Privacy Guides
Remove 'Test your Browser now' Button #430
-
|
This button leads to https://coveryourtracks.eff.org/, the problem is it checks our fingerprint against only 215,000 browser fingerprints, missing out more than 99% of all the browser fingerprints on this planet. We can't really know from this if our fingerprint is actually unique or not! Hence, it's better not to fall into incomplete tests, as half of the knowledge is always more dangerous! Fingerprinting can't be prevented to a point, until we implement something like 'Fingerprint Randomization' or 'Fingerprint Generalization', which Brave and Tor Browser do respectively. |
Beta Was this translation helpful? Give feedback.
Replies: 0 comments 20 replies
-
|
I do agree that the datasets that these sites have is tainted but I think that the point of including a "Browser Tester" is in seeing whether there might be anything sticking out in your browser fingerprint. |
Beta Was this translation helpful? Give feedback.
-
|
Like @NSABackdoors said, https://coveryourtracks.eff.org/ is just a browser tester and it's results are not conclusive of real fingerprinting resistance.
How did you know 'Fingerprint Generalization' works to a certain point in these browsers? You used a browser tester like https://coveryourtracks.eff.org/, right? Instead of removing the button, a warning could be placed in the section stating that the results are based of only 215,000 browser fingerprints and might not be accurate. |
Beta Was this translation helpful? Give feedback.
-
No. Fingerprint Generalization only works in Tor, it makes all the users look like the same. They mention it on their home page and you can find in countless other websites telling about Tor's fingerprinting protection. Brave provides protection with fingerprint randomization, which makes your fingerprint appear different to different websites, so you can't be tracked. Safari also makes more users to look like the same. Read here |
Beta Was this translation helpful? Give feedback.
-
Might not? It's surely not accurate! There are billions of browser fingerprints. It even doesn't forms 0.1% of all the browser fingerprints. |
Beta Was this translation helpful? Give feedback.
-
You're right. My wording was wrong. Anyway, as I said before, https://coveryourtracks.eff.org/ is a good tool for comparing browsers and I feel that the button should not be removed, instead a disclaimer should be placed like seen here. |
Beta Was this translation helpful? Give feedback.
-
|
Yes, we can include this table and put a disclaimer! |
Beta Was this translation helpful? Give feedback.
-
|
Regarding this, we did have plans to rewrite the fingerprint section privacytools/privacytools.io#1257 as it is woefully out of date. |
Beta Was this translation helpful? Give feedback.
-
|
Perhaps instead replace it with https://abrahamjuliot.github.io/creepjs/ ? |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for sharing, thats a really neat tool. |
Beta Was this translation helpful? Give feedback.
-
|
Deviceinfo.me also seems to be a Valid one with a Better & Clear Representation of Data. |
Beta Was this translation helpful? Give feedback.
-
|
Deviceinfo.me has unfortunately some errors in their detection methods |
Beta Was this translation helpful? Give feedback.
-
|
Can you list some of them ? |
Beta Was this translation helpful? Give feedback.
-
Are flawed at least for some browsers. Others experienced this, too: https://www.reddit.com/r/PrivacyGuides/comments/qix8tm/please_help_me_figure_out_what_is_blocking/?utm_source=share&utm_medium=web2x&context=3 |
Beta Was this translation helpful? Give feedback.
-
|
I am against simply removing it. Instead the topic browser fingerprinting should be covered in more detail, since it is important for privacy and way more advanced testing sites should be covered. The limitations of Cover Your Tracks in particular and other test sites in general should be made very clear. I don't see that much of a problem in the low number of collected fingerprints, which is obvious for most users, but way more in how strongly biased this data set is to a subset of privacy enthusiasts, which totally screws up a lot of the entropy values and is not obvious to most people, since they don't understand entropy. I also see a problem in the limitations of the testing methods of this site and people tweaking their browsers to get better entropy values without proper knowledge. |
Beta Was this translation helpful? Give feedback.
-
|
Just quickly read through reddit and seeing how many people are frightened by CoverYourTracks and AmIUnique results and trying to tweak their browsers, maybe a quick, temporary solution would be to delete the "Browser Fingerprinting" section, until there is a solidly written section about this topic. |
Beta Was this translation helpful? Give feedback.
-
|
I think we may have to do that, it's been horribly outdated for some time |
Beta Was this translation helpful? Give feedback.
-
|
A temporary removal of coveryourtracks seems to be needed, as (contrary to the enough warnings) some (many) people who come and see privacyguides.org believe its articles word to word. Yes, most are useful and I'd say it's good for most cases, but information can get quickly outdated. So, until a good warning/separate section is added it might even be better to remove, or at least hide(in a foldable section) it. |
Beta Was this translation helpful? Give feedback.
-
|
It's done. |
Beta Was this translation helpful? Give feedback.
Done.