Hacker News Reader

Comment by woeirua | Hacker News Reader

I think this speaks to the power of defaults. 96% of users have not turned this on because they didn't even know that its disabled, and when an app asks you if they can track you, the response is going to be a strong no. The reality is that FB's entire business model has been predicated on the default being that tracking is allowed.

If this holds, this will be cataclysmic for FB. You can't have a huge portion of your wealthy user base just disappear and expect advertisers to keep paying what they were paying previously.

BTW, this is also why the covid exposure notification apps failed miserably. They should have always been opt-out.

View on HN

"I think this speaks to the power of defaults."

Even more, it speaks to the power of removing choice. When app tracking is enabled by default it effectively removes choice because most users never touch settings (and may not even know they exist). Therefore users are not choosing anything. Someone else is choosing for them. "Tech" company employees will respond to criticism of their employer's behaviour with something like: "But users can choose to turn this off." That is not relevant if 96% are unaware of this "choice". The question is: why is tracking on by default. That was a choice the user did not make.

Only when faced with a screen asking the user if she chooses to be tracked do we get a chance to find out what a user would choose. And even then, developers will use "dark patterns" to manipulate the decision-making process toward a self-serving outcome.

Perhaps we should ask ourselves whether there should even be a setting to enable tracking. If the majority of users would choose to turn it off, if they were presented with the choice, then why even have it. This is essentially "privacy by design".

This is no different than when developers argue to HN that it is not worth catering to the preferences of privacy-conscious or minimalist users since those users comprise such a small percentage of users overall. By the same token, it should not worth catering to the small percentange of users who want to be tracked.

Depends on the tracking, IMO. Selling data to advertisers is a hard no, but as an independent developer, having access to crash logs and usage statistics is vital. Not only is it valuable to me, it provides legitimate value to my customers because it allows me to fix problems that they are experiencing or enhance the features they use the most.

Still shouldn't be on by default, but asking up-front seems perfectly reasonable.

If users don’t care enough to see the options available to them, it’s on them.

I feel like this isn’t respecting your user’s time. I’ve converted to opting out of relationships like this where possible; it’s not healthy. I wouldn’t enter a personal relationship on these grounds, I don’t know why I’d do it for some company.

A service provider makes one decision and it impacts N users. A user consumes from M service providers. A user evaluating whether a service provider is trying to “pull one over” on them in the TOS costs M units of time, 1 unit per service provider. A service provider deciding to choose the cow path as the default saves N units of time and costs the service provider 1 unit.

I engage with service providers that respect my time. If I suspect your default is user hostile, and provide a “we won’t exploit you in this relationship” as an option, I look elsewhere.

This breaks down in places though. Example: I haven’t found a communication provider that doesn’t default to building a profile on me and selling it to anyone willing to pay.

The lack of choice for me makes me wish a governing body would intervene and give me a marketplace where I didn’t have to waste my valuable time making sure the person I was engaging with wasn’t going to pull one over on me.

> BTW, this is also why the covid exposure notification apps failed miserably. They should have always been opt-out.

If Google or Apple remotely installed a hastily built tracker controlled by a government agency on my device I'd be pretty miffed about it.

Except it wasn’t hastily built, and the tracking isn’t controlled by a gov agency in the nefarious way that you’re implying.

I've seen all sorts of solutions put together by municipalities big and small from counties to states to countries and everything in between during this whole Covid thing and the thing that struck me is how easy it was for them to essentially track you with perfect precision and tie it back to a phone number.

> Except it wasn’t hastily built

They were built very hastily with little oversight with iterative beta cycles in plain sight.

> nefarious way that you’re implying

I didn't imply any nefarious intent what so over. All I implied is that your location data is being tracked and put into some database somewhere and you don't control it.

Who has this data? No idea. How will it be used a year from now? 10 years from now? No idea. Does it draw more energy on my device or use more processing power? No idea.

That alone is enough to not use it.

When you say Apple or Google installed, I assume you’re talking about exposure notifications that they built. I work at Apple and witnessed it. Nothing about it was slapdash and privacy was #1.

In this system, it’s not easy at all for government agencies to track you. That was very explicitly part of the design, as is shown in all the public docs. It’s not like how it works is a secret [1].

Your location isn’t being tracked and put into a database. Nothing about exposure notifications uses location.

[1] https://www.apple.com/covid19/contacttracing

Here's one example: https://play.google.com/store/apps/details?id=org.alohasafe....

No idea if it's using the official API or what.

I honestly just don't believe that it can work well enough in some places. Especially here in Hawaii where gatherings are on a beach and people leave their cellphones in their car or backpacks or whatever. I was even recently at a camp where people were sharing a blunt and there wasn't a single phone in sight cause it was out of signal range. Could have easily been a super spreader event and the tech solution would have failed.

What are you even trying to draw attention to here? This appears to be using the EN framework, which suggests it wouldn't have been approved for distribution if it collected any personal data or location information.

I get that you have an axe to grind here for some reason but I'm not even sure what point you're trying to make.

Exactly. It's not like Exposure Notifications are meant to solve every single spreading case out there -- many people aren't even opting into it. But we need all the help we can get, so why not chip away at the problem?

I have no axe to grind at all. I'm just saying why I personally don't use it. I don't really have an opinion one way or the other. Please don't be so angry with me.

What you are 'just saying' is FUD. You don't know how the system works yet you keep coming back to this thread to defend your incorrect assumptions.

And this is exactly why the apps failed: even HN readers don't know how they work.

And that's really a shame. It has the power to make an even larger dent in the epidemic. This is the cult of ignorance at play, where mistrust of all institutions (heavily fueled by need to get clicks to serve ads) means we can't have nice things.

I’m not ignorant just because I don’t want to be tracked, and slinging insults like this just reaffirms my beliefs :)

You are ignorant though. The whole point of EN is that you CAN'T be tracked.

Yeah, that’s the claim, but it conveniently ignores all the metadata that ties a particular device to certain places/times/events based on the other rotating IDs they’ve seen. It’s like Nintendo 3DS StreetPass but bad.

How do you ensure you're not tracked? That your location history is not sold to every bidder?

It's not an insult directed at a single person. I'm echoing Isaac Asimov's critique of American culture: https://aphelis.net/wp-content/uploads/2012/04/ASIMOV_1980_C...

There's some truth in what you're saying, but mistrust of institutions is completely rational in light of history. Empirically, ignorance in this context would be unskeptically trusting institutions.

That it strays into the irrational at times is as human as the inevitable corruption that fuels the mistrust.

For someone without full knowledge of the technology to assume (in this case wrongly) there's a privacy issue is entirely practical. It's an assumption that will be correct far more than it won't.

I really like what you touched on in this.

People are responding to fear of past president. One could argue that it is highly unlikely that it will happen again. One could even PROVE that it is NOT happening again NOW - but you are still dealing with someone responding out of fear.

I feel like we are seeing the exact same phenomenon with (SOME) people who respond out of fear of catching the virus.

I would be really interested to learn more about why some people 1) are fearful of government tracking 2) are fearful of the virus 3) aren't fearful of either.

Personally I've watched my self move between all 3 over the past year - it has been a WILD ride

It isn't fair to claim they were built with (1) no oversight, (2) hastily, and (3) out of plain view when you haven't even looked into how it works. AFAIK none of the implementations use location data.

In fact, Apple and Google won't accept apps that use both exposure notification and location APIs.

Your choice to be ignorant about the contact tracing frameworks is your own, but please stop spreading misinformation: https://www.google.com/covid19/exposurenotifications/#exposu...

There were many locations who built their own apps that did not follow this privacy model, or used it incorrectly, but for most cases "your location data being put into some database" is not accurate.

> There were many locations who built their own apps that did not follow this privacy model, or used it incorrectly

We agree then.

You mentioned Apple & Google remotely installing something, so the only logical conclusion is that you're talking about Exposure Notifications (which btw are opt-in, not opt-out, so remotely installed is a mischaracterization here). If someone built their own solution and put it on the app store, then it doesn't really have much to do at all with Apple or Google outside of being distributed on their platforms, and certainly would never be remotely installed (at least on Apple platforms... I have no idea what Oppo might do in China for example, but whatever they'd be doing would likely be because some country law dictated it).

No we don't, because the systems you actually specified are known for a fact not to do the things you alleged that they do.

There is no amount of mental gymnastics you can do to turn “app that tracks when yours and other phones are together and does <something> about it” into something that is inside my moral boundaries.

Then I'd suggest you're not being reasonable at all, because the way it works is entirely consistent with a privacy-first methodology.

That <something> by the way is to take the database that's only on your phone (and never uploaded) and show only you an alert. That's it.

If this is immoral, but letting covid spread around the world and have people needlessly suffer because you refuse reasonable interventions, then I don't know what is moral.

> All I implied is that your location data is being tracked and put into some database somewhere and you don't control it.

This is patently wrong. Exposure Notification apps (those using the official APIs, anyway) have no access to location APIs. The absolute worst they could possibly do is to log your IP address when you reach out for new data, but this is also true of every app you use, and every website you visit.

You don't really seem interested in understanding how they actually work, but if anyone else cares, https://ncase.me/contact-tracing/ is a pretty reasonable explanation.

> They were built very hastily with little oversight with iterative beta cycles in plain sight.

The amount of data that the Covid tracking API of Android and iOS makes available is precisely known. Germany's and a few other (https://joinup.ec.europa.eu/collection/open-source-observato...) are open source.

> Who has this data? No idea.

The users on their individual devices.

> How will it be used a year from now? 10 years from now? No idea.

It will be all but useless.

Except if you're in Singapore. Their tracing app morphed into a spy software for law enforcement: https://twitter.com/kixes/status/1384566617330229250

I’m assuming you didn’t read the white paper or anything about the implementation. No location tracking is happening.

You’re assumption about that and the full software implementation might be far apart. You can’t trust anybody with stuff like this as it will always be misused and abused until the perps are caught. The value of source code access is honesty.

If you have an iPhone and don’t trust Apple, you’re already compromised. Why would they wait for AirTags or Exposure Notifications to track your location when their entire OS is a black box?

That just doesn’t make sense to me.

The Canadian version of the app recently added “secure anonymous usage statistics”, which aren’t particularly anonymous (since all we have is their word that they don’t log your IP), and that aren’t particu secure (messages are encrypted with HTTPS but it’s not too difficult to infer the contents of the message based on their length). And, you can’t opt out of them.

I still have it on my phone (although at least within my house I’m blocking their metrics via pinhole) but I 100% understand why someone would not want it on their phone.

AFAIK they don't gain access to the BLE contact database on your phone, so what exactly is it that they could be transmitting of concern?

What are you able to infer about the contents exactly?

They're transmitting things like "This device showed a 'You have been exposed to COVID' alert". You can see the full list of events they send here:

https://github.com/cds-snc/covid-alert-app/blob/97bb16034140...

But the fact that they could be tracking which citizens have and have not been shown a "You have been exposed to Covid" alert, and there's no way to opt-out, is a little scary.

> What are you able to infer about the contents exactly?

Most days, you are not exposed to covid and you don't interact with the app. So most days, the app transmits a "background-check" event to the server. This means the contents of the message will be the exact same length every day.

As a passive listener, then, you can watch for days where the length of the message is abnormal, and then because there's only so many different kinds of messages that the app sends and they all have unique lengths, you can infer what metrics events must have been sent to the server based on the delta of the message lengths. You can figure out if a user has been shown a "You have been exposed to COVID" based purely on passive listening to traffic from the device.

but it could be used to essentially manipulate you to stay at home if someone went around town and then told the app that they have covid, which is nefarious enough

I hope you don't use Google Play Services, because you really don't want to know what it does.

Except that’s not what the exposure notification APIs did at all. Location is never tracked or transmitted to anyone. What exactly would you have been concerned about?

It is literally impossible to submit any exposure alert without including location data, and then it can reasonably be assumed that all the matched bluetooth device IDs were within the same general vicinity as well. Have you ever used a GeoIP database?

You clearly don't understand how the system works, there is no data related to location involved.

You clearly don’t understand my point. It becomes location data as soon as it’s triggered.

It doesn’t. I don’t know what to tell you. You should go back and re-read how it works.

Sure, let’s see: https://www.blog.google/documents/57/Overview_of_COVID-19_Co...

Right away I notice the glaring omission of “the surveillance state” in the wording of certain sentences like “People who test positive are not identified to other users, Google or Apple“.

“With Bob’s consent, his phone uploads the last 14 days of keys for his broadcast beacons to the cloud”, also giving away his geolocation in the process. My read of the “doesn’t collect location data” is “doesn’t explicitly collect location data”, like just because it doesn’t send a field ‘geolocation’ they can technically say that and not be lying. You should check out one of my very favorite blog posts of all time: https://kieranhealy.org/blog/archives/2013/06/09/using-metad...

This is literally an open specification. There is no geolocation involved here. You're talking out of your ass. Don't you think if geolocation was tracked people would have kicked up about this? It collects randomised Bluetooth IDs, that is it.

It’s impossible to know what metadata is stored, so I have to assume all of it is. Really weird how many people in this community are willing to defend surveillance now. As a former NSA director said, they want to “collect it all”. Why would this be any different? The crux of the NSA story in one phrase: 'collect it all' | Glenn Greenwald | The Guardian https://www.theguardian.com/commentisfree/2013/jul/15/crux-n...

Posting bullshit from 2013 is not going to help you here.

Unless you're using a pinephone they already have.

And even then the default OS on the newer pinephones uses KDE as a DE which had telemetry built in at one point.

I would be miffed about it for the thirty minutes it takes me to find the nearest body of water.

In Israel, covid exposure notification was can't-ever-opt-out.

What happened next was people stopped going out with their phones, because any exposure would result in a Police enforced 14 day quarantine

Its a bit shocking. I heard of a gathering that the police caught, because they noticed a large concentration of mobiles in one place.

Is there anything written about this?

I'm extremely skeptical that a big chunk of people in such a highly online country just left their phones at home. Maybe some of the Hardeim but I would guess 95+% of people wouldn't do this.

Seriously? Why would you be so skeptical of that?

Would you really bring your phone with you, when going out for a beer or coffee, even if it could easily result in you being forced to stay quarantined for 2 weeks?

Is your smartphone really that important to you?

Even though I have personally stayed socially distanced and worn a mask everywhere it was required, it makes perfect sense to me, after seeing how a lot of people have reacted to the pandemic, that a good portion of the population, would let their phone stay at home, so they would not risk a police enforced quarantine.

I do want to know whether I am exposed to COVId and get to be quarantined to not spread out the covid to others who I really take care of. Dont want to be a person that possibly make my family or friends sick.

Most people do. However, the precision of the Israeli system (Based IIUC on cell tower triangulation) was not good, meaning you'd have a nontrivial chance of getting randomly get flagged for 14 days quarantine (with legal consequences for not doing so); Some people were flagged multiple times within 3 months, and it's not entirely clear what the rate of FP was, but it was nontrivial.

I'm aware of quite a few people who selectively turned their phone off or left them at home in some circumstances, but I'm not aware of anyone who really gave up their phone use.

In israel, this was done centrally by cell tower triangulation rather than app/os, so no way to opt out even with dumb phones. However, reliability was not very good - e.g. people got notice that they must quarantine because a neighbour 2 floors up was positive, or because they stood outside a wall of a closed room where a positive person was on the other side of said wall. The appeal process was initially non-existent, then slow and cumbersome - so the risk of being "jailed" for 14 days due to false positive was small but serious.

I don't really get how trends like this one (and IMO that's just the tip of the ice berg – this is companies preparing for the inevitable legislation that will make most forms of tracking illegal within the next 10 years) don't affect facebook or google stock.

This IMO makes two scenarios likely:

1) The market is currently just dragging along, hoping this will all go away and we're really in a major ad-bubble that's about to pop.

2) Tracking... isn't that profitable after all. All the personalized tracking shit can be replaced with a simple "users who clicked on canon printer also clicked on cheap ink refill kits" type of model that gets the same results or better. Which also puts into question a lot of what FB and Google are actually doing.

I wonder if we are not overestimating the value of tracking and underestimating the value that Google brings by forcing some "quality advertisement" from their clients.

Google still has search intent to provide most of the value without tracking anything and it's by far the most effective thing anyways. FB still has their internal profile data.

Why would it affect Google's stock? Google has Android, and is pretty upfront about the fact that they're collecting literally everything you do on your phone. Android still has the largest installed user base, and is still feeding tons of info to FB.

FB's stock prices hasn't changed because we haven't seen if this really impacts revenue to FB. We know that banner ads are extremely ineffective. So if FB ads are still more effective than that, they will continue to capture the lion's share of ad revenue.

Not a huge fan of Google, but have to defend them here.

They have a pretty huge privacy section in android settings where you can see exactly what data they have on you and how to opt out and/or delete existing data.

When you set up android, it ask you if your want in or out. You can even set up a new android device without a Gmail account and disable all Google services including Gmail and search. I don't belive you can do this with iOS.

You can set up an iPhone without an Apple account.

That is mainly used to rescue a device, you can't really use it as your daily phone that way.

(Yes, I know there are some hacks to get around that )

Although, since you won't be able to install any software or use much of the preinstalled software, I don't think it would have much utility...

iOS can also be set up without a Gmail account, no Google services required either.

You are not wrong. But not right either.

All those nice settings (for google and everyone else) are only for the minimum they are required by law. That means anything that they can claim to be "anonymized" is fair game and not visible to you in any way.

...are they? In the US?

One thing I really do appreciate about Google products is that I'm able to fully disable personalization. My Youtube homepage is completely depersonalized, because I went through their privacy settings and switched off everything. I can't say whether or not Google is still secretly tracking me, but my biggest concern is actually that algorithmic recommendations put me in a filter bubble. Google lets me turn off the bubble.

Amazon, by contrast, does not let me depersonalize my recommendations. Neither does Netflix. There was a period of several years where I would periodically go through my Amazon history and manually mark everything as a gift. This worked for a time, but abruptly stopped one day. I assume Amazon decided I'd marked too many things as gifts and my selections should be ignored.

Google gives you the illusion. Ads are still targeted just fine.

Did you receive any ads in spanish or chinese recently? did you see any political ad to a party you vehemently oppose? All that is still decided by your profile and history. Just with hocus pocus anonymization on the front end.

So google is not that different from netflix and amazon. they just give you a blanket.

Google makes 4 times as much on iOS as they do on Android.

And that's because iOS users are more valuable in general. They tend to be wealthier, and have more disposable income.

> All the personalized tracking shit can be replaced with a simple "users who clicked on canon printer also clicked on cheap ink refill kits" type of model

That was how it was done for a long time. Advanced tracking was just a way to try and keep increasing revenue.

Whether it worked or not, only the higher ups at these companies can tell... probably.

And as an advertiser, I confirm losing a lot of money trying to advertise, not seeing revenue. Choosing to advertise 25-34 males (in addition to being the definition of sexism/ageism) who type “Jira time tracking” (because “Jira” keyword is reserved anyway) is useless. I want to put an ad on websites that talk about Jira. And that requires no tracking at all.

It's only a big issue if you try to advertise a specific item on a generic platform; but if you can target websites that cater to your specific business, you really don't need anything more. Of course it means ad networks lose out since they can't just target specific people, they have to only target specific websites or apps.

Personally in the old days of newspaper ad's, I always wondered what the point was of anything smaller than a whole page; what is the point of a 1x1 inch ad in the midst of 200 others. Even today I ignore 99% of what's on the page other than the content.

Point of those ads generally is that people were looking intentionally at them for a plumber, sparkie, piano tuner, gramophone polisher, used tap shoes or what ever else newspapers used to advertise.

In fact many people only bought the paper on ad day.

Whole different kettle of fish.

'users who clicked on also clicked on' is at the core of a lot of the tracking Apple is banning in many cases. This is why Google is pushing FLOC and 'anonymous' cohorts in Chrome.

This happens on every single HN thread on this topic. I'm not a total expert but there is a lot of arm chairing that's wrong.

Tracking is profitable and more so than contextual for a lot of direct response in a lot of industries.

Not in all cases, e.g. usually not for brand advertising. But it works.

For my clients is the direct data I have - and we have extensively tried every single other option. FB is the only platform that delivers strong ROI and we use the targeting iOS blocks. And yes - it's always retorted - I know 100% the donations come from the ads, it's incredibly simple to track even as simple as making a new page or individual product only linked from that ad.

So far this change is definitely effecting our ads and FB seems pretty buggy with the new ads manager changes. But they 'always find a way.' But so far it's not as bad as I originally thought it would be.

Plus you can still upload offline conversion data. most purchases have full contact info. We'll see if Apple comes after that, but they would then have to go after the entire ecosystem including Visa.

Because while FB definitely doesn't like that this is happening, it will certainly cement their position. 3rd party ad services that are less tech/ML-forward are way way more reliant on the ID tracking than Facebook is, and this actually will probably cement them as the only game in town. This is a very complicated situation.

I'll give my perspective which I think is very different than the rest of the HN crowd here in the comments. I think this mainly stems from the fact that I'm probably much younger than a lot of the folks here. I do a lot of random shopping online...

I opted into the FB tracking when I got the popup yesterday because I do FB/IG a lot and I click on interesting ads. Honestly, I tend to buy quite a bit from the ads that I see because they genuinely interest me, and a lot of them are from smaller businesses that I want to try out. If I see ads on FB/IG, at least I want them to be relevant to me rather than be generic.

Which is great Now you have the choice to choose whether you want generic ads or personalised

That’s what people wanted for so long The choice to choose , nothing more nothing less.

Im not sure why this is cataclysmic for FB - they’re mostly a first party advertiser. App tracking transparency doesn’t prevent Facebook from running ads on their own apps and using SkAdNetwork to track conversions (which works even if the user opts out).

Granted the fidelity and latency of SkAdNetwork is a concern but hardly enough to materially impact the bottom line.

FB SDK is basically everywhere (including e.g. Google's Waze app), and the data is incredibly useful to FB to segment users.

While Apple and it's ad business will be the biggest beneficiaries, it will not be a surprise if Google and FB also partially benefit. The clear losers will be all the smaller players.

> BTW, this is also why the covid exposure notification apps failed miserably. They should have always been opt-out.

Absolutely not. Using defaults because you know most people wont' change them to achieve a specific outcome that you, the default setter want, is exactly the same evil. Just because you're doing it for "good" doesn't justify it.

LOL. OK. How many lives could’ve been saved if everyone had Covid exposure notifications enabled during the pandemic? The answer is most certainly not zero. Maybe it’s hundreds or thousands.

How is going to be 'cataclysmic' for FB? They can still serve up ads. The ads won't be as targeted as before, but they still have billions of users. Advertisers will spend money on the platform. Perhaps they'll have to spend more since they can't fine-tune the ads as much.

The ability FB had to target ads directly to people was their biggest value add.

I don’t understand being strongly against one type of default tracking but strongly for another. Tracking is tracking as far as I’m concerned and I’m against all of it.

Is there anything to stop gating features/whole apps behind having tracking enabled?

The Apple policy requires that opting out is not treated adversely. See https://developer.apple.com/app-store/user-privacy-and-data-...

In the old Internet Explorer monopoly situation, Microsoft was forced to prompt the user which browser to use on first startup.

It was not an option of choosing an alternative from the default, the order was random. Same for the default search engine.

If this was setup in the same way i wonder what the ratio would be. With that i mean a fair yes/no choice, not the GDPR-joke where accept is a big green button and reject is hidden in confusing fine print 3 sub-pages deep.

Reply on news.ycombinator.com