Hacker News Reader

Comment by Doctor_Fegg | Hacker News Reader

I'd add nuance to this:

Ban _involuntarily_ targeted advertising.

I am 100% ok with Google, Facebook etc. knowing that I live in Oxfordshire, UK; that I love cycling; that I play the church organ. That's who I am. If they want to target cycling, classical music, and local ads at me based on this information, that's great. I would happily log onto these sites and click checkboxes based on this (or do it via browser settings).

I am not ok with them building up a detailed personal profile of my anxieties, day-to-day worries and preoccupations. Nor with them selling my information to other sites to which I haven't explicitly volunteered the information.

View on HN

I don't think this will work well.

If you give them that leeway, they'll claim everything is voluntary.

"Oh, you can go into [random place in settings that changes every week] and disable each one of your anxieties. It will remain disabled for 72 hours"

I think the point is no data collection: only targeting based on stuff you explicitly enter on a clearly indicated form.

Right, but then like "accepting cookies" or ToS/EULAs they'll just make explicitly giving them this permission a requirement for using their application, and we're back to where we were before.

I don't mean you put in a form like 'collect my age', 'collect my occupation', etc. I mean there's a form which you can enter your age, occupation, etc. If it's made mandatory you can just lie.

Why not both

1. ban targeted advertising as it exists now

2. if the user thinks they are missing out on recommendations or personalized deals, they can visit a website, enter their personal details, and the website will respond with ads.

We can really have it both ways.

This is fine in theory, but in practice you'd visit some site with an opt out cookie that gives auto approval just by visiting said site, who would then sell that to one of the huge adco's, and nothing would change.

Adverts pollute our lives in many ways, it's past time we stopped them IMHO.

these questions are in earnest, not meant as a put-down of your position.

how would you draw a line between what's okay to collect and what isn't?

do you think a line could be drawn that would work for most people?

would sites offer different levels of invasive tracking, and the user opts-in as far as they want?

what incentive would users have to opt-in at all?

I think the line drawn would be easy: they cannot collect the data themselves, they can only use what you fill in on a form they present. Then the line is basically whatever you feel comfortable telling them, knowing it will be used for advertising. The incentive would be the standard reason ad companies give for targeted ads: given there are going to be ads, people would generally prefer they be relevant.

This is fine on the surface, but still has the potential for problems. Eg., if I were Muslim, I might be quite happy to declare that I'm interested in Halal food and would like to see advertising relevant to my interests. Win/win, right?

Now, let's say I own an apartment block and want to advertise for tenants, but I don't want any of them dirty muslims in my building. I can target my advertising to exclude people whose interests include Halal food.

Or, let's say I'm the propaganda department of an adversarial nation state, and I want to stir up communal tension. All I have to do is find people with an interest in Halal food, and target them with ads exhorting them to join other Christians in the next Crusade against the Moors. Or whatever. This was actually a common technique in the 2016 election: identify voters who, via their social connections and profile interests, are likely on the fence about Clinton vs. Trump, and also are a bit homophobic -- then target them with "Queers for Hillary" advertisments.

It's shocking how easy it is to weaponize this stuff.

i like that, and i like using the advertising industry's standard PR line to justify it.

I don't know about the rest, but for incentive to opt-in, I could see it as a way to keep up to date on new products. I'm starting to get into woodworking, and I've voluntarily signed up for advertisement emails from some retailers because I want to know when something I need goes on sale, or when a new product comes out that would be useful. If I could pick which advertisements I'm targeted with, I'd be happy to turn off my adblocker, but since the current option is either block all ads or have pervasive tracking for targeted ads chosen by an ML algorithm, I choose no ads.

> I am not ok with them building up a detailed personal profile of my anxieties, day-to-day worries and preoccupations

Luckily no company cares about those things...enough to buy the information. Theres a dissonance between people knowing ad targeting is basically useless and them imagining it is super subtly effective. Ads arent effective at leveragong purchases. The data is conclusive. Yes the brands might appear in your head when faced with the monumentally unlikely choice of not knowing ANYTHING about various products in the category. Its far more likely you pull out a cell phone and check reviews or your family/friends recommend something or you will do a comparison of which has more shelf space or you will read the labels. More lkkely all of these things, none of which is "I saw an ad on twitch". In 15 years of digital advertising, this idea that demographic data (one of the pillars like geolocation), does anything to influence the market. With the corporate magic of attribution* the smoke and mirrors continue. *sales to people who saw an ad causes the adtech dept to be given credit for part of the revenue, as if the ad did something. No science that isnt decades out of date, just dissonance.

This is the direction we need to head in. ML/AI peeps dont wait for regulation. Push back on misguided attempts from mgmt and point them in different directions.

I have had experiences with execs running large video services for content producers/publishers who would openly say (and therefore influence underlings and the next gen) - we know there are people who are addicted to our platforms, they have nothing better to do etc so how do we get them to consume excess Ad inventory we are holding? Can you use AI to improve our identification and targeting of such groups?

My initial reaction to such kind of requests was to say thank you no and walk out. But over time I realized they just find someone else and stay happily in denial of responsibility. Its not enough to then call them out. The only thing that works is to point them in a different more constructive direction. Feel free to make wild claims about how profitable its going to be, if you are dealing with mindlessly ambitious douchebags who don't understand any other language.

Maybe it's just me, but I have this vague feeling that feeding total bullshit disinformation to management in an effort to redirect substantial amounts of energy and resources towards your direction of choice might not always work out well.

Unless management is genuinely mindless, it seems to me like it might work no more than once.

Perhaps you've had different experiences?

I wish I could say there are rosy outcomes when dealing with hard decisions. It's always a messy fight. How messy is a good signal to decision makers of how much support or resistance there is.

If you want voluntary targeted advertising, then read sites or sign up for newsletters that cover these topics. These sites can show you ads that are relevant for their audience, without having to track your activity.

Sites that are not niche specific (like newspapers) shouldn't try to microtarget each user. They lose control of the ads they show near their content, making them less relevant. I'm tired of reading news about the coronavirus crisis surrounded by Amazon's offers on some items I visited a month ago. I still think ads are better when they are somewhat related to the content you are reading, rather than to the user's past activity on other websites.

Targeted advertising is not the problem. The problem is it motivates them to collect too much data about you. And once they have it they can misuse it or give it (voluntarily or not) to someone else. There already have been cases when Google reported an innocent man to the police so he was put in prison.

Parent comment seems to suggest the same thing.

Ban the collection of data for targeted advertising. Instead the only data that targeted ads can use would be provided by the user voluntarily. Not collected. No checkbox for consent.

If user wants the advertiser (say Google) to know something - they need to enter that in on a form.

Reply on news.ycombinator.com