and for someone who absolutely does not want that, they could likely turn an option on to give random topics.

Right now though, one can argue that this actually gives more data than before (i.e. tracking + topics instead of just tracking), but try to think about the situation in the future.

(I'll still block ads.)

First, the new Permissions-Policy: browsing-topics=() (or the old FLoC Permissions-Policy: interest-cohort=()) will ban this functionality, which is nice if you don’t want to leak your user’s data to Google for advertising purposes. I just checked and my site is still sending it so nothing to worry about here.

Secondly, the initial list of topics can be found at https://github.com/jkarlin/topics/blob/main/taxonomy_v1.md . “/Finance/Grants, Scholarships & Financial Aid” is interestingly scary as a target group. They’re also weirdly American (“/Finance/Insurance/Health Insurance”, putting reggae under “World Music”, etc.)

Google's definition of what is a browser absolutely disgusts me. Please Firefox, Waterfox, Whateverfox, stay in the game for as long as you can because when you die, a piece of me will, too.

And that's where I stop reading. Now the software installed on my own devices is going to provide third-party sites with information about myself against my will?

'Thoughtfully curated' by whom? The user or Google? This seems like it's less about privacy and more about advertising. I wonder if it would have been more honest to call it Advertising Sandbox?

For instance, I think that if, by legislation, the publisher was allowed to send to the ad network three interest codes AND NO OTHER INFO, I'd be ok with that. I'd also be ok with a minor degree of localization (dma or metro).

The publishers could pick their three codes for all their users or, if they had some way of distinguishing "Sports Fan" from "Gaming Fan" IN THEIR OWN DATA they could send that (e.g. reddit would know my top three codes based on my subscriptions).

Then just eliminate tracking and monitoring, again by legislation.

Of course the list would have to be short, general and non intrusive.

In one fell swoop all the tracking cookies go away, advertisers get a single targeting method. The ad tech industry stops eating the world. Content becomes king again.

Google being in control of the codes, using their trove to choose the codes (in a suitable non-transparent mannter) while also reserving all that data for their own use is a huge problem.

The end result is still terrible. The only upside I can think of is that if this mechanism does find use in browsers, it could allow users to specify what ads they want to see rather than have corporate guesswork shove that stuff down their throats, and it could allow Chrome forks to fake this information.

It's sad that the only real alternatives are Firefox, which has strange and unfathomable priorities, and Safari, which is held back by Apple's desire to make money off the app store (and their own bland of technical incompetence and bad policies). There was a brief moment where Microsoft Edge seemed like a great alternative to Chrome, with its new Spartan engine, but the past few months Microsoft has been ruining the likelihood I'll ever open it again with their terrible anti-features and even an integrated lure into debt.

It seems Google has the same problem: they can't understand that I know what I am interested in, and that is not "travel and transportation", but it might be reasonably priced used cars and that trip to Estonia.

What are the odds of other fingerprinting and tracking techniques being dropped though?

This recent push of functionality that will make advertising look more private on the surface, another example of which is Edge's upcoming "Transparent Ads" [2] (whitelisted ads for trusted advertising partners), all of it merely prepwork for the day they sabotage ad blockers. The obvious next stage for them, after having successfully misdirected the narrative towards privacy being their primary use case.

Google must hate human beings, the shit they're pulling lately makes it obvious. So, Roko's basilisk... is this how it begins?

1 https://gdpr-info.eu/art-22-gdpr/ 2 https://techcommunity.microsoft.com/t5/articles/introducing-...

We had FLoC, where the browser determined topics behind the scenes and users had zero control over anything. Now we have "topics" or whatever, which is still FLoC, it's just that now users can view their topics and turn them off, which is better -- but the browser is still determining all of your information in the background and auto-sending this to lots of websites. They still don't understand what the criticism was.

----

I've made this same comment multiple times (https://news.ycombinator.com/item?id=28212558). I really do feel like there is something deeply systemically wrong with how Google thinks about user agency in regards to data/preferences.

I want to choose how I present online, and I want to be able to make that determination differently depending on the context. It's like saying that I want to choose what clothes I wear, and Google is saying, "okay, fine, we'll obviously still dress you, but if you don't like your shirt, you can tell us to change it, or you could even wear no clothes at all." No, get the heck out of my closet, I am capable of putting on my own clothes.

I don't want to just be able to hide some information about myself, I don't want to make a choice over whether or not I present at all by turning the feature on/off, I don't want to exclude certain categories that I'm embarrassed about, I don't want to have a single one-size-fits all identity that goes to every website, I don't want my browser to try and guess what my identity is, I don't want a browser-maker deciding what categories are and aren't sensitive to me, and I don't want to be forced to justify how I present or to guarantee that every piece of information I give to every advertiser is accurate.

I want to choose what (if any) persona I send to a website, on a website-by-website basis, and I want control over how that persona gets built, and I do not want that persona to change behind my back.

I feel like there is some huge conceptual divide here, but to me this is not a hard concept to grasp. The Topics API provides additional safeguards around FLoC, but does not change the fundamental nature of what FLoC is. It does not give me control over how I present online, it doesn't take Google out of the equation.

And increasingly it seems that Google is just totally incapable of building a feature that gives me autonomy over anything without them stepping into the process. When Google builds a dressing room/wardrobe, they need to be inside the dressing room while I change. They have to be a third wheel in this process. They need to have some pointless AI process running to make sure everyone knows their engineers are smart, or some pointless fiddling phase where my interests change every week so that everyone knows that Google was important to this process. I hate that they are incapable of stepping out of the dressing room and letting me heckin dress myself.