A facebook employee explained me how tracking works. Its not the email address Meta is concerned about. Its the IP, device identifiers and location. Meta doesnt care about the email at all apart from sending you emails for notification. Even with a fake email they exactly know who you are. Let’s say you visit CNN.com which has facebook tracker. Facebook has the IP and the device identifiers. Now you login with fake email account on Instagram, facebook knows that’s the IP ans the same device hence it “must” be the same person That’s how facebook creates shadow profiles.
Its not the email address Meta is concerned about. Its the IP, device identifiers and location.
This actually applies to the entire internet, look into fingerprinting. This website checks how susceptible you are to it: https://coveryourtracks.eff.org/
I like the EFF, but I don’t agree with the report this generates. There are two counters to fingerprinting: have the same fingerprint as everyone else (Mullvad Browser is based on this idea) and to have a unique fingerprint that changes regularly (The CanvasBlocker extension supports this approach).
Since most of the time I’m in Firefox with CanvasBlocker, I want to see unique fingerprints, but also that they keep changing.
There’s a decent bit in their site as to how fighting fingerprinting by trying to be more common can make you still stand out, so mullvad may not work out depending on how it implements this concept. Randomizing fingerprinting sounds like it could work (I haven’t researched it so I don’t have enough info to agree or disagree, but sounds legit at the very least) and expecting their report to understand that is beyond the scope of the tool. I mean, you couldn’t actually test that method is effective without recording it over multiple sessions/days/etc. Sure you want a unique fingerprint, but seeing a unique fingerprint once doesn’t mean it’s working.
Mullvad’s website has this nice widget that checks if your ip address can be found by dns too. Good for busting competitors
That site says “Your browser has a unique fingerprint” even thought I run Firefox, uBlock, Privacy Badger, and have privacy.resistFingerprinting set to true. My main problem may be plugins, once you have more than a few your set can be pretty unique.
Iirc unique != identifiable necessarily, because your fingerprint might be different while still unique the next time around.
Try going down the page and looking for the categories with more than a few bits of identifying information. I’m running LibreWolf with just uBlock Origin and Dark Reader (which I don’t think influences results) and I’m able to get nearly-unique, instead of unique (but I do get unique on default settings). TBB gets non-unique, which is a good set of results to compare to.
In my case I noticed that my fonts were really unique so I set
browser.display.use_document_fonts = 0
. Also I use my WM to set my page resolution to 1920x1080, which seems to have a better fingerprint than the default LibreWolf floating resolution of 1600x900 (and even the letterboxing resolutions, from what I can tell).I just spent some time testing again and checking for anything else. RFP does force a generic user agent, but unfortunately it keeps the version information and I can’t figure out how to change it with RFP on. Would be nice to set it to the ESR version used by TBB (which has lower bits), but I’m not sure if that would lead to a more unique fingerprint (if, say, a feature was detected that is available in later versions but not ESR).
It’s almost impossible not to have a unique fingerprint online with how stuff is tracked. Websites are tracking user agents, screen resolution, your GPU/web graphics renderer, etc.
The only way is to disable JavaScript, but good luck to using the internet without it
Never heard about Tor Browser? Nice “privacy” community you’ve got here.
OMG this scared me. I thought I was “enough” protected
what is the best way to regularly change your fingerprint? (I’m using firefox)
Tor Browser, the most trusted and reliable. Also maybe Brave.
deleted by creator
Firefox has containers that allows you to seperate all websites in categories where they can’t reach anything outside of it.
There’s a special one for Facebook
There are still…I dunno…probably DNS hops, IP, time’s of day, browser window size, browser user agent…
And if you access any page with any similar parameters on your phone or another household device on any site with FB tracking, it’s over.
It looks like in the last 7 days my phone has cutoff over 150,000 different tracking attempts and that’s just catchable ones and on my phone.
Canvas, webgl, audio fingerprinting and others I think are still effective through the container.
I don’t use Facebook but I’m 100% sure they have my data.
A lot of apps that uses Facebook login, debugger, React Native, etc. allows it to collect as much user data as it can and send it to FB servers because that’s the default.
I dont have facebook, and I explicitly tell family not to put my pictures on their facebook pages or mention me at all.
I’m still 100% convinced facebook has my biometric data, my home address, and what I ate for dinner last week.
The amount of data they collect is insane, and intrusive.
Every time it comes up, i’m reminded of a sex worker who was doxed by facebook because she in a parking lot that a former client was in, and it had used proximity data and shit to link her Sex Work Phone/Facebook Account, to her real Phone/Facebook account, which was then given to the client as a suggested contact.
Facebook takes biometric data from pictures that aren’t uploaded to the platform. All it takes is for them to have access to the filesystem of the user’s mobile OS.
This is why I fullstop do not let people take photos of me where I can help it. I’m fucking tired of being made a datapoint.
I’m pretty sure React/React Native doesn’t have any Facebook tracking built in. The dev community would crucify them for that.
I would highly doubt that. They don’t care, most of them not just willingly but intentionally insert google tracking code and similar.
Preaching to the choir but food to remind some people. Thats why you avoid or limit use of those services. Use tor or a VPN and use multiple layers of blocking such as DNS and in browser blocking. Also foss only applications where possible.
deleted by creator
Proton VPN is free and there is also riseup VPN. I run a tor node so I can at least vouch for that one 🙂. The more people that use it and run nodes the more it normalizes it. You can also use onion repos in your distro if supported. I know Debian had had then for a few years now and it uses the apt-transport-tor package.
deleted by creator
A tor node or relay is just someone running and configuring tor on their server to accept and forward connections to the tor network being part of the chain. Normally a chain has three nodes: an entry , a middle and an exit.
Onion repos are software repositories in this case for the Debian Linux operating system that contain firmware and package updates and are hosted as onion services accessible over tor.
Obviously. Device fingerprinting is much, much more advanced than that. IP, device screen resolution and type, etc. And that’s just the basics
Dont use facebook and have a dns filter that kills trackers and adverts such as Control D or pihole
I tried using control d and their dns location changer. Its crap
Really, i have been using the free one that blocks malware, ads, trackers, and social media (spybook, etc) and its been great.
Dont use facebook
Doesn’t really matter if you use it or not.
Strange game. The only way to win is not to play…
How about a nice game of chess?
Meta applies a myriad of advanced and complicated tracking methods. Email is a very popular and easy one. I believe the one you’re referring to is called a tracking pixel.
For example, some browsers block tracking Pixels, but if you’re logged into Amazon with an email address that Meta knows, they will sell your shopping habits to Meta to show you ads.
deleted by creator
Using a VPN does exactly nothing against cookies or device fingerprinting.
deleted by creator
I am using Firefox, and with a shit-load of add-ons that supposedly prevent unwanted cookies and fingerprinting. I use a VPN.
I was permanently banned from Reddit (for advocating firebombing nazis, as if that’s a bad thing). When I logged in to an alternate account, that account was also permanently banned. Any account I tried to create after that point ended up being banned within a week, regardless of whether or not I was using it. I checked online. Apparently this has become fairly common in the last 2-3 years.
While you can minimize your digital fingerprint, it’s almost impossible to prevent all digital fingerprinting. The EFF says that I have very strong protection against digital fingerprinting, but I’m still identifiable to a company with sufficient resources to devote to the task.
All you have to do to avoid reddit perma bans is to log in on a browser without cookies, on a VPN, and that’s it. If you’re still getting permabanned instantly on new accounts you’re doing something wrong. Maybe you’re using a VPN address that has been banned, idk. I ban evaded multiple accounts for years and only stopped because after I moved my IP changed and I don’t need to fuck with the VPN anymore
The main thing is you’ll be shadowbanned probably and you’ll have to fuck around with getting karma and begging the admins to un-shadow ban you if your posts don’t show up
deleted by creator
But the VPN company still has your data
deleted by creator
There is no such thing as a “good VPN provider”. VPNs were not created for privacy. They exist to allow individual users and groups to network together in enterprise environments
If you want more security use i2p or tor
yep this is why i use containers on firefox.
And FF containers are still no match for advanced fingerprinting.
The only way to protect against advanced fingerprinting is to use the TOR Browser or Mullvad Browser, to blend in with everyone else who shares the exact same fingerprint using those tools. The best you can do outside of those is to protect against less advanced scripts.
that is much more complex for normal people but yeah. tor can be slow as shit especially.
Tor is also not recommended for streaming since the p2p connection bypasses the Tor network
Does Firefox containers hide your IP, device info and location?
you can use firefox containers with container proxy to have different ips on each container. that said i run wireguard on my router itself so all the devices are behind vpn.
https://addons.mozilla.org/en-US/firefox/addon/container-proxy/
as for location, changing ip also changes your geolocation. there is also the location reported by your browser which you can change in about:config but this isn’t provided unless you give permission to the site asking it.
as for device info, each container is like having a separate firefox install. all the cookies are separated and isolated.
I think Firefox cookie isolation (not containers) should block that. Also, always use Noscript and block that shit entirely. You will have no Tracking anymore basically
While browser containers won’t work since you’re using the same IP anyway, blocking the trackers themselves would be more effective. DNS blocking, uBlock, and Privacy Badger can help block fb trackers on websites. So fb knows your ip, but at least they can’t track you across other sites.
And people dont share devices?
I have a family pc in the living room. We all use it.
Im not on facebook though.
There’s got to be more metadata involved in fingerprinting. The type of content you’re looking at. Maybe even deriving some sort of signature from your mouse movements.
deleted by creator