Show HN: DevicePrint – device fingerprinting without cookies | Hacker News Reader

Show HN: DevicePrint – device fingerprinting without cookies

21 pointssilverrump50 comments

Hi HN,

I built DevicePrint after running into problems with duplicate accounts and unreliable cookies in my own projects.

DevicePrint is a lightweight device fingerprinting tool designed for developers. It helps identify devices across sessions without relying on cookies.

Use cases include fraud detection, preventing duplicate signups, and security-sensitive workflows.

I'd really appreciate feedback — especially around privacy concerns or edge cases you’ve run into.

Link: https://deviceprint.io

View on HN

It doesn’t seem to be very good, but don’t worry, just keep prompting Claude and I’m sure you’ll get it sorted out.

Jokes aside, it’s cool but it’s not useful if it’s the first time I visit and I see I have 10+ past visits from all around the world… obviously this is not reliable and I wouldn’t use it for anything, much less anything serious.

>just keep prompting Claude and I’m sure you’ll get it sorted out

Anecdotally speaking, this is the case for most new Show HNs now :^)

Why actually try to understand a problem space? Far easier to prompt a turd into existence, polish it up with a cliché marketing page, and collect public validation from your fellow “hackers”

I've got 743 visits from one IP!

I somehow have over 1200 visits from 8 IPs with 99.5% confidence… first time I’ve visited the page lol.

Thanks for checking it out! We’ve identified the root cause of the inaccurate visit counts you saw and have now fixed the issue — the accuracy of DevicePrint should be much better now. If you have any remaining odd results, let us know!

“Hey Claude, users are noticing my product is fundamentally broken. Please shuffle some code, increase the confidence label to 99.6%, and spam the HN thread claiming I identified the root cause of a bug. Frame it as a small edge case. Do not under any circumstances empirically validate the supposed fix.”

Snark aside, I still see 9 previous visits from various countries, down from 900+ previously. It does correctly identify me as being in incognito mode, but if I switch to a normal tab I see a completely different set of previous visits.

AmIUnique.org has a good collection of non-cookie tracking techniques.

https://amiunique.org/fingerprint

I’ve visited 292 times. From Melbourne, Athens, Piraeus, Paris, Munich, Vantaa, Berlin and Kuala Lumpur. I’ve used Chrome, Firefox and Safari on both mobile and desktop.

What’s even more impressive is I’ve made all of those visits from all of those cities in the last few minutes.

You may have a bug.

You've (just) between the speed of my world tour - congratulations

It also has a 99.5% confidence that it has uniquely fingerprinted me, and says I have had 677 visits to the site, including Vietnam & Sweeden...

I think the 'unique' part of fingerprinting here isn't working unfortunately.

Good catch — that was exactly the symptom of the issue we just resolved. The over-counting was caused by fingerprint collisions in certain edge cases. The fix has been deployed, so it shouldn’t happen on new uses anymore. Thanks for the feedback!

FWIW I noticed this was down yesterday while you apparently applied fixes. I tried again just now, and it’s showing 51 visits from a range of countries and IP addresses. I don’t think your fix has fixed things.

Unfortunately this is an example of how AI should not be used.

You have to be able to understand your core technology/IP/logic - I feel that must have been significantly overlooked here.

Totally hear you — we appreciate the critique. We’ve now fixed a real issue that was skewing results, and the current version relies on deterministic signals rather than anything heuristic or generative.

Unfortunately still junk - unless I was just in Jakarta!

At least it's down to only 7 visits from 300 on the last version.

I love services that have not a single person as a contact/responsible for the site. /sarcasm

It's a red flag if you hide behind a contact form with no reachability beyond that whatsoever.

And as other said: 99.5% accuracy means you should have millions of working fingerprints, since mine and others are faulty as hell.

Ah, but it's "Trusted by developers and security teams worldwide!"

FYI, I'm seeing 99.5% confidence I've been to the site 62 times. I can assure you 've been there once.

I think it is called false positive :)

Apparently I went from Germany to UK in 29 minutes, pretty good.

It's a 99.5% declared confidence and says it used 30+ signals.

Assuming you've a list of VPN IP addresses, and travel times between countries, I reckon you should be able to rule out some false positives.

Would be interested to know what the "signals" were that produced the match.

I'm on domestic broadband in the UK (IPv4), according to dnschecker they're on a mainstream mobile provider in Germany. Could be a private tunnel, but those would be rare. Which raises the question of how the confidence rating is made.

I like the general page presentation, a good landing page except that you'll tend to put off everyone who gets a bad result for the example. That might be turned around with something showing "if this isn't you, well done on your browser security" and maybe some compelling stats on confirmed matches from testing?

It has 99.5% confidence this is my 10th visit. I've hit refresh once, but the rest aren't me. My other "visits" are from many countries, saying I've changed browser, IP, and location. They are using the same OS and browser though.

Same, it's all over the place. Whatever it is doing isn't a very strong fingerprint.

Seeing as everyone is apparently seeing themselves having visited multiple times when it wasn't them, including me, I'm very happy with the privacy of this system =) It cannot effectively track me

The aim is great, and this would be useful for many use cases, especially when buying traffic (ads).

But as others mentioned, it is far from being accurate. I got the same as others, multiple visits from multiple countries.

This reminds me of EFF's Cover Your Tracks, the rebrand of Panopticlick: https://coveryourtracks.eff.org/

This helps you see how your browser tries to block or deflect fingerprint and trackers. I miss their "You are one of x,000 users" from the old site but it still gives a nice summary of bits of info your browser leaks and how fingerprinting basically works.

It shows I've visited twice already, from different countries, IPs and browsers. I don't think this works. This open source one does work between incognito and normal session: https://fingerprintjs.github.io/fingerprintjs/

What differentiates this from http://fingerprint.com/?

Visited for the first time and it said I already visited 800+ times with a 99.5% accurancy - not very promising. From the code this also looks like very simple client-side fingerprinting + IP information?

We’ve just deployed a fix to the core matching logic that should significantly improve accuracy and reduce false positives. Feel free to retry at https://deviceprint.io and share your results!

Chrome, Windows 11, apparently, it saw me loads of times, but my first visit was today.

I'd love to use a reliable system like this to detect returning fake, banned, and bot users on my services.

Haha, same here, first visit but system saw me severals times already. I have a quite unique setup..

Works great! Thank you for fighting for users anonymity

Saint-Jean-sur-Richelieu sounds lovely! as does firefox, I should check out one of them at least...

Opening the site from two different temporary tab containers in Firefox yields different IDs.

Yes — browsers isolate storage between containers, so that behavior is expected. Fingerprints will be consistent within a given profile unless significant attributes change.

Thanks everyone for the detailed reports — the strange multiple visits and location jumps were due to a bug in how we combined signals. This has now been resolved, and subsequent fingerprints should be much more accurate across browsers and devices. Appreciate the testing!

I just tried again, from a different browser and different device. It's still showing multiple visits and locations from me - albeit only a few this time, rather than the hundreds before.

Happy to answer any technical questions or discuss implementation details.

Why are docs behind a login wall?

Good question — we’ve taken that feedback seriously. We’re currently working on making more documentation publicly accessible without a login, so people can evaluate DevicePrint before signing up.

People in the comment section has noted the site to recognize multiple page visits from them even though this is their first time visiting the site, did you test your service yourself on different browsers / os / devices ?

Perhaps he's posted this here to get some more test data

Works great, my device visited over 100 times already

edit: not only that, under past visits I can now see the ip address of other visitors, together with their rough location and browser setup. You may want to remove the "gdpr compliant" from the website :)

Don’t worry, it was 0/2 in detecting my browser or OS, like not even close, and I don’t do anything to obfuscate that, so it’s probably accidentally compliant

Thanks for pointing that out — we’ve corrected the display so no other visitors’ IPs or locations are shown. We take privacy seriously and have updated the site to reflect that.

I'm so happy this is bad. Good job!

Thanks for trying it and calling out problems — it helped us find and fix the bug. The issue has now been resolved and DevicePrint should give much more sensible results going forward.

thinks i have visited multiple times when i definitely haven't, did you test this on macs?

AI SLOP [ insert image for "Certified AI BULLSH*" ]

Definitely not working at all.

It shows I've visited all around the world, lots of times.

Nope. Just once, and from one location.

Reply on news.ycombinator.com