Cloudflare is known to use fingerprinting to detect scrapers For example, they use JA3 fingerprints and match them against the UA to block stuff like cURL while allowing OkHttp (Android clients) - but this can be easily be spoofed with packages such as CycleTLS [1].
I don't want to defend them, because they gate away a good chunk of the internet with their "bot protection", but unless you do PoW (which is also ecologically a nightmare), probably fingerprinting is the way to go - completely destroying the privacy of everyone involved.
Cromite, a privacy conscious fork of Chromium for Android, has constantly issues with CloudFlare Turnstile [2] because they (Cloudflare) try to fingerprint it in multiple ways in order to pass the challenge. The only way to get it to work would be to join the CloudFlare Browser Developer program - which requires signing an NDA. Rightfully so, the project maintainer didn't want to do it.
If you want to see the extent of what CloudFlare does to fingerprint the browsers, just have a look in the issue [2] and see which flags need to be disabled in order to allow CloudFlare to pass the challenge.
I understand both sides, but at least CloudFlare could be flexible enough to fall back to PoW instead of just blocking people from sending forms or accessing websites...
> I don't want to defend them, because they gate away a good chunk of the internet with their "bot protection"
They also gate away a good many people with their "bot protection". I am extremely worried about how so many seem to have outsourced the control over who can access their websites to a company, with no second thoughts whatsoever.
They sometimes have to comply with legal requests (which I understand), but at the same time they have a huge market share - which means that the internet is becoming less and less decentralized and more in their control. We've seen the effects of that in previous outages...
>I am extremely worried about how so many seem to have outsourced the control over who can access their websites to a company, with no second thoughts whatsoever.
I think the Web is on its last legs, anyway. Generative AI and LLM-instead-of-search has destroyed what little value remained.
> I don't want to defend them, because they gate away a good chunk of the internet with their "bot protection", but unless you do PoW (which is also ecologically a nightmare), probably fingerprinting is the way to go - completely destroying the privacy of everyone involved.
Bot protection with fingerprinting is just an illusion. Any signals like this which is on client side can be spoofed by an above average person. Fingerprinting is just way to consolidate the market for advertising business. Assigning Reputation to residential IP addresses and commercial blocks is is another approach to achieve the desired result. Providers would be a lot more careful to allow their IP addresses for misuses, however turns out that it would bring down the DDOS business on both sides, attackers and protectors.
Ironically, more than often its the same companies that invest in building their own bots and finding ways to stop bots from other companies.
it's all for nothing, because Cloudflare's scraping protection works about as well as a $5 padlock - good enough to dissuade bored teens, not good enough to dissuade even an amateur burglar. if someone wants to scrap your publicly visible data, they will. there's nothing you can do.
Exactly. I’m constantly amazed at how little you actually need to bypass CF, Amazon, Azure WAFs and so on (Incapsula springs to mind too). When you look at the code you’ve come up with, it’s actually quite small and compact.
More to the point, these systems actually help scraping because proof of work unlocks essentially unlimited scraping, in my experience.
That said - from my experience on the other side, sure you can’t stop people like me or you, but you can stop 99% of the others. That’s more than worth it operationally.
> but unless you do PoW (which is also ecologically a nightmare)
Can you expand? I don't see a problem with some napkin math.
5W load for 2 seconds is 0.002Wh (we have to let smartphones pass and not by doing PoW for 10s of seconds). 8 billion checks a day for a year = 8GWh.
I stand corrected. It's not a nightmare scenario (as for Bitcoins) - but I'm still of the idea that "useless" computations should be avoided (as we should avoid having 10MB websites).
In any case, according to some napkin math done by Kimi 2.6 (which by itself is probably already consuming more than all of my PoW challenges for the upcoming 5 years) - the situation looks incredibly in favor of PoW:
https://www.kimi.com/share/19e7ef40-a432-8912-8000-0000b4a71...
Which makes me wonder why CloudFlare isn't switching to this already
You can use Firefox with different profiles and configure it to launch particular profile directly, without launching default profile and using about:profiles.
Firefox with a non-default profile can be created like that:
./firefox -CreateProfile "profile-name /home/user/.mozilla/firefox/profile-dir/"
# For, say, cloudflare that would be:
./firefox -CreateProfile "cloudflare /home/user/.mozilla/firefox/cloudflare/"
And you can launch it like that:
./firefox -profile "/home/user/.mozilla/firefox/profile-dir/"
# For cloudflare that would be:
./firefox -profile "/home/user/.mozilla/firefox/cloudflare/"
So, given that /usr/bin/firefox is just a shell script, you can
- create a copy of it, say, /usr/bin/firefox-cloudflare
- adjust the relevant line, adding the -profile argument
If you use an icon to run firefox (say, /usr/share/applications/firefox.desktop), you'll need to do copy/adjust line for the icon.
Of course, "./firefox" from examples above should be replaced with the actual path to executable. For default installation of Firefox the path would be in /usr/bin/firefox script.
So, you can have a separate profiles for something sensitive/invasive (linkedin, cloudflare, shops, banks, etc.) and then you can have a separate profile for everything else.
And each profile can have its own set of extensions.
They actually have at least 3 kinds of profile:
1. containers - As they say its somekind of sandbox, technically a profile
2. profiles that are accesible through about:proflies, which they had for years, and probably the one you are talking about...
3. New profiles that comes with a pop-up much like how chromium browsers shows it
Odd - they've had that for years, but only on the command line. Wonder if it's different under the hood? They also have firefox containers which also never quite became a first-class feature (you have to install a plugin).
> Plus privacy.resistfingerprinting isn't enabled even when selecting "Strict" "Enhanced Privacy Protection" in the settings, great job there Mozilla.
For good reason. I've run that setting for ages but I kept having to disable it and add workarounds because websites would break in weird ways. Timezones in scheduling websites being messed up nearly made me miss a couple of appointments. There's no way to tell the user Firefox isn't broken without displaying a permanent banner like "if websites are broken in any way or you see weird glitches or your computer's time is wrong or fonts look weird or videos don't always work right, click here to disable fingerprinting protection".
Interestingly, Turnstile breaks with resistfingerprinting but works with fingerprintingProtection, I guess the latter takes this crap into account.
Is there a deal between Google and Cloudflare to make non-Chrome browsers harder to use? The pressure to use Chrome keeps increasing, and the amount of ad filtering you can do in Chrome keeps decreasing.
Speaking from the scraper’s perspective, I like proof of work; a ten year old 96-core server will cost a couple of quid to run for a few hours and will grab an absurd number of pages thanks to the access granted by repeatedly solving proofs of work. Small slick codebases too!
Depends on what type of scraping you're trying to stop. For the dumb scrapers that would try to scrape every page on a git forge (for which there are a bazillion pages for a modest project, because of how the site works), yeah it might deter them enough to stop. For anything high value (eg. reddit comments or retail prices), 10s of cpu time isn't going to stop them.
If it's high value, there isn't really much you can do that will be completely effective. Traditional captchas can often be beaten by AI, or by "captcha farms" where impoverished people are paid pennies to complete captchas. Fingerprinting can be beaten by using a full browser to make the requests. Basically anything you do is just a matter of making it more expensive for bots to access it.
Sure, the whole premise is exactly that proof of work reduces the value of scraping, while having negligible impact on users. If the data is so valuable that bot operators are willing to pay 10s of cpu, then other measures are necessary.
Nevertheless even for these high value cases, you can still argue that it disincentivizes the business model, it becomes less efficient.
It will not scare away bots but 10 seconds of wait (CPU or only a sleep) will turn away many real users. "This site is so slow, I'll use something else." A kind of reverse captcha.
With a tuned cool down period this isn't a problem, especially if you frequent the sites. OpenWRT uses Anubis and usually when I need to peruse their site I'm on a very low-end device. I prefer waiting much more over finding Waldos
But in principle I agree that there's no good answer to this, scraping _is_ useful and I bet most of us here had scraped something, it is AI company and their use of human's material for training without consent and return that led us to this (I know botting exists in forum since forum is a thing but it is easily solved by human moderators and keyword filter)
But after you’ve completed the Anubis PoW challenge for a site, it remains valid for some amount of time.
So it’s not quite as horrible as it sounds.
I have setting up Anubis for my own sites on my todo list. And I wish more people did it too. I don’t really mind waiting a little bit extra every now and then before the page loads. What I do mind is ReCaptcha asking me to click all the pictures with buses in them etc. And especially when I have to do it several times over before it’s happy. I’d rather wait a minute for a page to load than to ever solve a ReCaptcha again, if given the choice.
My guess is its an implementation error, not an hardware limitation. I have two 10-year-old devices and one passes instantaneously while the other halts for a good half minute every time.
One of unexpected outcomes from AI-induced hardware shortage may be that, in fact, compute won’t be getting cheaper and may in fact get more expensive…
Anubis is designed to stop a certain class of badly behaved bots. It intentionally doesn't run if a bot identifies itself with a UA, such as Googlebot, because then you can rate limit it or block by UA and with other tools.
Anubis is active when a user agent looks like a web browser (e.g. contains the "Mozilla" substring every major browser uses). The reverse proxy serves an interstitial page that does a proof-of-work check, validated server side, setting a cookie if it passes.
This means a legitimate user won't constantly get the proof of work check, because they already passed it. But AI bots rotating through tons of residential IPs to scrape your forum or git forge or whatever will be slowed down.
Overall, I like the idea. It's unobtrusive, privacy preserving, and seems to be working out well for a lot of sites.
Bots don't [currently] execute JavaScript or follow complicated redirects.
They don't now, but enough "high value to the bots" pages turning on JS or complicated redirects will simply result in the bot authors adding JS execution or redirect following so they can continue "botting" the sites they want to scrape.
It's a hole with no bottom. Each one-up on the anti-bot side will eventually be handled on the bot side.
I tested this extension that I've been using for a long time on the turnstile page and it got through, fwiw. I think it's a bit more subtle than how resistfingerprinting works but not sure what the privacy tradeoff is.
Doesn't this mean we just need to make the webgl fingerprint resistance implementation smarter? Instead of explicitly rejecting webgl access or responding with dummy data, respond with data that is random within space of N common and reproducible patterns. E.g. emulate webgl implementation of some low spec but actually popular devices.
The last screenshot in the OP article mentions that "a browser extension... adding random noise to canvas data" can be detected. Which isn't to say this perfectly detects all such randomization, but it's certainly an active part of the arms race.
All of those advanced features should be enabled on a per-website basis but unfortunately even browsers whose marketing focuses on privacy don't allow you to do that. Same with TLS root CA certificates, there is no way to configure that a certain CA can only create certificates for certain domains.
>Turns out it's because Cloudflare wants to have a fingerprint of your device via WebGL, the only reason for doing this would be tracking.
> So Cloudflare just banned all WebKitGTK browsers as I guess they put an exception for Safari.
This is false. I ran firefox with:
* hardware acceleration disabled (so software renderer, nothing to fingerprint)
* resistfingerprinting enabled, including letterboxing with default window size
* webgl disabled
* VPN enabled
* In a Windows VM
By all accounts this should be the most suspicious fingerprint ever, but turnstile happily lets me through. If they want to track people, they're doing a pretty bad job. My guess is that OP's browser is getting banned because his WebKitGTK has a weird fingerprint, not because of webgl or whatever.
> Such things are blocked in WebKit, and have been for years. Meaning it's tracking so awful that even Apple would block it, and as far as I can tell it's not the kind of privacy protection you can easily disable in it.
This is also false. Webgl fingerprinting works just fine on Safari. They might try to mitigate it by adding some noise, but that's not so different than what firefox does, and is certainly not "blocked".
I think you're making plenty assumptions yourself..
Official Firefox can be leaky unless you build it yourself with some build-time changes or use a fork with such[0]. Am I guessing right that you still have Webcompat, RemoteSettings, and Nimbus enabled still? How do you know a compatibility intervention isn't causing your browser to open the kimono just enough to "unbreak the page"?
[0]: I'm currently aware of Tor Browser, Konform Browser (am dev), Mullvad Browser, and to a certain extent Waterfox, LibreWolf, and r3df0x doing this.
> My guess is that OP's browser is getting banned because his WebKitGTK has a weird fingerprint, not because of webgl or whatever.
So why is Cloudflare saying the author got blocked because of WebGL?
> > Such things are blocked in WebKit, and have been for years. Meaning it's tracking so awful that even Apple would block it, and as far as I can tell it's not the kind of privacy protection you can easily disable in it.
> This is also false. Webgl fingerprinting works just fine on Safari. They might try to mitigate it by adding some noise, but that's not so different than what firefox does, and is certainly not "blocked".
Yep. Cloudflare and cloudflare's customers don't care about blocking people that use non-standard browsers (or accessible browsers, or feed readers, or whatever). Using cloudflare defaults is basically saying, "Only major corporate browsers released in the last year or two can access this site."
Adding noise to a canvas element is a mistake anyway. It means you can't develop a proper paint program using web technologies because your browser will mess with the image.
Bad optics aside, it doesn't actually reflect reality. See my other comment. You can enable basically all the privacy settings and still pass turnstile. Tor browser in a VM passes it, of all things.
It tripped "Canvas Randomization Detected". See the last screenshot.
Cloudflare's demo page still treats that as a pass, but complains about it. As is often the case with Cloudflare, I expect that they'll then take no responsibility for sites that use more aggressive settings.
...in the age of AI, does anyone have an actual solution for keeping out bots while preserving the privacy of humans?
Obviously this is terrible, but I think there's a possibility it's the least terrible option? Another option is IP reputation, which I think is worse. Or scanning a code with a non-rooted phone, which I think is even worse than that!
The only solution is regulation. If all content created by anyone has a copyright, how does an implicit opt-in (which is what happens if you don't create a robots.txt file for your website) for scraping make any sense? Moreover, even if you have a robots.txt, AI (or whatever) bots often don't respect it (or use workarounds - they outsource scraping of such "restricted" sites to unethical third-parties to get the data; Meta has even resorted to piracy, openly!). So clearly, the logic and the "honour system" has failed.
Cloudflare, Google Captcha, HCaptcha etc. are all shitty technical solutions because, as we are all discovering, it comes at the cost of our privacy (i.e. our personal data may monetise these services) and / or our computing resource and time. If current copyright laws aren't sufficient to prevent this, we have to acknowledge the system is broken. The answer could be enhancing it with some kind of Digital Millennium Copyright Act (DMCA) -like laws, but in favour of the creators against BigTech or rogue actors.
Or you could let information be free, at least the stuff that’s on the public net.
As for issues like bots overloading websites or using too many resources scaling laws will take care of it quickly, it’s not like you can’t serve thousands of RPS from a Raspberry Pi these days.
I don't think regulation will stop web scraping, not least of which because it can be done from locations outside the jurisdiction of the regulations.
> we have to acknowledge the system is broken
The system is broken. It probably takes, what, 10 seconds or less to use a residential or foreign proxy, 6+ months to internationally track and prosecute a single offender? So like a million times more effort going the regulatory route.
Just as criminal laws don't end all crimes, copyright laws and anti-scraping regulation won't end all scraping. But it will greatly reduce it and limit it to rogue actors. Two examples I can cite here are the laws against email spams and laws against unsolicited marketing calls - they had a definite impact in reducing both (even in India, from where I am, where implementation of laws are often lax).
Remote attestation should still be possible with a rooted phone if phone manufacturers weren't so shit. If the attestation happens at hardware level, it doesn't matter what programs or kernels you're running.
And identifying a bot that is acting on my behalf. Claude go search this topic is basically the same as Googling something and clicking on the results. Human driven AI searching needs to be in a different box than AI scraping for training data.
Hopefully it stays that way; "a bot acting on my behalf" is still a bot. At least it's often a well-behaved bot and uses a user-agent that can be detected and blocked.
You don't need a non-rooted phone to pass captcha checks, I have a rooted phone and can pass the captchas that ask you to scan a qr code. But I doubt phones without google services would manage.
They are not a problem unless you "believe" it is a problem. I estimate around 20-25K hits to my website from bots per day and I have all cloudflare protections disabled. Any decently optimized server should be able to easily handle that. (it's roughly 1 request every 3 seconds).
Yes and that is just the bot background radiation of the internet. I run a primary source of information site and these botnets are aggressive to a DDOS level. All to do some sort of scraping. Because they have sophisticated enough tactics to DDOS us if they wanted to. However I am not sure their objective as they have wasted enough of our resources to have scraped all our content 1000s of times over. That 25k traffic is a couple of minutes for us. And that adds up. 80-90pct of our traffic is this
What resources are you concerned about? An n100 minipc should be capable of serving something like a blog at 20k+ requests/second (or saturating its network).
Do it like plane tickets do, tie a ticket to an identity + buyback up to a week or so before the concert in case someone wants to cancel (or authorize the transfer and capture only a week before). Ask for ID and ticket at the entrance.
I'd simply check filling speed, even with browser's autocomplete humans are slow due needing click submit.
Then when it's "processing", do them in bulk and prioritize slower users. There's huge opportunity do bot checks after checkout without affecting user experience.
Also on product launches you could add unique field which requires user to input, for example that way bots can't prepare for launches.
Firefox has so much built-in tracking it seems they want to push me to build my own browser. For example every time you open the settings there are several ways they are sending out pings to certain extensions.
Also by default addons.mozilla.org is a privileged site so of course they include google tracking in it and they get the proper fingerprint no matter what you have configured.
If you are this motivated (I am!), how about joining forces on Konform Browser? Radio silence and remote third-party integrations disabled by default and generally sane and conservative defaults respecting old-fashioned notions like individual consent and data-protection regulations.
I wondered about that too. So they allege that bots
require that everyone now has to ID to the big service
providers. Very dystopian situation. Skynet is currently
winning the war.
A better solution would be to make webgl, webgpu and (especially) webrtc have some sort of prompt before they can be in any way used in that fashion, but this will absolutely destroy web ux Windows Vista style.
It's about explicitly deciding to allow certain capabilities on a per-website basis. No major browser allows defense-in-depth via fine-grained website permissions.
Even simply changing the user agent was sabotaged at Firefox, and choosing one user agent per domain is wishful thinking.
Fingerprinting is just an implementation, banning it will just drive these companies to invent new tricks. That's why the GDPR doesn't specify any technical tracking methods, whether you're using cookies or fingerprinting or a camera drone looking at the user's screen, tracking without consent or good reason is banned.
I doubt politicians care much about fingerprinting, though. They're more afraid of actual businesses getting attacked by bots than they are about Linux users with weird setups not being able to access some websites.
Your very sarcastic, uninteresting comment getting downvoted is not an indication that forum isn't intellectual. It's an indication that you aren't behaving intellectually.
