Facebook scraped call, text message data for years from Android phones

Heh I don't bother with the app and suffer with the web interface.

Annoyingly I now need to "Request Desktop Site" to use messenger or else it tries to get me to install the app. The artificial friction they're put in place has pushed me ever closer to just deleting my account.

Also, “maybe I guess” now means “yes forever” to them. How many inebriated TOS acknowledgements have happened?

They don’t get access to everything on your phone do they? As I remember (and this was years ago) they could ask for the photo dialogue to pop up and you could choose a photo but the app didn’t actually get access to all of them, only the ones you chose.

On the iPhone this was actually fixed in iOS 11. The standard image picker (UIImagePickerController) now runs out of process, and no longer pops up a permission dialog, but instead just gives the app access to only the photo(s) the user selected.

Source: https://stackoverflow.com/a/46594741

For what it's worth, Fuchsia, the new OS Google is working on, is capability-based.

Convenience, I’d expect. Not everyone has the email addresses of all their friends, and email has limits on file sizes, etc.

People are strange. My family all have iPhones, we all use iMessage, yet for some reason we have a group chat on WhatsApp which is where my sister and mam share photos of my nieces.

The functionality is no different to iMessage. I don’t know why this happens. It just does.

I think the main difference between the two mediums is that email is push-based, while Facebook is pull-based.

With email, you have to explicitly select recipients. You're essentially saying "here are my photos, I think they are relevant to you specifically". The onus is on the sender to figure out what's good for the receiver, and it's considered rude to send many frivolous or irrelevant messages. Think of how you grumble when you unsubscribe from some company's mailing list - this is an example of this social norm.

With Facebook, you say "my photos are here, anyone may look at them if they want". Figuring out whether the content is relevant is now the job of the receiver, not the sender. Facebook's UI is well aligned with this role: unlike email, where you must explicitly download attachments and mark messages as read, on Fb image previews are displayed inline and to never see a message again you must only scroll past it. Advertisements aren't considered intrusive, it's just content from another source that is (in theory) just as easy to ignore. The social dynamic is very different, and so it's used to send a different sort of message than email.

I have 75 relatives on Facebook who live around 3000 miles away from me. If I want to share a photo do I send a mass email to all 75 annoying those who aren’t interested, do I BCC everyone so I get repeated responses (and those responses aren’t shared with anyone else who is interested). Or do I post it on my Facebook feed where those who aren’t interested can ignore it (and it will be gone soon) and those who are can have a conversation around it without getting in everyone else’s way?

Edit: ok maybe around 40, not 75. But point still stands, it’s indirect communication rather than remembering about that cousin of an auntie’s grandparent’s nephew’s sister-in-law that I met ten years ago.

"Allow once" is different from "Allow just this one picture". With "Allow once", they can still scrape all your pictures, they just have to do it in one go.

Custom ROMs unfortunately can't really do much to implement the latter, they'd have to break compatibility with the whole Android ecosystem, which Google knowingly built this way.

LineageOS supports this.

That should be something that doesn't allow the app to get onto the Play store, but Google apparently allows it.

Android used to have the concept of an Intent. Does that still exist, and if so, why do apps still require full storage permissions?

Side note, there was also a project at one point to create Web Intents. I wonder what the web would be like if that had worked out.

Android has a lot of nasty corners around stuff like this, which can prevent being both streamlined (accessing all photos does let you optimize for your app, some people want that) and privacy-friendly (for people who don't want that). My personal favorite:

If you don't request camera permission, you can still open the camera with an intent. This is great!

If you do request camera permission, and you are denied, you can't use that intent either. https://blog.egorand.me/taking-photos-not-so-simply-how-i-go...

I honestly can't tell if it's this way to be intentionally hostile to users, or by accident, or if they honestly think this is a positive quality.

This would be easier to control if Android would allow users an option to see historically what resources the app requested and how it processed them (just list them, used them, processed them).

What legal system are you thinking of? I don't really see myself getting in legal trouble for revealing someone's contact information to a third party. It's a pretty normal thing to do. If I knew your phone number, and somebody else asked me for it, what's stopping me from telling them?

I hope the current controversy will inspire Samsung to take Facebook out of the mandatory bloatware set for all models.

you can disable, it is pretty much the same.

The only difference is the you don't get your space back since preinstalled apps occupy a different partition, but the app itself cannot run in any way.

Unless another social network comes along and FB will die out loud like Friendster and MySpace did.

If you think solid $ numbers are everything to keep "printing money" then see what happened to Kodak or Sony.

I created my Insta account before FB bought it. For some stupid reason I let it connect to FB for (limited) sharing some pics, rather than doing it manually - then that was it.

I don't want all my FB contacts to "see me" on Insta; I just have tonnes of people adding me with no real interaction. The integration is really quite obnoxious. I have in fact stopped using Instagram now due to this.

Yes! It's insane. It's probably using location data and IP address among other things. If they had even an ounce of concern about privacy they would detect and explicitly prevent making recommendations across accounts that someone is obviously trying to keep separate.

The best breakdown I've seen of this issue, including glib comments from Facebook: https://gizmodo.com/how-facebook-outs-sex-workers-1818861596

Who cares about the user?

but it exploits them for profit much, thus serving the primary mission: shareholder value

As someone who has rallied against FB for years, there's always some justification:

"It'd be too risky to their business for them to do something like that!"

"Someone would have leaked the secret by now!"

"There's no way a company that big could get by doing something that blatant!"

I think people just don't like uncomfortable truths, even when staring them in the face.

Everything you install on basically every desktop OS, by default, has access to practically everything.

I think most people would be surprised if they discovered that apps were uploading their emails. And photos. And tax documents. Just because they happened to be on the hard drive, and there were no permissions to prevent it.

Expecting FB to not do similar and respect basic privacy by default is reasonable. FB doing such things just because they can is not.

> READ

It says read, not steal and keep in their own servers.

If I thought that a file manager would steal all my data, I'd never install it. On (most) desktops there's no easy choice: apps will have permissions, but you might be able to sue the author for illegal access if they took all your data.

On mobile, suddenly, it's: lol, you gave it away!

That makes little sense. Yes, we should have better sandboxes - but we still have very few (none) usable general operating systems with Internet access that are "secure beyond trust".

Absurdly, I'd be more confident in a typical Debian install with thousands of programs from "main", some running in the kernel - than a typical Android device.

Because of presumed incentives for the volunteers/employees working on debian and the various upstream projects.

This is 99% about trust and incentives and 1% about capabilities.

What platform? Surely android as you can do that in iOS without those permissions and I would have thought you’d get rejected if you tried that on.

Noted. Didn’t know that was the case. Fle some reason, I imagined that it being open source would have meant a lot of outside involvement.

Not everything that's permitted is compulsory. I mean, suppose you frequent a store with a penny tray and you decide, since it's ok to take one penny, why not take all of them? And if you get away with that, why not start cleaning out tip jars? Maybe you will get arrested, but maybe not. I feel like computers are training people that everything you are technically permitted to do is worth trying. I'm not even sure what the rules should be, but rather I feel like losing a common sense of unwritten rules is losing part of what it means (or used to mean) to be human.

The app store is front and center on Mac. All apps are sandboxed by default. Steam is another sandbox. A browser is a sandbox.

The problem is that it's hard to write an interesting Mac desktop application that runs in a sandbox. The kind of complexities that require a full blown desktop application just don't fit in a sandbox. (As opposed to a game, mobile, or web app.) Whatever runs in a sandbox turns out to be just a prettier version of a web app, or a self-contained game.

We can wear bullet proof vests every time we leave the house too.

Put a few app developers in jail for what they do and render their businesses bankrupt and maybe we don't need to treat our phones as hostile to their owners?

> Back in the day, apps that did this were called spyware

Right?! Remember when ad/spyware that tracked every site you visited were considered devilish and flagged by virus scanners? Now it's the norm. Crazy.

There is some truth to that, but it's not that hard of a problem. Start by making a dependency tree, and only displaying the top nodes. "Do you want to share anything with app X?" yes/no and go from there.

Often the "it's too complicated" excuse is really a cover for "we make out money off your data and fund the development of this software to harvest it".

No, permissions were done in a substantially more granular way on iOS from the beginning of permissions. Contacts, photos, location, notifications all with their own permission request API to be fired at will.

Android started off with a blanket permission screen required to even install an app — all or nothing.

Android is a decade old. Our expectations of what apps would and wouldn't do, and would and wouldn't be capable of doing has changed. Back when the apps on my phone were in the range of being 400 KB to 600 KB, I don't think people even fathomed the complexity and power our devices are at right now, and their ability to secretly handle ever increasing amounts of our personal data without having a meaningful impact on device performance.

^this. As long as Google remains an advertising company with some incidental technology projects - they'll be hard pressed to ever fight internal culture/revenue enough to take meaningful steps towards privacy and security.

Just being able to write an app (code on the device) and deploy an ad (code on the Internet - possibility to run "code" like fonts, or trigger calls to site/unique.jpg) - would make preventing data exfiltration and/or tracking absurdly hard while continuing to cater to advertisers aka the paying customers.

I am on the same boat here.

Technically it is NOT google's fault to open those access to Apps.

Just like credit card CVV, the right to grant you the access to it, prohibit you to store in your server. For personal information, I think social networks need to be held responsible to live up to the same standard.

By granting Facebook permission to read my contact, my understanding is that they should only use my contact to match against their DB and find those who are on FB. I don't think this require them to persist all my contact/conversation history in their own server.

I’m curious about what you’re saying here and I’ll certainly be thinking about it more, but here’s something I was thinking about: maybe a step in the right direction would be to require permissions to offer a user configurable time limit. I think that users should be able to set the permission they grant to expire after a term of their choice—if they want to grant perpetual permission, fine. I think it would be interesting to think about the implications of apps having to come back to get reauthorization. When permission expires, maybe that just means that no more data can be collected; maybe it means the vendor has to destroy those records. Either way, I don’t want large companies to be able to exploit people who click a button they don’t read. Maybe companies should be required by law to charge users some form of consideration just like other contracts so that we can do away with companies dangling their product as a free carrot in exchange for swaths of personal data, and then leaking it.

I don't think there has ever been a human society in which people took "personal responsibility" in the sense you are using the phrase. I don't think it's humanly possible for even the smartest person to keep track of their personal data given the quantity of it, the multiplicity of tracking mechanisms, and the uncertainty about how it all works. Not to mention, even if we all read all the agreements we click through, that doesn't mean they are all accurate, sufficient, legal, or adhered to by the authors.

As I wrote in another thread, I have used LinkedIn for a long time, and I have never wanted it to spam my contacts, so I have always had it foremost in my mind to click "No" whenever it asks to import them. Yet at some point, it did it anyway, because it asks me if I want to connect with people who are only email contacts and not on LinkedIn.

Now if you had complete logs of everything I did with my phone and computer, you might well be able to prove in court that I inadvertently gave permission at some point - perhaps I didn't read all of the legalese on something, or perhaps my finger slipped and I forgot.

I can't imagine I would find anyone at LinkedIn who cared about figuring out what happened, regardless.

There is something perverse, in my view, in appealing to "personal responsibility" of individuals dealing with corporations, as it seems to me that the entire concept of a corporation is a way for people to work together as an entity without taking personal responsibility. The reason we have corporations is because it's impractical for people to be held liable for their screwups.

Your standard of “personal responsibility” includes the ability to audit code. I assume that you’re just irresponsible if a carmaker sells you a dangerous vehicle, because you don’t know how to detect the flaw? Doctors can just talk to you without consideration for your ignorance, because your failure to attend medical school represents your irresponsibility? If you can’t parse all of changing EULA’s in your life, you’re just irresponsible?

What’s obvious to you in your very limited field of expertise is not obvious to everyone. You shouldn’t insult everyone who isn’t a programmer by equating that narrow expertise with personal responsibility. I wouldn’t assume that your inability to understand a conversation between two surgeons meant that it was acceptable to harvest your organs.

This libertarian utopia obviously breaks down any time you're presented with a 200-page Terms of Service. Nobody has the time and/or skills to read and understand the content at the level required for "informed consent".

Which is why societies have come up with a far better method: collectively decide (or collectively choose people to decide) on reasonable limits for certain types of transactions.

My German law professor used to say that she never read ToS. Because under the country's law, they are either reasonable or unenforceable.

Such laws have nothing to with censorship. If you really need your users private messages, you just have to more explicitly present them with the choice, and respect their decision to say no without unreasonably denying them service.

The US has far more lenient standards for such one-sided contracts, but the basic principle is obviously the same: If Facebook were to add a paragraph giving them ownership of your house somewhere deep in the ToS, they wouldn't stand a chance in a court of law.

Saying I gave it to them is like saying a guest that robbed me should be able to keep what he stole just because I let him into my house.

If I’m friends with anybody who gave it to them, they also have my private txt messages with them and phone call metadata.

I have to think that violates phone tapping laws in certain states.