A news site doesn't need to do audio processing on my machine and read back the buffer? It doesn't need to use webgl to render triangles to an offscreen buffer?
Broad geographical targeting shouldn't be an issue. In fact, it's so easy that I think we should invent an ad-targeting API people can opt in to. If I could tell advertisers my geographic location, gender, age cohort, languages, and 3 interests, I'd not feel that's an invasion of privacy when they use that to deliver ads. I'd just get lawnmower ads instead of shampoo ads and that's fine.
The problem is, they don't want to use that kind of targeting. They want to target me with ads for a travel destination that my coworker mentioned in a Messenger message to another coworker five minutes ago. But we should just not accept that.
It's the browser that introduced many of these APIs in the first place.
That should answer the question.
(If you ask me if a browser should report how many cores the machine has, including physical and logical ones, I would answer never, every time. But not everyone thinks that way. https://developer.mozilla.org/en-US/docs/Web/API/Navigator/h...)
Because it's known that interrupting the user-intent with a unrelated consent-popup doesn't help.
Older people may remember that this was the approach of Java J2ME, where you couldn't provide a fluent user-experience as the user first had to answer huge popups about individual API-permissions with "Always Allow", "Allow now" etc.
You go to Facebook to see your cute nephew make spit bubbles. Do you care that they access your camera and canvas? What is a canvas? And sure access my camera, I want to upload photos!
It's fundamentally broken. They've studied the dark patterns, not to avoid them but to implement them best.
If it's a site I trust, and I think it genuinely needs audio, I give it audio. If someone doesn't understand and just does allow all, they never see it again either (and their ads are suspiciously well targeted).
We already have these consent popups for camera and microphone. Reading back bytes from an audio buffer seems like such a rare thing to do that it hopefully just stops happening.
We do this to check whether the browser is what it claims to be, not to track who you are. Similarly, font rendering, floating-point math results, WebGL renderer strings, and whether built-in functions are patched is all much harder to fake. These considerations determine how hard the proof-of-work puzzle should be, and site owners can see it as a risk score. We don't do cookies or browser storage, IPs aren't stored, and nothing gets used to recognize you on another site or on a later visit. The biggest players in automated traffic prevention can't (and usually don't) say this much, because of conflicts of interest between automated bot detection and their other products (e.g. ad sales and IP-profiling).
Your example is really fair though. I chatted with our engineers, prompted by this thread, and we found nothing in our detection that currently uses the audio hashes we collect, so we've started talking about removing the collection.
For example, on multiple instances, my spouse declined "location permissions" an app designed to work with a piece of Bluetooth-connected hardware, and would blame the hardware as being crappy.
"This smart lock app doesn't work."
"You need to enable this permission."
"Why do I need to give the smart lock app location permissions?"
It's not obvious to people that "Bluetooth can be used sneakily by big-enough companies as a makeshift GPS" even if the app needing Bluetooth access doesn't actually need a GPS location.
We should go further. A web-site with mostly static content should not use user-side scripting (Javascript) at all, it's unnecessary for it. It should be required only in rare cases where a site is basically an application and require manually turning it on each time such application is opened.
Then sites that want to track will just add some JS. Seems unlikely to help.
If such client-side behavior counts as “basically an application,” then “applications” is a larger set than seems useful.
If you don't care about your site being static then you can track it server side, but that has big implications for his you can just your site. Sticking a little flag in localStorage really is the best solution, if you can get past ideological purity.
Simple answer: The makers of 3 largest browsers by market share (collectively 90%+) are buddies with the people who want to fingerprint you. Only the 4th largest gives a shit.
I agree with you but really from a point of privacy and what/how data is allowed to be mined, shared, sold, etc. The implications to the business/ad industry are second order effects that are just collateral damage.
Last I checked it's because it was generally considered that preventing fingerprinting is practically impossible anyway. It's not like you need a lot of entropy for fingerprinting, and the web API surface is enormous. To actually stop fingerprinting you'd need to make your web browser more or less useless.
How so? There seems a fairly inherent tension between having things just work by default and hiding surface area for fingerprinting behind permissions pop ups. For example say you go to a maps website, how do you decide whether to share the users location with the website? Or enable webGL functionality to render some fancy layout on the screen? As a user having to enable all that stuff per-website is annoying.
Maybe there's some other way Im missing?
Google's business model has been ads from the beginning. Search, gmail, drive, office, chrome, everything they do is just to drive ad revenue further.
They don't care about the user. Never have, and never will.
And the outcome we are seeing is what happens when a defunct state isn't able to protect citizens from conglomerates. That's what the Kartellamt or the FTC is supposed to be for.
The company as a whole also really stopped caring at some point and are now following the “do be evil” paradigm
Not sure if I can't see this comment as anything else than hypocritical. It's always easier on the money receiving side, isn't it? How do you think the term malvertisement came to be?
AdMob is kind of the definition of inevitable Surveillance Capitalism and the prime example for necessary regulation through government.
On the web one is drive-by fingerprinted by sites they never once approved of or intended to visit thanks to the browser’s ability to redirect and embed without input from the user. The magnitude of the problem is far greater on the web.
To reframe it, in a given week the average person only uses a small handful of apps, but in the same span of time their browser has connected to thousands, maybe tens of thousands, of unique domains, most of which are tracking them in some capacity and the majority of which are tracking things most of us would consider invasive.
"You're protected from the methods I use here. You disabled JavaScript, so this fingerprinting trick can't run at all. You're still vulnerable to network-based methods. Read on to see what it does to everyone else.
From now on I’ll call you:…"
I've run sans JS as default almost since its inception in the Netscape days and I've very little difficulty without it. On rare occasions when I'm forced to use JS I'll cut-and-paste the URL into another browser.
I no longer use ad blockers as I just don't see ads with JS disabled. The other privacy precautions I take are the obvious ones such as no Google or social media accounts, flushing cookies at session's end, regularly randomize my IP and never let Chrome anywhere near my machines and make judicious use of multiple browsers (this is being written in Privacy Browser on Android).
Using NewPipe and sponsor block on PipePipe increases privacy and makes YouTube ad free.
Client-side scripting was a mistake which was obvious from the outset (or not severely limiting the reach of scripting engines).
What I find hard to understand is why alternative plugin JS engines haven't been developed that are under user control. Spoofing, randomizing and returning junk data to webservers would go a huge way towards stopping fingerprinting by websites.
I guess you use the web in quite a limited manner? (Or use that other browser quite a lot...?).
I've tried no JS various times over the years and always end up constantly whitelisting sites. It's even worse in 2026 and gets worse each year.
Banking, insurance, webmail, groceries, eBay, CAPTCHAs... not to mention what feels like 50% of the web that is behind either Cloudflare or similar, which almost always require Javascript.
Unfortunately this idealistic principle (no JS) attracts the worst people.
You're on Tor? You have JS disabled? I might as well warm up fbi.gov/investigate/cyber for the probable report I'll have to do...
"Unfortunately this idealistic principle (no JS) attracts the worst people."
Again, how did you come to that opinion? Where's your evidence/stats? Generally comments like that come from web/JS programmers who've realized that client-side users have them stymied when they disable JS.
An additional reason I avoid JS is speed, webpages become blinding fast when there's no JS. Once one gets addicted to the great rendering speed there's no return. I'm not against JS per se, trouble is you JS programmers have weaponised it against users, webpages of 7, 10 and even 20MB are now commonplace. Presenting such slow massive pages to users is nothing other than abuse.
I do not use Tor.
BTW, I should point out that my views about JS and privacy abuse were formed from my years working in govt. surveillance where the stakes are very high.
That does not change what I have dealt with. I am not a “JavaScript fanboy” remember I made a community that could work WITHOUT needing JavaScript and that lead to a bunch of grief for me which made me understand the extreme naivety of that original idealistic idea on the internet.
> language
I dont think js is needed for either of those things.
If we control the servers the clients are connecting to, I don't see how there could be much hope of hiding. We can spoof things like latency toward an upper bound, but we can never ever defeat the lower bound. Physics is still quite insistent about that one.
If we control multiple servers, we know where they are, and we can periodically compel our clients to rotate between them (see: DNS), we can engage in latency-based multilateration to put them on a real-world map. Depending on the number of servers and samples over time, this can be remarkably accurate. Determining someone's actual postal code just from network timing information is feasible.
VPNs obviously confound this in first order terms, but if we aren't careful they can make it even easier to identify our activity in other ways. We are trading concealment in the woods for concealment in a crowd. Both have their advantages and disadvantages. The VPN might also have a side channel that we aren't aware of.
https://www.usenix.org/legacy/event/nsdi07/tech/full_papers/...
Is there any reason a browser or plugin couldn't just spoof the data used in fingerprinting to random values? So it emits different values for each page or resource you load?
Spoofing 'Newly installed most recent version of Google Chrome on the latest version of Windows 11 in the US' without leaving any holes revealing something fishy is surprisingly hard, but that or 'Safari running on recent-ish iPhone running up to date iOS' would be the ideal target to blend in with the crowd.
It doesn't matter if it looks wrong as long as it looks differently wrong over time. Don't bother trying to blend in with the crowd because you can't ever expect to catch everything that can be used to fingerprint you. New techniques come out all the time. You can block a lot of information from even being collected. Limit what data you send at all, randomize what's left, change IP frequently (VPN) and you should be fine.
For a more proper solution, you could try to convince Mozilla that granting every website uncapped multi-threading and direct GPU access is a bit much. If webgl/webgpu truly is a must, then maybe each website permission should start with a single-threaded software-renderer that behaves indistinguishable on all supported platforms. And only after granting the website that extra permissions let it determine that a more powerful GPU is available.
https://jshelter.org/faq/#what-jshelter-configuration-should... :
First, see our blog post. Consult also other blog posts and other questions in this FAQ.
If you want to have the same fingerprint as many users, We suggest going for Tor Browser (do not install JShelter in Tor Browser). If you want to make cross-site fingerprinting linkage hard, go for the Recommended JShelter level. If you want better protection for the real data at the cost of having the same fingerprint on different sites, go for the Strict JShelter level. Keep NBS active. If you want to detect and prevent fingerprinting attempts, use FPD.
On the technical level there is no problem to block tracking - intriduce noise and make APIs unavailable until user permits to a web site. But that is not implemented.
The same with native applications. If your bank app requires you to share your location you can't spoof it. The makers of Android and iOS made spoofing of geo location impossible - you don't have that control, which means the phone is not your device. You have to give up your real location.
That's basically the ideal. It's become obvious that it's impossible to try to be identical. Web browsers are too leaky, and something will out you. You really don't want to try to blend in the "typical" users because none of them use ad blockers.
>coveryourtracks.eff.org shows you how unique your fingerprint is, with entropy for each measured data point.
Something to note - if a fingerprint is generic enough, that might be a _good_ thing!
Having worked with fingerprinting, anti-fingerprint browsers are very hard to properly identify. The ways to get around it are things like collecting more coarse fingerprints[0], as the overall one is fairly easy to change (e.g. make your browser lie about a collection of browser apis).
What's imo more difficult though is a fingerprint like:
- iPhone <new_version> - running os <latest_stable> - using default safari - bought and used in <common_location> (e.g. usa)
on a 5g mast in a decently sized city, you'll share a fingerprint with at least one other active user using a setup like ^
grapheneOS on the other hand stands out a bit more!
[0] - https://blog.castle.io/fingerprints-beyond-device-ids-engine...
Just because someone can theoretically demonstrate your fingerprint today doesn't mean you have the same fingerprint in a few days. It's that kind of stability that you actually need for consistent matching for advertising or tracking purposes.
90% of commercial fingerprinting services still rely mostly on the IP address. Once you remove the IP address, you're left with a set of configurations that have a very short half life - volume settings, installed fonts, software versions, etc. You can expect a good portion of your audience to degrade within a couple of days. (Whereas a cookie has a life of 2 years - practically though they only survive a few months).
What you actually pay for from third party vendors is a connection made around a more stable data point - a phone number, an email address, a user ID, etc. With one of these data points you can actually track people across multiple devices for as long as you want.
Suddenly the credit report was no longer a prerequisite to the mortgage.
But for real. You say you no longer live there, so I'm assuming you left for more opportunities. My wife and I are interested in countries with opportunities for engineers and privacy. Bonus points if there is engineering research opportunities.
Not asking polemically, I just have never studied economics and wonder if my intuition here is flawed
Some societies prefer huge risks for huge benefits, in those one (might) need a more detailed credit reports because everything is faster, people accept huge risks and the situation can suddenly change.
Some societies prefer more stability. Lower risks, lower surprises. Less opportunities, everything is slower, people can't even take that many risks, simpler histories will do th trick.
From what I notice though, housing seems to be a problem in both types of societies, irrespective on how credits are given, because there are other things involved (regulation, NIMBY, housing used as value, airbnb, geography, etc).
It feels onerous as an idea because bank statements, monthly bills, payslips, will bequests, stock vests and trades, bank robberies*, and investment income each have different ways of being tracked. At least most of these records of individual financial activity come with a PDF these days. It wasn’t that long ago that applying for a loan meant scanning in papers for your lender to review.
Computers are smarter now. Not only can they read documents (or at least match the shapes against shapes they’ve been told are real) but they can also understand the contents too (or at least match the words against ideas they’ve been told are real.)
There may well yet be a future Data Valhalla where individuals hold their life.zip themselves, and upload whatever bits of it are needed to get that Nordstrom store card / car financing / cheeseburger loan without having an intermediate credit agency.
*I feel there’s a kernel of a joke there about a bank robber asking for a receipt with the loot in anticipation of declaring it on their tax return.
Credit isn't like insurance where these things can be checked later during a claim (which can then be denied) - you need to know the person is a bad-payer before you send them the ten thousand dollar loan as that affects the terms, not afterwards when they default and you can't get your money back.
Potentially, with loans or other things, some kind of technical solution could work. You could have a personal document that you control, and you give permission for companies to write to the document.
With a loan, for example, if I give you a loan, I can also append to your document that I've loaned you the money; for every payment you make, I write to the log to say you've made a payment. If you stop making payments, I stop writing to the log. This would be cryptographically signed by me, so it can't be forged.
When you want to take out another loan, you can proactively share your document, the bank can see everything written on the log, and make a decision based on that.
For the latter it’s whether you’ve had a judgment against you, which is a matter of public record.
For the former it’s based on how much money you have and what you spend, including on servicing existing debt. Lying about this would be fraud. In that sense, my proposal does indeed open up a new avenue for fraud but I don’t think that’s the same as it being unworkable!
The Western governments have horribly failed of caring for their people - cost of living are exploding, and at the moment way too many people are papering over that with unsecured credits backed by their data being free to exploit, and everyone else not willing of doing that is increasingly locked out of existence.
You hit the nail on the head with the first statement. It required humans in the loop. It scaled linearly. Local branch managers had relationships with the people in their community, and knew who could and couldn't be trusted.
It was an imperfect system, where a human could be prejudiced against another human, and deny them access to credit.
Sometimes that would get discovered, and the human was held accountable for their actions, or had their actions upheld by a (org-chart-superior) human.
Similar to the imperfect system today, where the models a computer uses could be prejudiced against a human, and deny them access to credit. Dissimilar in that the model or the computer is rarely, if ever, held to account. "Computer says no" is a meme for a reason.
And every now and then we have to rescue banks that are too big too fail at the cost of everyone else, because someone dreamed up a way to increase short term profits that involves giving loans to people who will never pay them back.
> neither your car,
Praising a trillion dollar GDP, but can't afford a used car.
When I grew up I was told to not make debts if not needed. So many older people did indeed save up until they could afford a car or a house. Education was always mostly free where I lived.
Nowadays it seems like ordinary people would need to pay off a house their whole life and thus also need to keep income high enough to keep paying installments. This takes away so much freedom from people as they are trapped in their high paying jobs...
> we solved the problem with ramshackle one-room buildings with dirt floors,
there is a fallacy here. People with access to debt bid up the price of an asset, like a house, against other peoples’ similar access to debt. Sans debt access, the solution was not necessarily a small ramshackle house. The fallacy is the belief that the same asset price - whether opulent or ramshackle house, wouldn’t adjust to the reduced level from the bidding environment. The differential cost of construction for a nicer house can be countenanced in the access-to-debt case, but that is second order.
Illustrative straw-man example: Suppose, once one had a parcel of land, that it was trivial to construct one’s ‘starter home’ the ramshackle house if you like. Get one’s family and friends to help build in exchange for future return favors. As one advances in life, rebuild to have something more opulent. The rub is in the parcel itself. Unlike the would-be house, it had no cost built-in, just whatever price the previous owner was willing to accept to release their parcel. That parcel’s (and those around it) price reflects the market’s access to debt, or lack of access to debt. In the market without access to debt, parcel pricing reflects which buyers have accrued more savings vs other buyers. In the market with access to debt, parcel pricing reflects who can access more debt (through the various means we know of today - down payment ability yes, which parallels the non-debt’s savings case, someone co-signing, long term expected future earning potential vs someone retired, the parcel itself put up as collateral, perhaps other non-immobile assets put up as collateral, like the family oxcart).
The point is that there isn’t necessarily a 1:1 coupling between market’s debt vs non-debt access and whether the construction is opulent vs ramshackle.
A good number of European countries have only half the poverty rates - the dirt floors and carts you're talking about - of the US yet barely use credit cards.
That does mean banks can't know if you lie by omission and hide some unpaid bills, but they can raise the evidence requirements to compensate and I think everyone deserves a second chance.
https://www.malwarebytes.com/blog/news/2025/02/google-now-al...
There's a fingerprint demo on github called JSCreep (https://abrahamjuliot.github.io/creepjs/) that used to keep your fingerprint profile for 90 days and show you how many times you visited. Worked extremely well, and is still around, but no longer uses the API to retain fingerprint data, so you have to track your hash yourself. It would easily track most browsers across changes in IP address. Only Mullvad, Brave, or Tor would not leave an easily tracked fingerprint.
The demo presented in the royfish.dev link is easily foiled by browser extensions that scramble fingerprint data, the presence of which would itself be detected by more sophisticated profilers and used to define the fingerprint. I can refresh the page and get a new hash and adjective-animal name with the same IP. JSCreep sticks firm to the same hash for me, even with a new IP address and some spoofed browser data.
A lot of data can change over days, weeks, but if you can maintain an accurate fingerprint with tinier changes, and granted you have consistent visits from the user, could you just keep up?
Especially if one service is tapped into several websites, and can continually keep up with smaller data changes and distribute that info. Is this possible? In practice anywhere?
Let's say I'm logged in a famous search engine, they associate my fingerprint X to my account A. If they then encounter fingerprint X anywhere else, there is a good chance it was me, A, doing it. If the next day I get fingerprint Y, it gets attached to A as well and the tracking information follows it.
Does this make sense?
Even easier if, additionally, many of the smaller websites were syndicated into a tracking network?
Indeed. I checked this website a few hours ago and got one fingerprint. Just rechecked and got another.
Sorry, but that's simply not true. It's actually the default way of how people eg track people across installs in app stores ever since they started removing that feature.
Data harvesting gains most of its value because they sell some or all data to businesses which wish to leverage that data in advertising. If they cant advertise they wont need most of the data, which should at the least reduce the problem significantly...
Generally, how can society discover new products and services?
I would be open to having some kind of directory that i could browse at my own convenience, but that's ultimately going to devolve to paid placement.
I'm surprised you can't imagine a society where the consumer would be the one recommending products and not the producer. Any entrepreneur can find one first consumer to test their new product. And that first consumer have access to a great communication network to recommend that products to their social network — it's called "the Internet". Social networks are a great platform for people to share interesting things to their friends, family, colleagues and also to people they don't actually know ("the Internet" can connect you to strangers). Such networks cost little to operate when the content is created by users, instead of producers and their intermediaries (people favors short, insightful written text instead of catchy commercial once detoxed).
Being born in the 1990s, I did experience that "Internet", when advertising networks didn't exist and when people intentionally interacted to make connections and actively look for information. They were happy to tell about things they discovered, with little to expect in return ("influencer", aka mercenary, wasn't a job yet).
I want to learn about products that are useful to me, but not through ads. As it turns out, it is incredibly difficult to learn about new products that are useful, as those do not tend to show up in ads at all and even seeking them out is very difficult.
It's weird how no one has come together to make a website specifically designed around users finding product categories that will make their life better. For example, I didn't learn about height adjustable desks from an ad. Nor did I learn about balance boards from an ad. Or trackballs. Or essentially every other product that made my life better once I realized that product category exists.
What a myopic view. Also, the world is very much filled with shades of gray. I’m certain this dichotomy is indeed a false one.
Yes, and those people are free to visit equivalent of mail order catalog or go to the store or subscribe to consumer goods quality testing journal or to any of the enthusiast magazines about electronics or gardening or whatever where they perform independent testing and benchmarking or look up a video comparing ceiling fans across a century when they feel hot in the summer.
So yes, we should absolutely ban third party advertisements. People will still be able to discover good products.
Maybe I'm just not a typical person, but I have never woken up in the morning and said to myself, "You know, I would just love for a corporation to help me learn about their new products and services today." Never. Not even once. When I want to learn about something, I will actively seek information out about it, and I'm not going to look to advertisements for impartial information. The purpose of advertising is never to inform. It's to bombard and saturate.
We also cant get rid of cancer, it doesnt mean we shoudl close down all hospitals.
The spectre of possible interpretations is wide.
Some see only problems everywhere whatever the benefits might be, some see only new opportunities no matter the tragic outcomes, some see various blend of this and that.
The only winning move is not to play--er, to somehow not to have this kind of game in the first place.
The solution is explicit trust. Not fingerprints for users, but digital signatures for content. If it's not signed by somebody the user trusts, don't render it.
The user is the only one who can be approach that task without a conflict of interest. If somebody wants the user to run code that phones home about user behavior, make them gain the user's trust.
It only works if you re-architect the web into something that's even more censorship friendly than it currently is. I hate ads too, but the medicine seems worse than the disease in this case.
Hospitals are not perfect, people die, with your l;ogic we should close down them all.
Doing nothing and keeping advertising keeps a negative situation alive.
It was never about cookies. Thats what the industry made you think. You don't need to show consent banners for cookies (like auth sessions). You need consent for tracking. This includes profiling. And it doesn't matter what you use, fingerprinting or tracking cookies, the consenst banner is required either way.
Consent is required. The banner is just one mechanism, but not required. A website could hold off on all tracking until a suitable point in the usage of the site (like creating a profile, for example) and ask at that point.
The banner is an artefact of the hostile way most sites choose to try and hoodwink the user into being monetised and ultimately to piss users off so much that they complain about the legislation rather than the lack of online privacy.
Better to just harass them into a habit of dismissing the banner.
They don't seem to call it GPC, but their "ask websites not to track" feature now uses it instead of GPC.
if you write to your reps, they might just change that. The Eu is one of the few democracies left. Slow but powered by the people
The exact definition of "legitimate purpose" is unspecific (which is important to be able to adapt to development of new services) which leads to some lawyers being careful.
I also claim that some lobby groups promote having too many and too big consent banners to move blame from tracking companies to the EU ("stupid EU forcing all those banners!") in order to reverse the policy on tracking, which seems to have some effect.
Generally speaking JS code that makes request to server is covered by ePD while actually processing that request server side is covered by GDPR (there are some nuances to this).
The problem with this view, though, is that even the very agencies propagating these rules do the exact same thing! You can go to any site from an EU arm and get the same banners.
At this point, you can't really blame it on bad-faith actors, you have to accept that it's a bad rule. Which is not to say I know a better one!
Malicious compliance is not the law-maker's fault.
Not to mention the large number of "legitimate interest" exceptions that are absolutely not justifiable.
The main problem with the legislation is its lack of teeth and enforcement.
In fact the law even specifies “no pre-ticked boxes” but every banner always has “accept all”. And it’s 50/50 to see “reject all” even though it is a requirement that denying must be as easy as accepting.
They should pay a bounty for reporting non-compliance.
That's not entirely true. Websites that don't actively try to fuck over their users (because their purpose is not shoving third party ads at you, it's providing information about/a platform for their company's product) often provide a modal with two buttons:
'Accept' 'Decline'.
This, IMO, meets both the letter and the spirit of the law. The funny thing is, for any website actually doing that, if you click 'Accept', your data will likely only be used to improve it.
> Global Privacy Control detected. Advertising cookies are turned OFF.
Yes, I love this. My browser is already sending "do not track" requests.
If you ask my 70-year-old mom to decide whether she'd like to allow the WebGL API, she'll go bonkers. If anything, the constant cookie banners have just trained her to consent to stuff out of muscle memory. You could probably design a "Are you ok to sign over the deed to your house" question in the rough shape of a cookie banner and get 90%+ agreement.
> We Care About Your Privacy
> We, and our 228 partners use cookies and similar methods to recognize visitors [...]
228 partners should make anybody concious about the fact that something is wrong and you probably should not consent to that at all. Also, nobody cares about your privacy when they think sharing data with 228 parties is okay.
Or alternatively how weak the EU messaging was I guess
1) ePrivacy. the OG banner law. requires you to inform the user (but not consent, as I understand it) if you store data on a user's device. includes cookies, localStorage, indexedDB, etc. does not affect fingerprinting 2) gdpr. requires you to ask consent for any kind of user data processing, unless you can find another legal basis
eprivacy was about cookies, but forgot fingerprinting. gdpr was not specifically about cookies. but gdpr (or at least its past interpretation by the dutch privacy authority) didn't require consent for limited first party analytics.
which led to situations where you'd get the impression that you wouldn't need a banner for first party tracking using cookies .. well the gdpr didn't require it, but the older ePrivacy rules still did.
we still have the digital omnibus law package coming up which is supposed to simplify and resolve the overlap between the two above, and reduce banner requirements at least for SMB just analyzing their website visitors
It does. The current version says "storing of information, or the gaining of access to information already stored, in the terminal equipment". Reading e.g. navigator.userAgent is information already stored in the terminal equipment. If you then use that in any way that gets communicated to remote server (e.g. directly or calculating some fingerprint) you read that information in manner ePD describes.
The directive itself has been transposed incorrectly in many jurisdictions. For example from what I understand the Dutch law added an exception (roughly "for obtaining information about the quality or effectiveness of a supplied information-society service, provided this has no or only minor consequences for the user's privacy.") which is not permitted under ePD. ePD exception list is exhaustive and the only other exception can be done via Article 15. The exceptions done via that are related to things like national security and certain criminal prevention/investigation.
Of course that then raises the issue on how you get invalid exception invalidated. I don't know Dutch procedural law well enough to answer that.
If people have a problem with fingerprinting, they need to focus on making it illegal to trade data tied to fingerprints, not trying to make users all look the same (we aren't!).
I have a hard time figuring out why quality values are even included for Accept-Language. They either have one of the preferred values, or they don't.
I don't know about expressing ties. If there really is a tie and the client is just as happy to receive either language of equal Q, then it doesn't matter to the client that one is consistently preferred over another. Perhaps it could matter to the server if e.g. the client indicated that "en" and "zh" are tied, but the "en" content can be served in fewer bytes than it can serve the "zh" content, but it seems like such a niche use case.
edit: RFC 9110 offers an explanation: "Note that some recipients treat the order in which language tags are listed as an indication of descending priority, particularly for tags that are assigned equal quality values (no value is the same as q=1). However, this behavior cannot be relied upon. For consistency and to maximize interoperability, many user agents assign each language tag a unique quality value while also listing them in order of decreasing quality."
Fun anecdote: I remember seeing a screenshot of a website selling a scented candle called "basic bitch", which was machine translated so that the idiomatic meaning was entirely lost. So the translation ended up meaning something like "foundational female dog".
Say I have a remote employee who is only able to access your internal systems via his yubikey. His fingerprint is "Genuine Antelope".
Next week, he accidentally damages his yubikey. He has another one in the drawer, but I need to authorized him. I create a new session for him on the provisioning portal (so he can enroll the new key) but now his fingerprint is "Interesting Elephant". What do I do with this information?
It might be better to log the exact forensics and show the administrator the differences, e.g, "User has a new font called "my_handwriting.ttf" since a week ago", or "appears to be a totally different processor, and the new system has Russian fonts installed".
They should be authorized ahead of time.
> What do I do with this information?
Not much. You can't authenticate a person using their (public) fingerprint information.
Your problem boils down to: "How can I verify someone's identity over the internet". You can exchange emergency recovery codes ahead of time (e.g. in an envelope), you could trust public keys from both of their yubikeys, but they can be stolen.
Yes, but it's a signal! There's a cost to spoof it. A scammer could cheaply start a VM that clones your font profile, but at what cost could they replicate your CPU SKU and GPU combination to provide identical webGL timings?
Other signals include the ability of your employee to show their face on a video call, to sound like themselves on a voice call, the ability contact you from their home IP, to receive official mail at their listed address, to produce a second 2FA token you'd previously registered (but not activated) or repeat a code word you'd exchanged. Any of these things can be spoofed, bypassed, copied, or burgled, but each one imposes some cost on the attacker.
The question is how to surface this information in a way that equips help desk staff and doesn't deceive them into authorizing unnecessary risks.
At {{job}} we did briefly consider rolling out fingerprint.com, but settled on ja3/ja4 at the CDN layer.
It does mean that sites that uses 2FA (so every site nowadays) flag me as a coming from a new device each visit, and I have to have my phone nearby every time I want to use my computer, which is quite annoying. If you use GrapheneOS, you can obfuscate your phone too on WiFi networks (and sites, but GrapheneOS not needed for that, just Brave). You can't obfuscate yourself on cellular networks though, unless you have a ton of sim cards you rotate through.
I would bet money that Brave cannot defeat CreepJS. The only thing I've ever seen do it, ironically, is Safari in private mode.
Also just using your computer normally and visiting regular websites already tracks you regardless of your browser's privacy abilities, and I promise data brokers don't care what browser you use, your overall online behavior is still unique.
And phones!
So what sort of benefits are you reaping from your anti-fingerprinting measures?
> So what sort of benefits are you reaping from your anti-fingerprinting measures?
Mostly pride on knowing that I'm preventing companies from doing to me exactly what I wouldn't consent to if they asked me outright.
Sure its just for ads (mostly). Sure the chances of my data leak leading to a personalized cyber attack are slim (I don't actually believe this but you might). Sure, [niche scenario] [non-zero probability] might be prevented. Sure there is a meatspace equivalent to the cyberspace invasion of privacy.
I wouldn't consent to stalking in meatspace, for any reason, and conveniently to my knowledge no one is stalking me physically. I dont consent to stalking in cyber space either, for any reason, and I do know that companies do, and they make it an unending uphill battle to let them know and force them to not be blatantly disrespectful. So I will make it as hard as possible, within what I think is reasonable, to make it useless or difficult for them to stalk me online.
And yes, online things have affected my physical life, so these niche applications have occurred to me and my efforts have reduced the negative impacts, to a satisfactory level, and I will continue to kick against the pricks, because it works.
A better question would be how are you being screwed over if you never bother to try to stop companies from collecting as much data as they possibly can about your private life? It can cost you a job offer, or cause you to pay more than your neighbor for the same products/services. It can get your data leaked, or sold to scammers. Companies use the data they collect on you for things as trivial as deciding how long to you leave you on hold. The more data they have on you, the easier it becomes for them to use it against you.
Feel free to give companies every bit of personal data they want from you. They'll be thrilled to use that information at any time and in any way they think it will benefit them. Until we have laws against surveillance capitalism the only person looking out for your interests is you, but only if you can be bothered. It'll sure help companies take advantage of you if you decide it's not worth trying.
Anywhere I go in meatspace, people could be collecting data about me and my behavior. That has always been true since the beginning of time. An old lady can sit on her porch and take notes about my outfit, my comings and goings, and every driver who comes to my door. Today, they're saying that Target can definitely hear our WiFi/BT MACs as soon as we get into their store, and they see/track everything on camera everywhere inside. My Groucho Marx isn't going to hinder or change that!
All I can do, honestly, is be judicious about what sites I visit, and be forthright in my interactions. I should cultivate good relationships with trustworthy, reputable companies, and stay away from the rest. I cannot ultimately prevent the buying and selling behind the scenes. We cannot realistically prevent collection of everything possible.
I don't shop at Temu or Alibaba, and I don't search on Yandex. I checked my "About Me" on Amazon yesterday and it's shockingly bland and shallow. However I interact with Google, YouTube, Microsoft, and Cloudflare, it's a foregone conclusion that they know everything. My extensive 19-year history on Wikimedia projects is a shockingly rich treasure trove, if anything just my waking hours of every day since 2007! Your own karma score 22,495 in 14 years is probably a fantastic track record of when you were awake and using a computer. If you're so paranoid about Internet privacy, then you wouldn't be so willingly revealing that stuff in public.
I'm using adblocking DNS again, but awhile ago I sort of dropped the pretense of paranoia, because I understood its futility, and again, life's too short. If evil corpos want to do evil to me based on what they think they know, they can do their worst. I am a Christian man accustomed to suffering and infirmity. I got nothing to lose in this life, except integrity.
Yeah, but you can very easily _limit_ what is collected and sold behind the scenes just buy switching browsers, changing browser settings, or using a couple privacy extensions. A little goes a long way.
Your only option is to never buy anything online and only pay cash.
I probably wouldn't go as far as never buying anything online, but paying cash IRL should be something done by everyone to protect their privacy. There are a few other easy things that can massively improve your privacy with minor or no inconvenience: Linux on every PC/laptop, GrapheneOS on phone(s), uBlock on every browser (or a DNS-level sanitizer), car without cellular modem (pull the fuse), non-Google and non-Apple email (ideally your own server or a small provider in a friendly jurisdiction). Also, don't use social media and text over Signal instead. Of course, TVs and other home appliances should be kept off the Internet. Overall, those are some very minor sacrifices for anyone with a little bit of maturity. It's very far from living off the grid or anything like that.
That's a very early 2000s view of the issue. Ads are not what all the surveillance is for, it's just the publicly acceptable excuse for tracking you that companies love to use because nobody cares which ad they get shown. Being able to build a profile of which websites you visit can provide a ton of information about you and that data is increasingly being used offline. It stopped being about ads ages ago and now what you do online can impact your next job offer, what prices you pay, which policies companies apply to you, etc.
You can never limit all data collection. Going to the store and paying in cash will still have you tracked by Flock cameras. The point is just to limit what you can, where you can. Every scrap of data collected about you is going to eventually be used against you later by somebody. No point in giving them anymore ammo than you have to.
What if "fingerprints" are contained in the values of Cookie: headers
It would be great if a wave of lawmaking started in the USA, because most tech behemots operate from there. I would expect EU to take the lead though IF such a wave would start. The EU has still a fair share of christian/conservative MP's that keeps doing the bidding for companies by pushing chat control, but on the other hand also has some progressive and knowledgeable representatives doing great work, so there might be a glimmer of hope. We need media campaigns and public education though. Political leaders are selected for their political abilities and as such rarely have leadership qualities. If the electorate moves, the leaders will follow it. So we have to start at the bottom.
I understand Google having Chrome send way more info. Why do all other browsers go along with this?
Make it make sense...
Why would it be a good idea to expose unsuspecting citizens to false claims by the sellers?
This royfish blog example, or anyother, including that fingerprint.com gets new value everytime.
That should give some hint about how difficult it is to do in real life, at scale.
Now we have a new (nearly impossible) problem to solve. "How do we prevent bad actors from abusing pretty much any feature of a public system which can possibly be abused, or reliably catch and punish the people who try?"
Don't you people remember how it was done back when web browsers didn't have cookies, or could be configured to reject or ask for each cookie? The server software automatically added a PHPSESSID parameter to every link, containing the session identifier.
If you're going to be a bitchy pedant, you'd damn well better be CORRECT.
Second, despite the dumpsterfire that Mozilla’s leadership has been in a decade or so, it is very robust with privacy and anti fingerprint features out of the box, the containers has also stabilised and works great.
Using Brave EFF reports that I have a randomized fingerprint.
Using Firefox or Edge EFF reports I have a unique fingerprint
"randomized fingerprint" -> that's a bad result, it means the site can detect you're spoofing info
"unique fingerprint" -> this is good, ONLY IF you refresh and it still says "unique fingerprint", because it means the site sees you as a brand new visitor each visit, and doesn't detect any (what advertisers would call) nefarious behavior. however, if you visit again and the second time it says "nearly-unique fingerprint", that's a fail because your two visits were linked
The ideal result you should aim for is "non-unique fingerprint". That means your device is indistinguishable from countless others. Blending in to the crowd is what you want.
See Fingerprint.com, despite Brave's protections you'll likely find it works reliably.
I love that things like this are coming too light. But, I fucking cannot understand how anyone would’ve ever thought this was OK.
They made some improvements August this year
https://brave.com/privacy-updates/38-webgl-webgpu-fingerprin...
I just installed it, I want to know if everyone gets the same hash (same identity for everyone).
Same thing the other way: no need to have a banner for a simple login cookie which isn't being used for privacy-invading tracking purposes.
I’ve checked my network. I’ve monitored egress traffic. There is no residential proxy. I’m not on a VPN. If anything, services should be able to track me entirely by my home IP that doesn’t change that often.
My best guess is that I’ve committed the crime of not allowing third party cookies? Or maybe the crime of not being logged into a Google account 24/7. It’s not browser or device specific. My iPhone is just as likely to fail the cloudflare checkbox as my desktop with Firefox + uBlock.
I haven't seen a YT ad in years while in lurk-only mode (no login), which is nice, but I also see cloudflare everywhere , which is not great.
One of the nicer things about LLMs, is that they arrived just in time to erase my dependency in search.
This should be absolutely illegal, one entity is capable of making my life extra difficult and there's no one to appeal to.
Actually they are about reading from or writing to terminal equipment of user that is not done for "sole purpose of carrying out the transmission of a communication over an electronic communications network" or "strictly necessary in order for the provider of an information society service explicitly requested by the subscriber or user to provide the service". Some safety-related exceptions are possible via ePD Article 15, but those need to be separately added to the local law. That's why things like functional cookies require consent as well.
Of course there is also the GDPR layer on top of this.
The fingerprinting comes from enumerating properties of the terminal equipment and their values, and to do that you have to read them, for which you need the explicit consent. (Except in some rare cases, but if you are that case, your workplace will also have a legal department.)
Additionally, as you said, GDPR adds a lot of additional restrictions, with provisions regarding creating profiles of persons (what is commonly refered as tracking).
The thing the site is doing that isn’t “strictly necessary” is always tracking.
It also limits the exception in relation to e.g. employment.
So a canvas/font/WebGL fingerprint used for tracking needs the same consent as a tracking cookie. The guidelines even suggest passively collected protocol signals can fall under it (para 43), though that's where it gets argued.
The weak point is enforcement. A cookie shows up in devtools and in every cookie-scanner audit. A fingerprinting script looks like any other analytics JS, so it's much harder for a DPA to spot and prove.
...it's something i hope no court will uphold. but we know what happens to european courts even looking sideways at American monopolies working as front for fiveeyes.
imo, eff should be starting trials to stress this instead of blogposts.
The whole cookie thing has always been bs
If you've ever worked at a startup, you're probably familiar with the fact that no one gives a crap about this type of stuff early on
The hell with pretty much anything, the only thing that matters is revenue
People only start to care about this after the fact if you've grown to large enough size
If you think stuff like this or right to be deleted stuff is moving the needle anywhere but at the biggest companies, you've been misled
That's the real painful truth of this stuff, most of the time it's just lip service anyway
Wth is anyone going to do anyway? Go through a resource-draining lawsuit against a 20 person company where it will take a year of discovery just to maybe-sort-of prove they've been mishandling your data?
It's going to take some pretty radical new laws for this type of stuff to ever have any meaningful impact
Fingerprinting itself is also not new.
The article's points still stand, but the trick didn't work this one time.
Furthermore, in corporate environments where boxes are set up the same for many users, there are often, or too often, two different users and two different boxes with the same exact fingerprint.
The whole cookies discussion is a successful diversion by the ads industry. It was never about cookies.
Anyway, it's a simple script for the sake of demonstration. I'm sure you'll get a different hash on other browsers/devices.
This is backwards. The antitrust concerns of removing 3rd party cookies led to them developing alternatives.
I think Google needs to be forbidden from causing more damage to the EU market. And there need to be replacements too, once we forbid this Evil entity from sabotaging EU citizens any further.
We also need real open browsers not controlled by greedy Ad companies.
I am also aware of the irony of using thorium here right now (based on the adChrome code base). If I use Firefox, I can not listen to audio since mozilla claims only systemd-pulseaudio using people can listen to audio. Well, chrome plays audio on youtube (which is also owned by Google, do people begin to see how big of a problem Google has become) just fine, so I have no idea what Mozilla devs are smoking. I could recompile firefox, to just use ALSA, but who in his right mind wants to compile a 1% usage base legacy browser that uses ancient technology to compile its outdated code base please?
See: https://www.linuxfromscratch.org/blfs/view/svn/xsoft/firefox...
Mozilla gave up decades ago on fighting adGoogle. And it can be seen via small things too.
Ladybird may be a competitor at one point in time but they ban people from their github repository, which also sends the message "do not contribute to us because we don't care about you if you have dissenting opinions". Well, so much for real competition, when projects get all emo early on. You can't beat the evil adGoogle empire with wishful projections about "conduct of people". That's not going to work. I don't subscribe to any random ruleset made by random people I don't know and don't want to know when they resort to censorship. But hopefully more and more people realise that something has to be done about Google one way or another.
> I look at all this data and I don’t see people. I see patterns. Patterns emerging out of patterns. A living, digital fractal state of the economy, society and human psyche. The crystal ball of the 21st century and it’s made of silicon! That’s the product.
> And to the highest bidder it goes!
> What does that make you then? A gear? An ox? A volunteer?
This does absolutely nothing for me. If this resonates with you, can you explain?
I can understand worry about a tyrannical government, and if you have something like a gambling addiction I can understand about ads. But the meta-level privacy fear about deidentified patterns - I just plain don't get it.
It's truly totally ok with me, yes.
(I get the non-anonymized concerns, or risk of becoming non-anonymized, that's not what I'm confused about)
Is the implication that the machine knows me so deeply that it doesn't matter if it connect my device ID to my name? That it knows me deeply as a person regardless of what identifiers are used?
If that's what meant, I have to say: 1. That's an absurd overestimation of the current capabilities of adtech 2. Even if it were true, I'm not sure I'd be that bothered by it? What I'm bothered by is what could be done with that intimate knowledge. And if the answer is just "show me the absolute most optimal ads", this doesn't really bother me.
That's why the quote I highlighted confuses me.
On Safari, fingerprinting doesn't survive in incognito mode. Same with Firefox.
Good thing I use different computers for different purposes, though it looks like using different browsers seems sufficient.
Different computers doesn't solve a lot of fingerprinting methods as well.
But at least the moment we start using websites the largest players will know it’s us. Nobody else swipes and types exactly like me on my screen size of iOS over the relay.
a native app with 0 permission prompts gets you more bits than web
Now on a desktop PC sure I believe these fingerprints are highly descriptive. And in android land they probably work okay because of the diversity of products and configurations. But iPhones resist fingerprinting just because they all give the same answers.
The web provides:
time zone language and locale (accept-language header too) iOS version screen size and pixel ratio text size dark/light mode reduced motion ad blocker extensions for safari
An app provides (without additional permissions):
IDFA advertising ID (unique and cross app) IDFV (vendor ID) AdAttributionKit (which apple does obscure somewhat) Battery level, charging state, volume, headphones Various appearance settings again (font, dark/light, reduced motion)
Just slightly more permission granted via prompt can get you:
System boot time and uptime Disk space Keyboard languages and extensions File timestamps
Then additional individual permissions (which people agree to all the time for specific use cases)... at this level they can single you out easily: location contacts bluetooth motion sensors
To Apple's credit they're trying to make fingerprinting more difficult and attempt to ban it entirely, but it's not at all impossible.
Some of these guys sound like they erase their fingerprints off the plates when they visit a restaurant, and ask people to forget about them when done talking.
Like it or not, as web gets more complex, it will recognize you more. If you don't want that, just don't use the web - perhaps buy a farm and life off-grid.
I just don't get what's the appeal to this endless damn cat and mouse game.
On the web you can do that much more efficiently. That scares people.
As in any defense/offense situation, it’s all about balance.
"By the way, if anyone here is in marketing or advertising... kill yourself..."
In particular because "I have nothing to hide" means "I have nothing to hide until now, and because of the corporations and states that have access to my data are lead right now by people I trust not to misuse that data"
But if tomorrow the government change, and suddenly your searches for "gay bars near my location" becomes illegal, you'd rather this info not be in everyone's hands...
You took a guess. You guessed wrong
I have not "disabled Javascript". The software (pl.) I'm using to request and read web pages has no Javascript interpreter
I can also insert a CSP response header such as "default-src: 'none'" and get the same result ("...") even when using software that runs Javascript with Javascript enabled
This "data" does not seem very useful for advertising except perhaps something like "don's waste time trying to target this user as they will not see or hear the ads"^1
What are these "network-based" methods exactly
I never see any ads when using the www. I attribute this mostly to not using popular software controlled by companies or individuals that advocate for internet advertising. This software could be something as complex as an operating system or a "browser" or something as simple as a Javascript sourced/planted in a web page
Whatever these "network-based methods" are, they don't seem to be very effective at targeting me with ads
I generallly stay in textmode. Most internet ads seem to be graphical and/or audiovisual in nature
Hypothetically, if a "developer" tried to target me with text-based ads, I could easily filter them out
I already filter the text in response bodies using a local forward proxy
For example, the "score" on HN comments I submit never goes below 1 in the HTML I'm reading. In theory this helps avoid self-censorship
Here is an excerpt from rewrite.lua
if host=="news.ycombinator.com" then
body = string.gsub(body, "%(<span id=\"karma\">%d+</span>%)","")
body = string.gsub(body, "-%d points","1 point")
body = string.gsub(body, ">0 points",">1 point")
body = string.gsub(body, "width=\"%d+\"></td>","width=\"0\"></td>")
end