Comment by ramoz | Hacker News Reader

Comment by ramoz | Hacker News Reader

Parent Full thread

It's important everyone becomes educated about the fact that virtually every mobile app sells some part of your data that leads to some private company possessing the ability to draw a circle around your house on map and then detect all patterns of life without any PII.

I don't think it's fair to pit this as a US gov't surveillance problem. It's true though - the Government missions involved, where this type of data is relevant, face rather compelling offerings especially in complex times; e.g. immigrant/refugee surges where understanding the flow of people up to the border is important for stability.

All of this data comes from Big/Small/all tech; usually branded as "ad tech" or "mobility data". And the supply chain is rather murky, masked, and rebranded/repackaged numerous times between a network of data providers & downstream businesses.

Working close to Government... I've never seen any mission specifically seeking large-scale citizen-based tracking capabilities. US government in recent timeframes have seemed adamant about not purchasing any US-based location data & are cautious even for non-US based monitoring; especially as they learn more about the origins & scale of mobility data.

These sorts of things are compartmentalized.

The FBI, ICE and DEA absolutely do scaled collection. DEA/ICE was known (ie NY Times article) to be tracking cars via LPR/camera a decade ago up and down I95, for example. Individual cities and other jurisdictions share LPR and camera data. Things like fusion centers and drug task forces probably get access to various forms of intel.

Civilian government is different - they are subject to lots of controls and audit, and usually take data protection seriously.

> usually take data protection seriously

Do we need to list the number of government data breaches/leaks over the past 20 years? My own PII has been in several.

They may take it seriously, but their actual track record isn't really that great.

We need to start facing the fact that PII just isn't private.

It's not necessarily personally identifiable either (considering all the different ways to fake it, make mistakes, the inability to correct mistakes once the information is out there, and the unreliable methods used to verify identity)

Spoken like someone that's never had a partner with impulse control or anger issues. Keeping PII private is, IMO, like freedom of speech or privacy: it may seem trivial and unimportant when you don't need it, but it sure as hell matters to some people, and can keep them safe.

It matters to me a great deal, but I've just come to face the fact that much of what I and the various organizations who collect PII from me consider private is effectively publicly accessible, no matter what lip service to privacy these organizations give.

Our SSN's, names, ages, address, phone numbers, emails, locations, habits, interests, political opinions and affiliations, sexual orientations, medical histories and countless other "private" and identifying information are in countless databases and available to those willing and able to pay for, spy on, or steal them

I think there are some steps which can be taken to help mitigate this:

- use a VPN to obscure rough IP based location lookups and if the VPN has shared traffic it also make tracking more difficult

- limit apps which are useful with geolocation to use rough geolocation

- If on ios, opting out of app tracking will at least disable Apple’s advertising ID.

- Be mindful of granting bluetooth and network access.

For apps which require precise geolocation this becomes a problem (ex: google maps for navigation). If you can, find an alternative with better privacy incentives (ex: apple maps).

For the stronger privacy willed when signing up for a service think about what information is identifying and whether or not it is worth withholding:

- use unique email addresses (ideally with a unique domain or a shared domain; ex: simplelogin)

- use a credit card with a unique number which can make payments under a unique name (ex: privacy.com)

- use a unique name

- phone numbers are a bit more tricky (ex: talkatone)

None of these things are perfect, introduce their own risks, and may trigger fraud alerts but I think they help in some of the situations listed above.

I admit this certainly isn’t realistic to expect of most people.

If you or anyone else have thoughts on this or know of other services, I’m all ears.

I always get annoyed how these programmers can have a good sleep at night given what they have done. Note that blaming it on the marketing VP is not fair. If even 50 % employees have a thought this tracking can be stopped.

The developers may not be aware of the full consequences of what they've been asked to do.

I was recently discussing the Uber revelations in an ethical tech group that I run. The most shocking part for me was that at one time the app was designed to behave differently depending on whether the user was categorised as law enforcement based on their usage history.

This surely required complicity at all levels, from management down to engineers and testers.

But someone point out quite astutely that such a feature can be generically framed as "optimise/adapt behaviour based on historical usage". It makes business sense to categorise a user's profession and alter ride costs based on that. All that's needed then is to give a higher-up the control over a dial that effectively nullifies law enforcement's ability to get a ride.

I can believe that some employees of these companies are genuinely shocked and surprised that this is being done, but few will refuse to do it, and fewer still will quit.

The sad fact is that our surveillance society was built with the willing cooperation of countless developers for whom money was far more important than the privacy of their users.

Framing this as "money vs privacy" is disingenuous. There are many other factors at play. It takes a lot of courage and social/emotional skill to be able to say "no" to a work request in a way that is assertive, respectful, and doesn't lead to becoming a pariah. Consistently doing this in the face of deadlines and incoming requests is a big investment of energy. Switching jobs whenever you find yourself in that position is also a big investment of energy. Not everyone has that energy to spare, for example if they have a young family, are going through bereavement, divorce or moving house, or have a health condition etc... .

if you are that talented enough where companies will continue to offer you $$ to program whatever they say, then yea you do have a lot of POWER and can say NO..

Your financial power is mostly irrelevant to the effort and skill of saying no.

I don’t find it is generally one programmer enabling it. Decimation of privacy often occurs slowly at most orgs. One exception at a time. One “critical” temporary need on top of another. Often in different teams. The person who built the UI didn’t build the GPS modules. The person who built the GPS module didn’t build the data store. The person who built the data store didn’t deal with report exports. And that person didn’t deal with their privacy and compliance policies- or sales pipeline. Lots of people wanting to say yes and do a good job. Often leadership is caught up chasing a dollar

People are very willing to do almost anything against "the bad guys", defined by whatever ideology they were brought up with. Everyone doing this imagines that they're preventing crime and terrorism, and can probably point to a case in which it was.

"How do you sleep at night?"

https://youtu.be/GO0JaecRWy0

Minus the beautiful ladies.

Programmers selling out is the goal in and of itself these days.

It's gotten to the point where we just sorta accept that people want to work at places like Facebook and Google. There's so many of them that the thing to do is treat it like a morally neutral job or be seen as a weirdo.

And when the largest and most pervasive global surveillance systems history has ever known "aren't so bad", then nothing is.

yup..how do we change this

Exactly this. But I think it's similar to the "Matrix". They're part of the system.

Responsibility is still probably more in the product VP realm, but with a few internal "cambridge analytica" shops where the devs/data scientists know exactly what they are doing; however or more jazzed up by their desire to advance in their research domain.

For vast majority of devs, any level, the way data pipes out to different business units from your appliance at a large business is often obscure/unknown. You integrate SDKs and API's that are black boxes, you send data out to warehouses with 30 analytics teams measuring/creating new data products you'll never see.

the programmers sleep at night because doing this is the only thing that lets them feed there families.

like it or not, there are always people who will see this as the lesser evil because of their personal circumstances don't give them other options.

I'd agree if we were talking about low-skill work where people are just scraping by. Programmers have the luxury of choosing from a wide range of places to work. We're all in a position where we can refuse work we find unethical, even if it means taking a pay cut.

Yes, but… I’ve seen that it is often just shades of bad. There are so few morally pure companies out there- they are all willing to do bad things for money. Or the vast majority of them. And it is hard to evaluate that upfront. Even the most virtuous will bend privacy for the right stakeholders

I should go work for Google? Because they clearly value privacy?…

You say that as if we all started out fresh from college with the ability to pick and choose the kind of dev work we do. Not everybody has the kind of safety net starting out that implies the ability to do that.

you are right but then you should work to get to a point where you can and will do whatever you want..you can destroy a system 2 ways, within or create your own

yup if you are that talented enough where companies will continue to offer you $$ to program whatever they say, then yea you do have a lot of POWER and can say NO to whatever you want

True, as long as there's money in it, but I'd bet the majority of people in adtech aren't in terrible positions, so I don't think that's the root of this. Do people generally care about acting ethically without circumstances forcing them to? Beyond paying lip service. And if you think so, do you think they generally care in the context of surveillance, where most of us haven't had first or second hand experience of any obvious cost?

Listen to Jordan Peterson - especially Maps of Meaning lectures and/or book.

Then i start having dreams that my dad is giving me a lecture but my dad is also kermit the frog

Don't fall down that rabbit hole.

https://www.unikumnett.no/2019/10/12-reasons-why-no-one-shou...

C'mon, it's just someone with some opinions you don't like, not the bogie man.

It's not a rabbit hole to hear out someone you disagree with. Unless you're either afraid 1) they might be right, or 2) the listener doesn't have critical thinking skills. I get that #2 is scary, and it's a legit issue in society, but they shouldn't be listening to you either.

Also, there are good tear-downs Peterson's message online, the one you linked is... not.

I used to think that - despite his overreaching when it comes to politics and culture - he was probably a decent enough psychology researcher and self-help author.

However, having been through a fairly significant psychological journey myself and then reading the 12 Rules for Life, I'm quite worried that his rules have the potential to prolong or exacerbate psychological insecurities. Some of them, e.g. "Pursue what is meaningful" and "Make friends with people who want the best for you." are absolutely fine. But there is nothing there I can see about self-compassion or self-acceptance. If anything there's a general trend in the opposite direction of encouraging self-criticism. This probably works in the short-medium term for people who experience pleasure from validating their own self-critical thoughts, but I fear in the longer term that it will postpone or diminish their potential to love themselves.

Admittedly I have only read the rules themselves and not the whole book; in fact I'm reluctant to read the rest if that is the best he can do.

Just a quick link to alert people who may not realize who/what he promotes.

I read all sorts of terrible stuff, and it probably influences me more then I'd like, but it's easy to jump in the middle of something that "seems" reasonable on the Internet.

Jordan Peterson is indeed a good person to help answer, "Why do they do it?" Because the answer is always money, and Peterson is willing to be repeatedly make a fool of himself and compromise his supposed morals to pander to people who give him money and attention.

> virtually every mobile app

Reminder to readers to use F-Droid

naive question, if you use f droid and you have location services on, use a bunch of social apps, have google maps on it, use it normally, what extra layer will f droid have with all of those leaks anyways?

F-droid is an app store that carries only FOSS apps that do not have any Google API's. These apps are generally safe. Some may require your location such as maps to function. The difference with most of these mapping apps based on OSM.org is that they do not transmit your location anywhere.

You can still install other proprietary location harvesting apps on your phone. If you do so, you are just minimizing the number of data harvesters.

To ensure nobody but you knows where you are: 1. Have only safe apps installed. 2. An OS that does not send your location to Google, Apple, or others 3. Your SIM must not be transmitting to the Telco's. This means no SIM or airplane mode.

You can make emergency calls without a SIM card. The phone identifies itself to the network (with IMEI), even if there is no SIM. The telcos, obviously, map IMEIs to SIMs. (Perhaps needless to say, but for basic operational purposes they have to quite efficiently triangulate your position, to know which cell tower to instruct your phone to use.)

Having no SIM does not help you conceal your phone's location at all, only perhaps to make it harder to map it to your identity.

I don't know if any location data might leak in airplane mode, but I would not be surprised if some did, for example, through NFC or Bluetooth.

This depends on what airplane mode means on your phone. With GrapheneOS, airplane mode blocks baseband transmission, but you can optionally still turn on Bluetooth and/or WIFI. If you dial 911, airplane mode is immediately turned off for the call to go through and stays off until you turn it back on.

WIFI triangulation or bluetooth mesh can be used for location tracking in particular areas. MAC address randomization and generic phone ID makes this more difficult, but broadcasts of known devices in WIFI or bluetooth scanning can give you away.

Others may know more than me on this.

> If you dial 911, airplane mode is immediately turned off for the call to go through and stays off until you turn it back on.

I didn't know that, that's really smart and so simple. I like it. I don't know why such trivial thing is so amazing to me.

> I don't know if any location data might leak in airplane mode, but I would not be surprised if some did, for example, through NFC or Bluetooth.

Which is why hardware kill switches are the future (see Librem 5 and Pinephone).

The future of never happen.

The worlds worst privacy offenders make/sell phones. Google. Samsung. Apple.

Google, no need to explain.

Samsung? All their apps, their keyboard, collect, collect collect.

Apple? Walled garden of share data with Apple.

And beyond this, for example on Samsung builds, the GPS daemon calls home(Qualcomm? looked into it a year ago...) to update agps data, but also... provide tracking info.

None of these people want hardware kill switches.

Which is why we should support the ones I listed.

> I don't know if any location data might leak in airplane mode, but I would not be surprised if some did, for example, through NFC or Bluetooth.

Not sure about NFC, but at least BT gets disabled in airplane mode.

Bluetooth stays enabled on iOS in airplane mode. You have to disable Bluetooth separately.

Weird choice for a "stop all radios" toggle to not stop a certain radio.

You forgot the IMEI.

Unless you wrap that phone in a Faraday sleeve, what you mentioned does not prevent tracking.

It Depends.

Any f-droid app that depends on location uses the standard android location APIs. That means you're submitting what RF networks (mobile and wifi) are visible, and Google then passes back an approximate location. Once you have a good GPS lock, the same data is submitted to Google again for inclusion in their database.

If you avoid installing Google Play services, you'll have to use a framework like microG that re-implements the Google APIs. microG lets you plug in a number of backends, with a default of sending the same data to Mozilla Location Services instead of Google. You can easily turn this off and use local-only location providers that either build their own database any time you have a GPS lock, or use downloads from opencellid.

Unified-nlp is another option to replace the Google location backbend. It similarly allows you to select location options or build your own local database.

None. F-Droid is just an app store whose apps tend to be very safe. It isn't a program to protect you from apps that you installed from other stores, or from google.

Also curious of this

And now you are the weird outlier that doesn't have the same data spewing forth instead of Average Joe making Average Data.

Exactly. Bruce Schneier - Data and Goliath.

i suggest you download netguard and block all apps from accessing internet https://netguard.me/

Are the apps different if I download them from F-Droid? For example if I download Signal, is it different then the Play Store version?

In the case of apps that are available on both f-droid and on the play store, the difference is that the f-droid versions will be compiled without google play services. This can limit functionality (for example, receiving notifications while the phone is asleep), but does increase privacy somewhat.

Significantly, there is very little overlap in the apps available on f-droid and the apps available on the google play store. Most play store apps siphon data, and most f-droid apps don't.

i think a state level actor would find it much easier to compromise f-droid and its app then the play store.

at least with google play you know they're selling your data so can adjust your behaviour accordingly.

Or on iPhone, switch "location services" off.

I don't trust Apple either

F-Droid then DNS66

Both are bad situations but the government has more power and control than tech companies.

> Working close to Government... I've never seen any mission specifically seeking large-scale citizen-based tracking capabilities.

They tend to keep those kinds of things pretty secret. We didn't know about all kinds of spying they do on us for even decades after the spying started.

> Working close to Government... I've never seen any mission specifically seeking large-scale citizen-based tracking capabilities.

Further and rarely understood, but it never is about what the government currently does, but what it could do when change occurs.. like the Nazis could exterminate the dutch jews (was it then?) better because they had those registers.. like what if the next Putin/Orban/Trump takes over and wants to take some more steps..

General principle of data economy is one pillar for civil rights! Its abuse by companies right now is an super annoying but comparably small issue.

> what if the next Putin/Orban/Trump takes over and wants to take some more steps..

What are you implying here by bring up those names?

Reply on news.ycombinator.com