Re:No cookie consent walls, scrolling isn’t consent, says EU...
by est31 on Wednesday May 06, 2020 @12:36PM
(#23093077)
You might think it's just one bit of fingerprinting information, but if 99 of 100 users click "accept" and one user clicks "decline", it's suddenly become 6.6 bits. Now combine that with other fingerprinting data and you have uniquely identified the user.
It's the same problem with pages where most users are logged in: the few who aren't are suddenly such a small group that they become identifiable through other means.
Re:No cookie consent walls, scrolling isn’t consent, says EU...
by jcheng on Wednesday May 06, 2020 @05:26PM
(#23096505)
Can you explain how you got 6.6 bits?
Re:No cookie consent walls, scrolling isn’t consent, says EU...
by user2342234 on Wednesday May 06, 2020 @06:11PM
(#23096956)
From Shannon's formula for information content[1]. The Idea is that the less likely an event is the more information it imparts.
I = -log2(p)
In this case they used p = 1/100
Re:No cookie consent walls, scrolling isn’t consent, says EU...
by aDfbrtVt on Wednesday May 06, 2020 @05:57PM
(#23096815)
Because it uniquely describes a user within a group of 100 other people. 2^(-6.6) ~ 1/100
Re:No cookie consent walls, scrolling isn’t consent, says EU...
by marksomnian on Wednesday May 06, 2020 @06:16PM
(#23097005)
log2(1/100)
How is a opt out cookie the same as tracking cookie?
Isn't there a clear difference for a cookie:
Compared to a cookie: Sure the first one can tell server "this is a user that had visited the page in the past". But this is nothing compared to the second one which tells server "this is this specific user"