CreepJS is much better (and scarier) at fingerprinting you than EFF. I've not managed to completely fool it yet but I've got my score down to 0% trust, meaning the fingerprint it generates is pretty useless. I suspect the only way to totally fool it (by which I mean spoof my devices) would be to turn JS off completely.
this post was submitted on 01 Nov 2023
97 points (97.1% liked)
Privacy
31914 readers
484 users here now
A place to discuss privacy and freedom in the digital world.
Privacy has become a very important issue in modern society, with companies and governments constantly abusing their power, more and more people are waking up to the importance of digital privacy.
In this community everyone is welcome to post links and discuss topics related to privacy.
Some Rules
- Posting a link to a website containing tracking isn't great, if contents of the website are behind a paywall maybe copy them into the post
- Don't promote proprietary software
- Try to keep things on topic
- If you have a question, please try searching for previous discussions, maybe it has already been answered
- Reposts are fine, but should have at least a couple of weeks in between so that the post can reach a new audience
- Be nice :)
Related communities
Chat rooms
-
[Matrix/Element]Dead
much thanks to @gary_host_laptop for the logo design :)
founded 5 years ago
MODERATORS
On Safari 17 every time I visit the site it claims it's my first visit, despite a trust score of 57%. Not sure if I'm interpreting the results wrong or ITP is just doing its job.
I'm not 100% sure but I don't think creep stores anything on its github incarnation so it'll always look like it's your first visit.
iOS 17 Safari (especially with enhanced fingerprint protection on) is really good at fingerprint protection. It rotates a few data points like canvas ID so that it makes you look like a new fingerprint each time.
Fingerprint analyzers can find out lots about your fingerprint that way, but if your fingerprint keeps changing, it becomes difficult to identify you. Unique fingerprints don’t mean anything if your fingerprint keeps changing.
That's what I was kind of thinking/hoping based on the results, but I wasn't sure if I was understanding it right. Thanks for elaborating!
Imagine I keep a log of everyone I encounter… their race, hair colour, eye colour, glasses shape, accent, gender, fingernail length, ear lobe shape, everything. I would probably encounter the same people every so often, and I would be able to recognize them from my log.
Now imagine that one of them started dying their hair and putting in coloured contact lenses, and they changed it up every day. I may be able to collect all of the details about them. They’re very unique. But… I couldn’t match them against anyone in my log, even though I’ve seen them multiple times.
Having a unique browser fingerprint is perfectly fine if it constantly changes. They can collect all of those details about you, but if you keep changing key details, they won’t be able to recognize you.
load more comments
(2 replies)
Wow yea this seems really good. And scary. Too bad it doesn't seem to work with mullvad browser
I get 0% on CreepJS with default DDG browser set to "strict", with a crowd blending score of 27%.
I get 40.7% with Mull + adblocker and 66.5% with FF + adblocker
"Your browser has a unique fingerprint"...well that isn't good...
Unique among the people who use that website. So if nobody else if you're configuration ever tried that website.... You would be unique
The bits of entropy are the more important parts of the results. The lower the bits the better
Please also consider things like canvas spoofing. It will create a unique fingerprint that is different every time.
I'm not sure how to read this report. It says my browser is unique and random with strong protections.
More of my stats:
Fennec (privacy badger + unlock origin): 1 in 23301.0
Fennec private tab (privacy badger + unlock origin): 1 in 20712.44
Firefox hardened (arkenfox + privacy badger + unlock origin): 1 in 37281.6
Firefox hardened private tab(arkenfox + privacy badger + unlock origin): 1 in 31069.5
Mullvan browser (dafaults with unlock): 1 in 147.48
There is also fingerprint.com, which I tend to trust more since it's a company that literally sells fingerprinting tech to other companies.
It managed to identify me while using the Tor browser on "Safer" (doesn't work on "Safest" due to JS). Edit: this is likely due to an issue with my install, and not the browser itself.
How did it identify you via tor? Were you using the browser bundle? Completely vanilla?
Did you refresh your session between tests?
Completely vanilla, fully stopped and restarted the browser. This was right after the 13.0 update.
For what it's worth I just tested.
Tor browser 13.0.1, plus U-Block origin, fingerprint.com did not identify two different sessions
I should test again then, not sure what happened
Just did a fresh install on Linux (fresh download too) and unfortunately, with no settings changed except security to "Safer", it once again identified me across multiple sessions.
FWIW it does change my ID if I resize the window enough to jump to a different size letterbox.
Edit: forgot to mention, the fresh install got a different ID ti the previous install.
Something is very curious about your install.
if you want to debug this: in tor browser, double check your using tor, try out whatismyipaddress.com, change circuits and make sure it changes again.
Look at the bits of entropy that coveryourtracks.eff.org shows, it could be something funky like an environment variable letting in system fonts.
Yeah I'll do some investigating, good to know that the Tor browser isn't at fault though (I probably should've operated under this assumption in the first place).
I got "unique among the 185,973 tested in the past 45 days"
Edit: this is using Firefox Android Nightly with UBlock + Canvas Blocker
Within our dataset of several hundred thousand visitors tested in the past 45 days, only one in 4330.4 browsers have the same fingerprint as yours.
Currently, we estimate that your browser has a fingerprint that conveys 12.08 bits of identifying information.
Using Mull with NoScript through Mullvad
Same here Edit: except unique in 186,012 tests
load more comments
(1 replies)
mullvad vpn + mullvad browser + a bunch of extensions: 1:26830.0
fingerprint.com does not track me
Mullvad browser + extensions is pointless, might as well use LibreWolf or just harden Firefox yourself.
The point of the Mullvad browser is to not stand out from the crowd; by installing extensions you are definitely standing out.
librewolf doesn't do auto updates.
I don't need to harden firefox myself, mullvad comes pre harded.
You're right. What I meant was that you lose Mullvad's fingerprinting resistance by installing extensions, but if you're only looking for a hardened Firefox with auto updates then it's fine.
Librewolf Flatpak autoupdates lol
That's great! I'm glad you have a package management system that works for you!
https://librewolf.net/docs/faq/#how-often-do-you-update-librewolf
It should however be noted that LibreWolf does not have auto-update capabilities, and therefore it relies on package managers or users to apply them.
Am I looking for a high number or a low number?
Looks like Chrome randomized my fingerprint but Firefox doesn't. Does that mean I should be using chrome instead of FF?
Within our dataset of several hundred thousand visitors tested in the past 45 days, only one in 93387.5 browsers have the same fingerprint as yours.
Currently, we estimate that your browser has a fingerprint that conveys 16.51 bits of identifying information.
But also
Your browser has a nearly-unique fingerprint
I don't get it
Almost no browser has the same fingerprint as yours, which makes it nearly unique.
There's a couple issues going on here. Number one is it's unique amongst the people who go to EFFs website cover your tracks. That's not all of the internet users. Hell that's not even most of the internet users. It's pretty niche community.
The bits of identifying information are the critical key here. 16 bits, 2 ^ 16.. 65,000 different possibilities. Each piece of information you give, makes it a little bit easier to track you. Things like language, time zone.. The more bits, the easier it is to identify you. The less bits, the more you blend into the crowd.
This is why multiple people, including myself, have talked about fingerprint.com they're professional service, who's targeting websites, who want to track users. So they're incentivized to track as best as able.
Even if you've got a great EFF score, you should always check fingerprint.com, to see if they can track you.
Mine said -
Your browser fingerprint appears to be unique among the 184,486 tested in the past 45 days.
Don't get it either, just stock FF on stock Android
Stock browsers give a lot of information, supported system fonts, supported system languages, time zone, canvas size, browser window size, there's a lot of data that leaks out from the browser itself.
Install a weird game that installed a weird font into your system? Well now the entire world can uniquely identify your font combination as you.
Makes sense, thanks!
Though, no idea how that would affect your average Joe so I'll not worry about it!
Your browser fingerprint appears to be unique among the 187,041 tested in the past 45 days.
Same here. Is there a way to spoof a more generic fingerprint or something?
load more comments
(1 replies)
I've an unique fingerprint, but different fp results in every test run, with mostly wrong sys specs, only it shows correct my country, nothing else. Same in Browserleaks.
On LibreWolf, which I use to surf daily, I got one in 180k+.
Afterwards, I tried Tor Browser -which is honestly almost never used- and this was a lot better at one in 6k+. Though this was only in "Safer" mode, I tried testing it on "Safest" afterwards, but an update screwed it up and I somehow couldn't get it back to its standard opening size.
Interestingly, my best result I got once again on LibreWolf. This time, I changed two things:
- Enable letterboxing
- Disable Javascript entirely through uBlock Origin
This resulted in a one in 800+. I am interested to know how Mullvad browser users fare on Mullvad VPN.
I see. I too use similar settings on Librewolf: should try it.
Within our dataset of several hundred thousand visitors tested in the past 45 days, only one in 4244.39 browsers have the same fingerprint as yours.
Currently, we estimate that your browser has a fingerprint that conveys 12.05 bits of identifying information.
Firefox mobile with various addons, most important of which is probably NoScript
Doesn't work with Javascript blocked.
1:46157.75 using mobileSafari on iOS with NextDNS.
view more: next ›