Hacker News Reader

Comment by TekMol | Hacker News Reader

I am not sure if this change is really a bad thing.

I don't really know, how the current wild west of online ads works and who stores which data about users.

But if Google throws their weight behind the idea that a common set of informations is used, which is completely under the control of the user, easily accessible and changeable, maybe that is a good thing? Maybe no more "header bidding" which sends the users data to a gazillion companies?

I guess it just needs a browser API call to see what data is stored in your browser? So it is easy to check?

It might also be leveling the playing field. So small websites can get ad revenues similar to big websites. Currently, small websites get something like 1/100th per pageview of what big websites get. If "the user is interested in toasters" is all that is needed to raise ad revenue, maybe that is fresh air to finance the indie web.

View on HN

Browsers simply don't need to cater to the advertising industry. It's extremely unfortunate that the world's most popular browser and the world's most popular mobile OS are both owned by the world's largest online advertising company. It's also extremely unfortunate that this kind of concentration of power is legal.

Catering in what sense? Giving more power to advertisers? Or recognizing that its how the platform is used? By his description it sounds more like the latter.

Browsers are referred to as user agents. They act in favor of users.

This comment appears completely unrelated to mine. I understand you think your point needed emphasis but why in response to my question?

They are currently called user agents, right? And arent they currently utilized by advertisers? Does this change give more power to users, advertisers, both? Its an honest question.

They actually do: https://www.gov.uk/cma-cases/investigation-into-googles-priv...

A big problem that the web has is that it has no standard for micro-transactions. As such, the web is powered by ads or subscriptions. That's a fact.

Of course, we can talk about the "indie web", the one in which enthusiasts often get overworked without pay, and then have to provide support for free to entitled users, while begging for donations to keep the servers alive (e.g., Mastodon, etc). But IMO that's not sustainable. Servers, electricity, people’s time aren't free.

If ads go away, large portions of the web will be locked behind paywalls. You could say that's not your problem, however, you're a privileged individual that can probably afford to pay for YouTube or whatever else you need, whereas the poor can't afford it.

Ads going away also means that your average online publication can only survive if subsidized by a bigger player, like Facebook. Especially smaller newspapers, as there is only one NY Times and subscriptions fatigue is real.

People that want to ban advertisements are basically the NIMBYs of the online world. Fact of the matter is that absolutely all browsers are funded directly by ads, including Safari and Firefox (both funded by Google actually). Brave Software had the audacity to speak against the Topics API, while having plans to also do client-side targeting, but in a way that keeps the results under Brave's lock (data for me but not for thee). Vivaldi spoke against the Topics API while defaulting to Bing and encouraging users to not block ads from their partners. I vaguely remembered that even Firefox flirted with client-side ads in the new-tab page.

The Internet needs valid business models. Wanting ads to go away is like wanting capitalism to go away. And the people proposing it have no better solutions than having the government subsidize it all or just letting it crash and burn.

I don't care if ads go away or not, but I care a lot about the tracking that ads bring. Getting rid of that alone would remove a whole lot of the problem.

Moving the tracking into the browser itself is not a solution. It just makes the browser into an agent against the user rather than a user agent.

This is a good point actually.

What I like about the Topics API is that it seems to put the user in control. And at least in the EU, I noticed they at least ask for permission and with a Reject button. And the dialog is clearer than your average GDPR consent form.

The question is if the user is in control or not. For now I tend towards cautious optimism.

> A big problem that the web has is that it has no standard for micro-transactions

As a user, that's not my problem. And a browser is supposed to be a user agent — acting in my interests on my behalf.

> however, you're a privileged individual that can probably afford to pay for YouTube or whatever else you need, whereas the poor can't afford it.

I can afford it but I have zero subscriptions. Why would I pay for something if I can get an even better experience for free with an ad blocker?

> Fact of the matter is that absolutely all browsers are funded directly by ads

Well maybe we should finally, at last, decide what the scope of the web should be, and stop uncontrollably expanding it into the operating system territory. It helps to remember that a browser is merely a hypertext document viewer with support for macros. Most people still treat it like one.

If browsers could be "finished" without requiring constant maintenance busy-work, maybe they won't need as much funding.

And by the way, the SerenityOS project is building its own web browser from scratch with a small team that runs mostly on sheer enthusiasm (and a whopping $110 000 in donations from two companies): https://ladybird.dev

> The Internet needs valid business models

The commercial part of the internet, which I will not miss the slightest if it goes away.

> As a user, that's not my problem. And a browser is supposed to be a user agent — acting in my interests on my behalf.

I find it absolutely hilarious when HNers pull out a dictionary and act as if the issue is a misunderstanding of language.

The issue is that no way companies will give up the old tracking methods, they are just gonna sandwich together the old with the new and get the sum of the two.

Sites that get really big, usually do "give up" the old way of sending out the users data to a gazillion of advertisers. Because they have the inhouse resources to segment their users:

https://www.reddit.com/ads.txt

Meta, Google, Reddit, Twitter etc won't send out your request data to show you ads.

But unless you are a giant, it is not feasible and sites resort to send out request data to countless data brokers to find out what users are interested in:

https://www.washingtonpost.com/ads.txt

If the users interest is available via the browser, that would enable smaller sites to do what larger sites do. Show toaster ads to users interested in toasters without the data brokers being involved.

> If the users interest is available via the browser, that would enable smaller sites to do what larger sites do.

And how is that a good thing? The problem is the data collection itself. That larger sites don't directly sell or share the data they collect is better than if they did, but it certainly doesn't make everything OK. The tracking is still happening.

Thanks, I think people are not really digging into the implications here.

A lot of that is because, I suspect, that advertising is taken as being evil as an axiom. Ads are evil, therefor anything related to ads is evil. Or, for some, ads that are targeted are evil. It feels religious when I talk to some people - I have spoken with many who genuinely believe that an ad is an inherently bad thing. They resent someone advertising to them in all forms, under all circumstances.

But ads are also a large part of how the internet works. I wrote hundreds of blog posts using wordpress.com for free, and it's because they would put an ad on the site.

So if we step back and ask "how do we allow for targeted advertising while also ensuring user privacy", I think we end up with something like this.

I don't want advertisers to know things like my name, my usernames, the sites I visit, etc.

I am ok with advertisers getting some opaque token with tags like "Programmer, American, Energy Drinks" or whatever. Especially since I can control this in my settings.

Finding a helpful line between helping the right ads get to the right people, thus ensuring the web continues to be free, and also making sure that you're not one data breach away from having your entire persona exposed, seems like a reasonable path forward.

No advertiser needs to know anything about me. They’re not a useful function of society and indeed are deceptive, harmful, and dangerous. Advertising needs to be put on a short leash with a choke collar until we can kill it.

> No advertiser needs to know anything about me.

They do if they want to deliver ads that are relevant.

> Advertising needs to be put on a short leash with a choke collar until we can kill it.

Kinda proving my point. Nothing in your post actually substantiates why advertising is inherently bad, you just state it as if it's true.

"They do if they want to deliver ads that are relevant."

I don't have any interest in them doing so, so this seems to work out.

I'm not interested in being manipulated, nor providing information about myself to make that manipulation more effective.

You're not really engaging with the conversation, which is kind of my entire point about the group of people (like yourself) who say "ads bad" and that's it.

I gave a case where advertising allowed me to write a blog for years, without paying any money. Indeed people make these choices all the time - mobile games are a good example where people will not want to purchase the game for a dollar but are happy to watch an ad every 30 minutes.

Unless you're willing to accept that advertising has some potential value I don't think there's any point in us discussing solutions.

Though I'd prefer no advertisements, I think it's worth distinguishing between targeted ads and those that are simply based on the page content.

If you run a technical blog and run advertisements that fit that theme, I am less adverse to that. I see the tradeoffs there, and can choose to visit your page and see your content knowing the exchange that's being made.

I was responding specifically to your concern that the advertiser would no longer be able to tailor their advertisement to me if deprived of information about me.

(My preference is to directly fund people who make things I enjoy, and I do practice that. I also think most mobile games funded by ads rely on predatory models to be profitable, which isn't really ideal. The whole thing is a mess of misaligned incentives.)

Thanks, so I think we both agree that some advertisements are fine, which is the point I've been trying to make. How do we make all advertisements "fine".

Direct payment would never have worked for my personal site, I could not pay and certainly my dozen readers weren't going to. But that site made my career. Ads allowed that.

As for what advertisers need to continue funding that, while not being awful, I think that is the interesting question.

> > No advertiser needs to know anything about me.

> They do if they want to deliver ads that are relevant.

There's a simple (and much older) alternative: deliver ads relevant to the content being seen, instead of tracking the user. If the user is reading a page about cars, deliver ads about cars (and car parts, and car insurance, etc). If the user is reading a page about My Little Pony, deliver ads about plushies. And so on.

Presumably if a user is reading about cars there's a good chance they already own a car. In effect you're saying "the existing internet can work with lower revenue for ad companies" and to that I am sympathetic.

> They do if they want to deliver ads that are relevant.

Advertisers can still attach ads to relevant content without knowing anything specific about the person who might see it. Consider an advert for a car attached to an motoring article.

> substantiates why advertising is inherently bad

I'll try tackling this. Advertising is inherently bad; it's the source of a great many of our problems.

It's inherently manipulative. It tries to persuade, coerce or deceive us into buying things we wouldn't otherwise. This is a huge driver of obesity and ill-health (pushing UPFs, alcohol, tobacco etc.) and climate change (generally increasing consumption).

By definition, it absorbs our time and attention. It interferes with our enjoyment of sports, written articles, and wastes our time sitting through video and audio adverts.

It corrupts services. It provides an incentive for companies to maximise advertising revenue instead of the best experience - take Google or Amazon, who design search to push ads rather than what users ask for. Or social networks, who use algorithms to maximise engagement instead of positive interactions. Or news orgs, which publish clickbait fury-inducing articles for the ad impressions.

It sucks a lot of our brightest workforce into useless work designing ads and ways to push them. It absorbs a lot of economic activity into zero-sum competition between rivals running to stay still.

I appreciate that, but you're not really talking about what would replace it or how the web would survive without it. Like, the reality is that ads are enabling non-ad content today, so I think it's fine to say "and it's bad for reasons XYZ" but can you not see the value?

> They do if they want to deliver ads that are relevant.

They can insert car ads on thedrive.com, tech ads on arstechnica.com, or topic-relevant ads on subreddits.

This approach appeared to "work" on radio and TV.

> Nothing in your post actually substantiates why advertising is inherently bad

Nothing says "pollute the air" like a "Go Green!" Exxon Mobile ad. Or how about those "flushable" wipes that aren't flushable and cause damage to water treatment facilities, yet continue to be a legal product [in the US]? Ads directly encourage harming the environment.

Why does "hate ads" need some form of empirical reason to rid this scourge of humanity? We all have preferences which may or may not follow a logical path.

Capitalism doesn't need defending; evil is a feature, not a bug. ;-)

I think there's an ecological angle to it. The more targeted ads can become, the more valuable any individual ad display becomes, and the fewer ads need to be delivered to support the content. People making compelling content with well targeted ads can provide less obnoxious experiences (and they will stand out vs other sites that spam the hell out of your eyes). Basically, there's a dynamic at play where when ads are less targeted, you should expect to see more ads and those ads that you see are all the more "annoying" because they're not relevant.

I really am open to whatever this Privacy Sandbox is about but I just haven't found anything that explains it. The popup I saw actually seemed sort of interesting where it's teasing this interaction where your browser is saying "Hey, you seem to be interested in these things, can I tell ads that you're interested in them?" and I really sort of like that approach. But... I don't understand where that list is coming from, how new items get added to the list after I've reviewed it, etc. I do NOT want to have to POLL some preference page to make sure things I don't want in that list creep in. Or maybe I could explicitly add things I know I am definitely interested in from a list? I'd prefer it was more upfront and asked me each time it found a new one or however it works.

So... I actually see a way this could be empowering to users if done correctly. It's my lack of high-level understanding what is going on and my experience that Google seems to have become mealymouthed lately so I have a base level of distrust. It comes off as yet another "hey you need to stop what you are doing and research something Google's doing to you" again and I don't know if I have the time for that and I have the expectation that it's going to be a waste of multiple hours to try and find any useful info from Google itself. So it may be good or bad but really I don't give a shit about learning about things at the level of a browser implementer. In the mean time it's always easy to just say "No" and nope out of Chrome.

Perhaps if Microsoft allows it into Edge then it's at least been filtered through a Google adversary.

Edit: so I decided to take a peek and yeah of course privacysandbox.com is just your typical Google goosechase. "How does this work?" answers with essentially "We don't know! Go see developer.chrome.com or developer.android.com and see if you can manage to find something there!" Literally linking to the top level of those sites rather than anything useful. Fuck you too, Google.

> fewer ads need to be delivered to support the content.

Rhetorical: What company says "new tech? I can do less of this thing that makes money!" (or government, for that matter).

> Fuck you too, Google.

That's the spirit.

Are all ads bad? Probably not, but you'd quickly learn to distust all ads of one out of 10 jumped out and tried to shoot you in the face.

There is just far too much money in trying to hoover up all of your data and sell it off to someone else for the 'bad guys' not to try it constantly. When even governments are telling you to use ad blockers to protect yourself you know the industry has long let the problem get out of control. Users are going to protect themselves.

But this work from Chrome is an attempt at limiting the ability to "hoover up" your data and sell it (ie: this pairs with disabled 3rd party cookies) without killing advertising as a whole.

> if we step back and ask "how do we allow for targeted advertising while also ensuring user privacy", I think we end up with something like this.

That's assuming that targeted advertising is acceptable in the first place. I don't agree that it is.

> That's assuming that targeted advertising is acceptable in the first place. I don't agree that it is.

That was the subject of the entire first half of my comment.

Ads are evil. They represent the efforts of a multi billion dollar industry using the output of millions of dollars of psychological research to manipulate you into buying things you don’t need. The average person has no chance against this psychological manipulation.

Burn it to the ground, and let something sane take its place. Micro payments and its kin don’t work today because ads are easier. Create a space and a need for them, and watch them flourish.

I'm fine with looking for alternatives to advertising, but so far nothing has been successful. People are choosing ads.

It’s almost like there’s hundreds of billions of dollars at play to make people incentivized to push ad ecosystems as much as possible

I think mobile ads are an obvious example of users choosing ads. Mobile apps can almost never actually charge money - people will not pay a dollar to download an app. But they'll happily accept advertisements in their app.

i avoid the app store like the plague because it's awful, but it's basically impossible to find options that aren't ad supported. and the lack of purchasing incentive has a lot of other factors into it as well aside from "i want ads". there's also massive user distrust in paying for apps because they're so often garbage in general

i search for apps these days by googling for FOSS and filtering results to reddit.

Ads are communication. Are you saying that communication is evil?

I regularly see young and naive folks online saying things like "supporting your favorite creators by buying their product isn't capitalism", or "self-promotion isn't advertising". All that cognitive dissonance must really hurt.

Ads are a subset of all communication. So no, communication as a whole isn’t evil, ads are. No cognitive dissonance required.

Nice fallacy though.

OK, advertising is communication "promoting a person, product, service, event or a job opening".

Is that kind of communication evil?

This isn't about ads which I doubt we'll ever get rid of. This is about privacy invading targeted ads (which Google enables as the 800lb gorilla ad network).

Those are evil and unnecessary (and honestly unclear in terms of efficacy these days).

The promise is that Google is some "unimpeachable authority" that preserves your privacy instead of selling you out, when in fact, they are not and they do just that.

That the modern web economy is based on this unethical behavior doesn't make it just or even an ideal economy.

Can non-google use the flock? I really haven't bothered learn anything about how this works but I had the impression Google does some sort of semantic integration on its end that adds meaning to the tracking numbers.

Basically, I would be pretty upset if my browser is disclosing interests in plain text to websites I visit. I think Google's idea is that they play bouncer over the semantics. So I'm skeptical the method can be used by non-Googles.

I also suspect Google will break things often enough that a competitor ad network couldn't really keep up. But that's just my very uninformed read of whatever this all actually is.

Understanding this whole thing had been on the back burner. I am open to something clever (and some things I have read long ago about it hint at clever) and have a similar position you do, but everything I have encountered about this is either super technical or sky-is-falling FUD or flowery PR bullshit.

The pop-up's vague language and smooth-brained cheerful explanations just made me take the easy route of turning everything off in the privacy settings and uninstalling Chrome. I usually use Firefox anyway and all the work machines have Edge, so who needs it. "I don't know WTF this is and don't have time to figure it out. I turned the settings off in preferences and uninstalled Chrome." is how I'm answering questions from friends and family.

Yes. Other browsers can implement FLoC. Any website can also use the browse APIs for retrieving interests.

The 2nd half of this is not correct - only domains whitelisted by Google can use the browse APIs in Chrome. You need to apply to Google with your business's DUNS number to get your domain whitelisted for you website to able to call the Chrome Privacy Sandbox APIs.

Here is the process to apply: https://github.com/privacysandbox/attestation/blob/main/how-...

It's the only browser API I can think of where Google approves, rather than the user (eg: normally, the user is prompted when a website wants to use the browser's webcam or location APIs - in this case, Google handles the approval for the user!)

Why not both? We have to click those dumb cookie things anyway.

Google's pretty fond of privacy things that "hurt advertisers", but hurt all advertisers other than Google harder than they hurt Google.

But this change levels the playing field, doesn't it?

Previously only Google and other big data brokers could figure out that a visitor who looks at a jokes site is interested in toasters. Now every site can.

That's the carrot. The stick is it's the first step in a path to neuter cookies. Which hurts all the advertisers, but Google the least.

    hurts all the advertisers, but Google the least

How so?

Owning the most popular browser, most popular search engine, most popular email platform, most popular ad platform, most popular video platform, etc. They don't need 3rd party cookies. They can mostly reconstruct what was lost.

It takes away more information from everyone else. Widens the gap despite slightly denting what Google can see.

But Google is late to the "remove 3rd party cookies" game. It's not like that was their idea.

They're late because they weren't allowed to, with the most popular browser: https://www.gov.uk/cma-cases/investigation-into-googles-priv...

That's pretty distinct from "They pushed to remove 3rd party cookies to hurt competitors".

I stick by that. It's not the only move where they tried the boiling frogs approach around widening their control gap. AMP is another example. Manifest V3 also. They've become pretty skilled at making a trojan horse that has an outward benefit with the real intent hidden.

That they end up making concessions here and there to some regulatory bodies doesn't do much to convince me it's all about privacy.

> It takes away more information from everyone else. Widens the gap despite slightly denting what Google can see.

Good. It would be even better if it had a greater impact on Google as well, but I'll take whatever gains I can get.

Is it completely under control of the user? Do you get to choose the list of topics to send to websites?

My interest list:

Ad/content blockers

/End

Reply on news.ycombinator.com