Browser Fingerprint Detector | Hacker News Reader

Browser Fingerprint Detector

fingerprint.goldenowl.ai

98 pointseustoria66 comments

I'm deeply confused by a lot of the privacy discourse here. There seems to be opposing goals between preventing the fingerprinting mechanisms and just preventing uniqueness. Under the "preventing uniqueness" model, my Linux computer with custom Firefox and no fonts, and no js, etc. is the "most fingerprint-able" because it's the most unique. Whereas grandma on Windows and Chrome is "less unique," and therefore in some sense less fingerprint-able.

I think there are a few potential problems with this view that I never see discussed:

- Firefox sends some dummy data when making use of privacy.resistFingerprinting, and so you should get a unique fingerprint _every time_ you visit a site, so the fact alone that you're unique might potentially not matter if you're _differently_ unique every time you visit the site. Is there a flaw in this line of thinking?

- My understanding is that the primary utility of browser fingerprinting is for advertising / tracking. In other words, the bulk of the population an advertiser would actually care about would be the huge middle of the bell curve on Chrome using Windows, not the privacy nuts on Linux with a custom browser config. In other words, if "blending in with the crowd" really worked I would think that tracking companies would fail against the most important and largest part of the user pool. If anything, it's more important to target grandma as she will actually click on ads and buy stuff online compulsively.

Can anyone speak to these points? I often feel like the pro-privacy people are just crawling in the dark and not really aware of that real-world tracking is actually occurring vs. what might be possible in a research paper. Maybe I'm just the one that's confused?

"... so the fact alone that you're unique might potentially not matter if you're _differently_ unique every time you visit the site. Is there a flaw in this line of thinking?"

No, you're thinking correctly and the odd discourse that you (and I) see is based on two implicit assumptions:

1) Your threat model is a global observer that notices - and tracks and exploits - your supposed perfect per-request uniqueness.

2) Our browsers do not give us fine grained control over every observable value so if only one variable is randomized per request, that can be discarded and you are still identifiable by (insert collection of resolution and fan speed or mouse jiggle or whatever).

Item (1) I don't care about. I'd prefer per-hit uniqueness to what I have now.

Item (2) is a valid concern and speaks to the blunt and user-hostile tools available to us (browsers, that is) which barely rise to the level of any definition of "user agent" we might imagine.

I repeat: I would much prefer fully randomized per-request variables and I don't care how unique they are relative to other traffic. I care about how unique they are relative to my other requests. Unfortunately, I am wary of browser plug-ins and have no good way to build a trust model with the 12 different plug-ins this behavior would require. This is the fault of firefox and the bad decisions they continue to make.

> Unfortunately, I am wary of browser plug-ins and have no good way to build a trust model with the 12 different plug-ins this behavior would require. This is the fault of firefox and the bad decisions they continue to make.

I see so many people paranoid about browser extensions and I really don't see the point. It's like any other software. If you trust the author, install it. If you don't trust the author, check the source code, install it (ideally from source), disable automatic updates and subscribe to the changelog. Is this any different from any other thing you install on your device?

>- Firefox sends some dummy data when making use of privacy.resistFingerprinting, and so you should get a unique fingerprint _every time_ you visit a site, so the fact alone that you're unique might potentially not matter if you're _differently_ unique every time you visit the site. Is there a flaw in this line of thinking?

Yes, because those randomized results can be detected, and that can be incorporated into your fingerprint. Think of a site that asks you about your birthday. If you put in obviously false answers like "February 31, 1901", a smart implementation could just round those answers off to "lies about birthday" rather than taking them at face value.

>- My understanding is that the primary utility of browser fingerprinting is for advertising / tracking. In other words, the bulk of the population an advertiser would actually care about would be the huge middle of the bell curve on Chrome using Windows, not the privacy nuts on Linux with a custom browser config. In other words, if "blending in with the crowd" really worked I would think that tracking companies would fail against the most important and largest part of the user pool. If anything, it's more important to target grandma as she will actually click on ads and buy stuff online compulsively.

The problem is all this fingerprinting/profiling machinery ends up building a profile on privacy conscious people, even if they're impossible to sell to. That can later be exploited if the data gets leaked, or the government demands it. "I'm not a normie so nobody would want to show ads to me" doesn't address this.

Advertisers try to reidentify and match you against their database, the less information you give them and the more randomized it is, the less certain they can be, its you again.

If i use my locked down firefox with a VPN where potentially a hand full other brills like me come out on the other end, i am not concerned about them building a profile of me.

>Advertisers try to reidentify and match you against their database, the less information you give them and the more randomized it is, the less certain they can be, its you again.

This assumes the randomization is done properly, otherwise it just turns into a signal of "installs privacy extensions", which can still be used for targeting, as a sibling commenter has mentioned.

> otherwise it just turns into a signal of "installs privacy extensions"

The more such signals, the merrier, the parent of your comment addressed that ("other brills"). Instead of going in circles, it would be better to encourage people to evaluate and use such extensions. Telling them that they don't work is a self-defeating half-truth.

You are correct, the discussion is often unthoughtful and spun.

> the bulk of the population an advertiser would actually care about would be the huge middle of the bell curve on Chrome using Windows

The middle of the bell curve in the USA would be an iPhone and there is very little you can customize. So many people have the same model with the same settings that trying to track by fingerprinting is effectively useless.

Yes, PC/Linux users have more to track. They are the minority though. I'm not saying therefore ignore this issue. But grandma is using her phone. Not a PC.

> Firefox sends some dummy data when making use of privacy.resistFingerprinting, and so you should get a unique fingerprint _every time_ you visit a site

This assumes the fingerprinter can't filter out that random data, and that the feature is actually useful. Some of things it does sound like sites might fail or cause problems. Setting timezone to something else seems like I'm going to make a reservation for 7pm only to find out it was 7pm in another timezone. other things it doesn't might not be good for grandma. CSS will report preferred reduced motion as False. CSS will report preferred contrast as No Preference.

I definitely agree with you point, but I think that's what I'm wondering about? Can it actually be filtered out, and are tracking companies actually do this in practice, or is it like when someone says they bridge an airgap by making two computer's RF spectrum do funny things? Possible in a lab, yes. Something most people need to worry about? No.

I'm not saying this _isn't_ the case for tracking -- I just don't have much of a way to know what techniques are actually being employed in real life.

The comment by gruez is accurate IMO.

Creepjs actually tries to detect what your browser is lying about and takes that into consideration (or not) based on its heuristics.

I'm still not aware of any FOSS browser (with JS actually enabled and functioning) that can produce a random fingerprint ID on every refresh of the creepjs test site.

But please prove me wrong.

I've never heard of creepjs, are there more resources about it and where it's used?

I used to work in adtech a long while back. We found that our system could effectively target people who tried not to be targeted. By that I mean we realized a better ROI that without said targeting and click-throughs & conversions were happening for our customers at a nice rate.

At the end of the day the object of the exercise is generally less about building a perfect profile of a person and a lot more about getting said person to buy something. We found our system worked very well at figuring out what ads worked on privacy-conscious people and our customers saw a nice ROI from it.

In fact, it turns out pro-privacy technically skilled people cluster nicely and it's entirely possible to sell them stuff, and their attempts to be less 'profileable' than normal actually helped our mission (which was advertising, an endeavor that in my experience doesn't GAF about violating a given person's privacy in the way the pro-privacy crowd often thinks it does).

Take from that what you will.

Can you describe in more detail what sort of techniques were used to target and track people? What sort of privacy mitigations were feckless?

It wasn't so much that privacy mitigations were feckless, it was the fact that people who did things like falsify their User-Agent strings tended to cluster into distinct groups very nicely, and hence it was easy for the targeting algorithms to feed them effective ads, landing pages, etc.

The targeting system went "oh goody, privacy geeks" and was able to very effectively do its job. This is because ad tech systems care less about you as everdrive the named individual with privacy interests and other human aspects, and more about you as some potential consumer of goods.

While it's possible to use the systems to profile people in the sense that a stalker might, that's not really the intent (in the way people like to think of it). I (in the past tense, I don't do adtech anymore) honestly don't care about you, I just want you to buy shit from the people who pay me to sell you their particular flavor of shit. If you hiding your exact name or browser details or whatever makes that more likely (it turns out it did), then hooray! There's no conflict there, where to some there would be (because their assumptions about motive are all wrong).

In terms of what techniques, we found machine learning (stats) way back then did a pretty good job of clustering people based on things browsers return (monitor resolution, OS, etc.) coupled with time of day, search terms, and other things you can't really suppress. A completely contrived example might be pushing expensive pediatric electrolytes to someone with a large-screened Mac looking up baby flu symptoms at 2 am. The "system" did a far better job of real time targeting with this stuff than any human could, and the things it would cluster on were often rather unintuitive.

Thanks, this is a really useful reply. With regard to identifying privacy-focused users in your system, I'm sort of imagining the following scenario:

- user has a bunch of privacy mitigations and tweaks in their browser

- user logs into your commerce site, searches for stuff to buy

- commerce site knows who the user is since they're logged in, and per your comment can infer whether or not they're wealthy, have kids, etc. based on the user activity and whether or not the user is likely have an expensive screen & GPU, etc.

Does that sound right? That's really interesting, and something I'm embarrassed to say I hadn't really thought of. In other words, I've spent a lot of time worrying about cross-site tracking, and advertising domains, etc. However, if I'm purchasing from Amazon they know it's me since I'm just shipping to my own house. Even in a scenario where my browser is magically un-fingerprint-able, it's obviously me since I'm using my account and shipping to my house.

In other words, I may potentially have prevented a bunch of cross-site tracking and fingerprinting. Perhaps when I go to washingtonpost.com they don't know I'm the same person that Amazon knows about. (that might be a best case scenario) However, by virtue of the fact that my privacy config is operating all the time, Amazon has also learned something about me I didn't necessarily need to tell them -- ie, that I'm privacy-focused.

Do you think that's a fair assessment, or am I missing the point?

> Amazon has also learned something about me I didn't necessarily need to tell them -- ie, that I'm privacy-focused.

Exactly, and it turns out privacy-focused people tend to be relatively self-similar in many ways and, at least back in the day, were easier to advertise to.

Now, Amazon (or whomever) still doesn't know your name, but they are still targeting you and, what I found super interesting, is the fact that it was often more effective targeting than if the person was just part of the bulk of the population. It's kinda like if the system observed all the nonconformist teens like to wear Doc Marten's.

Informative, thanks for sharing.

  > Whereas grandma on Windows and Chrome is "less unique," and therefore in some sense less fingerprint-able.

I got highly unique on FF so I tested Safari on a M2 Air. Still says I'm highly unique. I'm on a university campus internet, there's thousands of people with that exact same setup. I don't think I've ever seen a finger printing site that doesn't say I'm very unique.

I think the problem I have with these types of sites is that they do not really offer advice on how to become less unique and how to protect one's self. It's probably pretty easy to identify machines through things like canvas fingerprinting or through all the other things that the browser actually exposes. Many privacy browsers like Tor or Mullvad will just send no data to those. That makes them "unique" because there's not many people using browsers that do that but it's unique in a way that makes you fungible. There's unique as in "uncommon" but also unique as "differentiable." I can't understand how these sites never make that distinction.

This is the paradox: Imagine walking dressed in red in the middle of a crowd dressed in black.

Being unique makes one easily identifiable and requires less effort to correlate one's past activity, while non-unique ones are full of noises and low confidence.

I recently wrote about the limits of these kinds of fingerprinting tests. They tend to overly focus on uniqueness without taking into account stability. Moreover sample size is often really small which tends to artificially make a lot of users unique

https://blog.castle.io/what-browser-fingerprinting-tests-lik...

This is great, and exactly the kind of nuance I almost never see when this topics come up. Thanks for posting this. Far too often, the pro-privacy crowd is much more _upset_ than they are precise, and to the point of your article are spending extra effort without really accomplishing much.

> Fingerprint Collection Failed

> This can happen due to several reasons:

> [...] JavaScript Errors: When any of the 24+ fingerprint collection methods throws an error [...]

So when any of the browser APIs it exploits aren't available, it just fails instead of using that as a datapoint in itself. I'm unimpressed.

This is why privacy must be enforced by states, their laws and a powerful public enforcement agency.

You cannot expect people to technically protect themselves from tracking.

(you can invite them to not use abusing services though)

> (you can invite them to not use abusing services though)

First, you'd have to define how one can determine what an abusive service is. Is Facebook an abusive service? Is some random website that happens to use FB's SDK an abusive service? How does a normie internet user find out the site they are using has abusive code? Some plugin/extension that has a moderated list that prevents a page from loading and instead loads a page dedicated to explain how that specific site is abusive?

> Is Facebook an abusive service?

Yes

> Is some random website that happens to use FB's SDK an abusive service?

Yes

Now write it down and get a majority of the population to agree with you.

Still nowhere near as good as creepjs: https://abrahamjuliot.github.io/creepjs/

Doesn't even detect common browser extensions.

Wow, this blows it completely out of the water. Even detects battery level, free storage, fonts etc

It depends on the browser you're using, Brave is obfuscating a lot of this info, for me using Brave on Android it shows 100% battery while my actual battery is 62%.

On Firefox on android almost everything except the basics you expect are "unsupported"

It has file system free space, but it's wrong.

Sure, but is the fingerprint at the top stable? It is for me despite most of the tests being blocked, spoofed, or randomized.

One thing I find odd is that, on LibreWolf, a lot of these fingerprint tests are disabled or even worse, randomized. How is it able to generate a stable fingerprint?

Ah... I was here wondering why browsers don't just run sites in a built-in virtual containers..allowing the same reports of the same hardware for everyone. especially for WebGL and canvas fingerprinting.

I suppose someone might say it is about performance of going through a virtual layer? I understandit might break specialized 3D web-apps...but for common web-browsing? idk. Do people regularly use web-based app that need direct access to a GPU to be fast and functional? But surely, an exceptions list could work.

I am sure I am missing something, but what?

people regularly use Google Maps and video conferencing apps (Zoom, Meet, FB Messenger, ...) all of which use the GPU. Maps to be able to customize the map based on what your point of interest is. VC apps to do background blur, addons, etc...

BUt can the gpu not be virtualized?

The original is panopticlick from EFF: https://coveryourtracks.eff.org/

I do not see how this is better

Perhaps I'm missing it but does it explain what aspects of your setup contribute the most to your score or suggest remedial actions? I wasn't that surprised to find that my standard setup is highly fingerprintable (for one, I use Firefox which alone is enough to single me out in a crowd) but I also tried using a vanilla Chromium install via a popular commercial VPN and still got a rating of 100%.

I'm curious as well. Ran a stock Vanadium config with Mullvad enabled, and got 100%. Maybe Vanadium isn't as focused on fingerprinting as I'd thought.

Running Chrome will make you highly fingerprint-able since it has so many APIs that can identify your hardware and software configurations directly or indirectly. It doesn’t help you “blend in” at all.

Looking at the JS, in the `calculateUniqueScore` function - it is just checking how many features it was able to detect (it gives a weight to each summing up to 100).

It is not checking how unique you are based off of some data-set it has.

This site also has plenty other such "issues"/"bugs" feels like it was quickly vibe-coded without much care.

Well, perhaps my system is better than I thought. The site could not resolve anything until I turned JavaScript on (JS off is my normal default state)—it continued to poll and displayed nothing indefinitely (had to be terminated).

When I turned JS on it could not determine my machine type, got screen reso wrong and it provided incorrect location info (map and coordinates were wrong). Moreover the browser type was incorrect (but then I automatically randomize the browser type on each launch).

I never realized my battery charge state could be read.. The json data file is pretty impressive.

  },
  "battery": {
    "charging": true,
    "chargingTime": 0,
    "dischargingTime": null,
    "level": 1
  },

Interesting, even with a VPN on mobile safari on an iPhone over a carrier connection I get a uniqueness score of 100%. This is a neat tool, but I'm skeptical of its accuracy. I've run similar tests of uniqueness in the past and this just isn't accurate.

Yes, they claim 100% but the hash changes.

I figured it out! I disabled scripts so the code of the website never ran. Now am not highly unique! (Neither am I “not highly unique” though.)

So, what's the solution to all of this? Are there any settings I need to modify to Chrome to not allow certain info to be queried?

yes but then you become a "suspicious user" and you have to fill 100 CPATCHAs every day

at this point browser fingerprinting is a feature, not a bug

To be frank, in my book, relative to inadvertently being fingerprinted and tracked wherever I go, I consider being consistently faced with “let’s confirm you’re not a robot” popups and pages to be a minor inconvenience.

Consider that all those CAPTCHAs are fingerprinting your browser anyway, and probably also your biometrics (through your inputs while solving each CAPTCHA).

I second this. I tried to use Tor browser for a day in place of my regular browser. Many websites wouldn't open and the ones that dud asked me to fill in thousand captchas.

Even the unmodified Firefox browser with a few of the privacy settings turned on break a lot of sites.

Use a different browser altogether. Chrome is never ideal for anyone who cares even a little bit about privacy. Use [Brave][0].

[0]: https://brave.com

On my privacy hardened Firefox, this website does not work. NoScript deactivated, Adblock deactivated, still does not work.

Is there any library I could use for HW finger printing? I'd like to detect and ban evasions.

https://github.com/fingerprintjs/fingerprintjs

https://github.com/abrahamjuliot/creepjs

https://github.com/thumbmarkjs/thumbmarkjs

Why does this have a domain of .ai, what exactly it is doing AI related?

.ai is a ccTLD. Being AI related is not a factor to get one.

But they are considerably more expensive than more common TLDs, so if you’re getting one you presumably want it specifically and understand the association users will make.

Or it could be that the .com domain was already registered and unavailable, so they started browsing the other TLDs to see where they could find something and felt like .ai is new/hip/trendy

Which we know is not the case here, if you just visit the domain (instead of the submitted subdomain).

https://goldenowl.ai

This is very much an AI-centric website.

Maybe, but there are still many reasons to get one and it does not make anybody less legitimate than AI startups (which was the parent’s point).

Besides, they do sell AI-related services.

> and it does not make anybody less legitimate than AI startups (which was the parent’s point)

Was it? I’m interested in what exactly in their post makes you say that. I see confusion, not any accusation regarding legitimacy.

> Besides, they do sell AI-related services.

I know, I checked the main domain. My point was simply that if you spend extra money on a domain which has a strong association with something, it would be expected that whatever you put on it is associated with it (which indeed is the case). Otherwise you’d be wasting money and confusing potential users, which isn’t generally good business practice.

kinda wish I didn't visit this website as I don't know what goldenowl privacy policy is

Fingerprint Collection Failed

Yay, I am safe. I use Brave. Everyone should use Brave.

Brave Browser is not detected, lol

My browser is in a constant battle with Cloudflare but HN loves them...

Reply on news.ycombinator.com