Fascinating visualization , as someone building productivity extensions, it is always sobering to see exactly how much surface area the browser exposes.
It's horrifying to see that our browsers give up anything at all. I feel like we need a new model where we just get served the content, and we don't serve up anything to the content provider. I feel ill.
How would you prevent that? If you want to use a feature, you need to know if that feature is available. So you add methods/functions to ensure that you can fail gracefully when something is not available. That availability is the fingerprint. I don't blame the browsers as much as asshats using the tool for malicious purposes. We're not going to ban hammers because someone uses them for things other than driving nails.
The APIs it's relying on to show you that information do implement some anti-fingerprinting mechanisms - things like available memory will be put into rough buckets rather than returning the exact amount you have.
Fascinating visualization , as someone building productivity extensions, it is always sobering to see exactly how much surface area the browser exposes.
It's horrifying to see that our browsers give up anything at all. I feel like we need a new model where we just get served the content, and we don't serve up anything to the content provider. I feel ill.
How would you prevent that? If you want to use a feature, you need to know if that feature is available. So you add methods/functions to ensure that you can fail gracefully when something is not available. That availability is the fingerprint. I don't blame the browsers as much as asshats using the tool for malicious purposes. We're not going to ban hammers because someone uses them for things other than driving nails.
This doesn’t seem right. 1 in 6.2 billion for Firefox on an iphone, not including ip/network uniqueness?
Seems more plausible that anti-fingerprinting is throwing it off?
Some of the data seems straight-up wrong.
I know this computer has more RAM than reported and it's definitely not on a 4G network either.
The APIs it's relying on to show you that information do implement some anti-fingerprinting mechanisms - things like available memory will be put into rough buckets rather than returning the exact amount you have.
Yeah seems like it doesn't account for anti-fingerprinting at all.
If your ISP issues an IPv6 address, isn't that pretty much game over for anonymity/uniqueness?