Hacker News Reader

You are plainly ignoring the actual privacy sandbox in favor of the clickbait. Google is killing 3p cookies which kills google competitor products. But the sandbox is purpose built to create new infrastructure for the competitors (and google) to rebuild in a privacy focused way. The privacy sandbox is a way to create an alternative to IDs (like google’s sign in option) and 3rd party cookies. You can go through proposals and see how these can work here: https://github.com/w3c/web-advertising

Look, I’m not a google fan but you can’t just ignore half of what google is doing. Even then, you’re missing that they are other ID projects that are closing ground on googles ID.

If the user is using Chrome, and is logged into their Google Account...

What does it do if the user isn't signed in to Google?

Google's engineers would have to have an unbelievable level of contempt for users to propose a feature that claims to improve privacy by signing in to Google.

It is a well known stereotype of programmers that they have contempt for users.

It will probably nag you all the time like it does now for each YouTube video you open without signing in.

> Google's engineers would have to have an unbelievable level of contempt for users to propose a feature that claims to improve privacy by signing in to Google.

I, for one, find that level of contempt totally believable.

>What does it do if the user isn't signed in to Google?

Well, then they just use you to train their fire-hydrant, bus, bicycle, and stair image recognition algorithms.

"It is difficult to get a man to understand something when his salary depends on his not understanding it."

Obviously that just gives Google a monopoly on tracking. So that's cool, I guess ...

Centralization is not always bad. Google is a single company and can be regulated so you can opt-out.

I guess I just don't care who makes money off fingerprinting as long as I can choose that no one makes money off me personally.

I'm leery of any all encompassing systems that you can supposedly turn off with a single toggle. For example, you would think that not having a FB profile is the ultimate opt out but they have all your data as a shadow profile.

The REAL danger of fingerprinting is not on "some company making money of me through ads", it's that basically one or more entities, i.e. companies and/or governments, can identify you, your habits, political affiliation, localization etc, completely. And then they can modify the internet you see, censor what you can say etc. I couldn't care less about the ads part.

this post doesn't deserve downvotes. Googles size also makes them accountable. Decentralisation is useless if you end up in the wild west with 100 companies, none of which can secure your data.

I'd rather be on a well-regulated network that has the resources to have proper security in place rather than on fifty tiny ones who scrape very detail of my life and serve me crappy ads, this is basically how porn-sites work.

> Decentralisation is useless if you end up in the wild west with 100 companies, none of which can secure your data.

It's not useless, nor is it realistic to imply they are all unable to secure "your data" through their narrow lens. Treating "your data" as monolithic, necessarily implies attribution through centralization - all exchanges between multiple parties are at least a 2 way transaction of identity. It's the consequence of decentralization that there are gaps, not a fatal weakness.

The regulation is so far, USA or EU, not very promising. A slap on the wrists and no transparent investigation or explanation how data is handled, will not convince me.

This is an egregious violation of privacy of the unsuspecting Chrome user who conducts their business while signed in to Chrome.

To make things worse, couple of years ago, Chrome started mingling Chrome Sync and cookies for accounts.google.com which meant that signing into Chrome Sync also automatically signed you into all of Google services. The cookies reappear even if you delete them, just because you are signed into Chrome Sync. Now, that same identity (<you>@gmail.com) is not just used for Google's own websites, but also made available to anyone who uses Google Analytics - which is 90% of the web - including your cross-domain and cross-device history. Of course, none of this is new, but I shudder every time at the thought of how much information the user is giving away about themselves.

How is that identity made available to anyone via GA? Google does not allow user identifiable PII in GA.

Are you conflating the aggregate anonymized data in GA with the user level visibility Google has into the raw data?

"...conflating the aggregate anonymized data in GA with the user level visibility Google has into the raw data?"

Isn't that kind of the point? I see no difference. Who cares what individual site owners happen to be able to see if Google is going to monitor, track, send back (telemeter?) way more data and be able to analyze and store it forever?

It's all about who might possibly get the data at a later date than who happens to house it presently.

EDIT: In fact, come to think of it, why are there two things being 'conflated' here? Why is Google collecting more information about Google Analytics-using sites than it's exposing to site owners?

I am not sure if you use and know how Google Analytics works. You are conflating two things here.

In GA4, with Google Signals enabled, the "identity signal" is a psuedonymous identifier unique to your Google Account. I don't know exactly if it's a hash or a join key or what, but the point is it's stable across devices and domains.

It does not directly contain PII. Google can join it to PII, website owners cannot.

It is not aggregate. You can pull the hit-level data from GA, including the identifiers being used.

That’s interesting, Google

> “disable personalized advertising”

On this bit, tracking concerns are not just for advertising though. Extending your reading on how they replace cookies with the Google id, I’d see them replacing ad targeting with other uses of the data anything that wouldn’t straight be labelled as “ad” (scoring, black listing, indirect target matching, etc.)

I'm a Google engineer who interacts a little bit with this kind of stuff.

There are a few settings for this kind of thing. You can take a look at https://myaccount.google.com/data-and-personalization

The "Ad personalization" bit is probably what the parent comment is referring to. But it sounds like you're interested in the "Web & App Activity" bit, which will turn off the non-ads usage of your data. To a certain extent at least, since there are some grey areas.

For example, I'm on the team that sends Google Shopping emails. If you click a button to track the price of a specific TV, we'll still send you an email if that price drops even if you've opted out of "Web & App Activity". But if you've just been browsing shopping pages for TVs, we won't send you a general email about TV deals if you've opted out. Both of those cases are in some literal sense "web activity" but it's still pretty clear what the user expects.

But you might imagine- if you're tracking the price of a specific TV and opted-out of Web & App Activity, should we send you an email if a near-identical TV drops in price? We probably wouldn't, and we don't have anything like this today, but it's not quite as clear. And Google has so many features across different teams, I can imagine there's probably at least one where some privacy reviewer made a different call than what you would have made.

Forgive the grumpy old man approach here, but can't this all just stop.

The price of a TV (or almost any common product) is driven more by large scale factors (cost, competition, features and brand) than the latest 4 dollar shift because of some promotion in some channel somewhere.

We are optimising for the wrong things. Computers are supposed to be our agent in the digital realm, reaching out for us, in our best interests.

If we turn off all storage of personalised information completely and utterly, 95% of everything I ask can be answered from context (please show me cheap TVs).

Look, i don't object to the idea behind the sandbox (it's always been weird that browsers tell the server what fonts and other settings exist. I mean who ever optimised for that)

What bothers me is that it's a google id. Just let me have a few U2F ids - this is my shopping id, track it if you may. when I take it out of the slot stop tracking me.

I am sure I am feeling extra grumpy today, but when we stop tracking and trying to find ways to make me buy, and start finding ways to make my life better, that's when we have a digital revolution.

Yes we need (again) an inversion of control. At first advertising links only tracked origin (website), so announcers could target right website to put links on it. But soon advertisers realized you could gain more granular information in using more active technology like cookies, gifs, flash, etc. So they began tracking users and their journey through all websites with all privacy problems it entails. I think in terms of privacy the only level acceptable is the one you have to deliberately have to engage with to start "tracking" (that means advertising content must be displayed alongside publisher content and not personalized nor tracked through advertising company if the user hasn't engaged with).

I also think that marketing/advertising is bad since it distorts value perception but i remain pragmatic and we can't ban all advertising that easily.

I wanted to expand on this a bit.

If we are ever going to manage the flood of data out there we need digital agents acting on our behalf. The AI behind the facebook feed or the google ad tech is a very early beta of those agents - and "on our behalf" may seem a stretch. But that's not just a cheap snide remark - I think on our behalf is a fundamental issue of how they will be built and designed and regulated.

Richard Thaler has promoted Paternalistic Libertarianism - which seems to me a very good default setting for digital agents working on out behalf. It's not however that simple.

I hope we shall see a Medical approach to managing data - where everything is shaped by the best interests of the data subject. But there are three major world views - let's call them European, US and Chinese for over simplified ease.

The Us view is caveat emptor - one should be rugged individualist enough to be able to calculate the best co-pay arrangement and by extension work out ones preferred facebook privacy settings. If you get it wrong you will be prey for the payday loan industry.

The European view that by default it should be good for the individual (although as a Brit I should sneak in that the state gets to define good. And I suspect the Chinese view is it should be good for the confucian society - and the state may have some say in who is in and out of that.

As we are entering the new era of regulation of tech, these conflicting world views are becoming entangled - and we coders will be on that front line.

Choose a side - and be prepared for some very weird outcomes.

This is replacing one way of tracking with another. Can we stop it with the tracking entirely instead?

I would be happy to voluntarily provide a list of topics I would like to get ads for if I knew that this helps get rid of the cookie and tracking mess we have right now.

Think of a questionnaire a la Netflix that you spend 3 minutes answering to so that it gets your preferences.

The current system we have in place is not just creepy, but also crappy. I see it on YouTube, both the ads and the video suggestions are very poor.

Do you use any signals from people who opt out of Web and App activity to feed into models that are used not just for measurement but for targeting?

What if someone is identified by a model as being in market for a TV and then opts out? Would they still be classified as in market at that point?

I work in digital media, feel free to get technical with your response.

I'll hedge this by saying I can only speak for the teams I've worked on, plus my somewhat limited understanding of company-wide policies.

The answer to both questions is no. If you opt out, your data is not used for modeling or targeting or anything. Perhaps some internal reporting that isn't used for anything other than like PMs wanting to understand user behavior? Even that I'm not sure about.

If you are identified as being in market for something based on activity and then opt out, you will no longer be classified as being in market. That classification will be deleted- though perhaps not immediately but within some reasonable time frame, say 24 hours or so.

And yet there is no way to turn off tracking, because the data will be collected anyways, and will be used one way or the other eventually.

As long as this is connected to a Google account, well, then that's one thing. One doesn't need to agree to those terms.

Google is continuously trying to transcend this. Look at AMP, which is an attempt to run the entire internet through Google for tracking purposes. The strategy is always the same: Promise some sort of consumer benefit under the condition that everything will be tracked. Yes, one can opt out of targeted advertising, but Google still owns the data and uses it internally.

This new play is the exact same thing. As is all other efforts to track people, logged in or not.

Your choice is this:

Have an account and have everything you do tracked by Google (or the next company), give them all your data to use in their algos, and agree to terms that state that they can do whatever they want with the data. Perhaps then you can disable targeted advertising, although that also doesn't really work, does it?

Or, second choice, do not have a Google account. However, then you are subject to subversive attempts to track everything you do and use the data whether you want it or not, illegal or not.

This goes against some basic tenets of law in many countries. My data is my property, not yours. Google is powerful enough to not care, but make no mistake: People working for Google and making these decisions should be in jail. Any malice and hate targeted toward them is 100% justified.

An interesting thing I've noted on disabling personalized advertisements: My SO, who has personalized ads on and lives in the same household, sees ads based on my behaviour. So the ads aren't personalized to me, but they certainly are to my SO.

Depending on the type of ads being served that cross-profile-pollination could lead to info leakage and problematic and/or weird situations. It makes me want to be able to build a fake persona rather than a ghost/anonymous profile

You're assuming competence that isn't there. It's a shared ip address, there is still plenty of tracking that is dumb enough to assume that it means that you're the same person and that gets thrown into her 'unique fingerprint'.

I'm not sure who gets the worst deal from tracking, the companies buying ads or us having to see them.

I can’t seem to find the name, but someone built exactly this: an agent that sits on your computer and generates noise. Visiting random web pages, performing random searches, etc. You’d then get random, non-personalized ads.

"Ad Nauseum" -- I think it's a browser extension.

That said, I have heard a few times comments mentioning that it's easy to filter out (from the ad provider side.) I hope not, but if it is, I would only hope someone smarter than I would send a PR. :)

In other words, they're personalised to your IP address, but since only your SO has personalised ads turned on, they see what's personalised to both of you.

Interesting way of looking at this: monopolies stifle markets and hurt consumers. But consumers of the ad market are not the end-users, they're the companies buying ads. Therefore, by strangling the market, it is actually possible for Google to be helping people for whom that market is a bad thing (which is most of us).

"Google Signals," huh.

That sure doesn't sound insidious...kind of like a large Internet ad company trying to make sure that they're the only one who can track your activity online.

Don't worry, though. If you were using any of the now-dead competing advertiser networks, you can target using this information for a small fee! They sure know how to benevolently push forward the wheel of progress to fight to ensure your absolute privacy from everyone but them.

Not that I'm a fan of these 'Signals'--or even cookies for that matter.

> If the user is using Chrome, and is logged into their Google Account, then you can use that instead of cookies.

How is the Google Account ID shared if not by cookies?

"Oh _these_ aren't cookies! It's just that Chrome sets and sends custom http headers to all Google owned websites. _Totally_ not cookies! _Totally_ GDPR compliant!"

Set-X-Totally-Not-An-HTTP-Cookie: sessionToken=bigiain; Expires=Wed, 17 Jan 2091 10:18:14 GMT

How do you kill fingerprinting, though? It's impossible without limiting what web developers can do with their sites.

Many ways! You reduce the nearly unlimited responses for various aspects of a defined few - such as a basic list of fonts or the nearest screen resolution. You only provide random data unless the user specifically grants fine tuned info. You stop providing long lists of capabilities and instead leave it to the implementation to assume.

Providing a “baseline” user agent instead of a detailed specific version could help for example.

The do intend to kill fingerprinting [and cookies].

But only because they have a far more efficient way of tracking users _and_ make sure they're the only one who can benefit from this.

It's unbelievable how absurdly anti-competitive and anti-consumer google has become, yet nobody seems to care.

I wonder if they are also doing this to get around the GDPR related cookies stuff. Don't need to have a banner if you don't use cookies. I assume this would also apply to sites using Google Analytics as well

GDPR is actually technology agnostic, it doesn't mention cookies or any other technology. If you store extra PII in any way, cookies, fingerprinting, localstorage or whatever you need to ask for consent with opt-in.