4 ms·

> maybe the simple answer is that refusing cookies doesn't actually do anything? I agree with all the other comments that are saying this is probably fingerpri

by jefftk 3y ago

> maybe the simple answer is that refusing cookies doesn't actually do anything? I agree with all the other comments that are saying this is probably fingerprinting, but you can check whether refusing cookies is doing anything. Two ways: 1. On a page where you refused all cookies but are seeing targeted ads, open developer tools and go into the "Application" tab. Open up Storage > Cookies. Do you see anything listed? You should see nothing there. You also shouldn't see anything in the rest of storage, since "cookie" consent is really "client-local storage consent". 2. You can check whether cookies were sent on particular network requests, like the ones to the ad companies. Open a new tab, open devtools, open networking. Then paste the URL in the url bar. Find an ad request in the networking tab: do you see a "Cookies:" header? If so, it sent a cookie for you.

dndn1 3y ago
The truth about cookie behaviour can be found, as it's a browser-driven feature (unlike fingerprinting). But at least some analysis of the cookies will be needed, as I think essential cookies under some definition/interpretation are permitted? Now what exactly that means I'm not sure. But the first thing my cookie banner does when someone declines the banner is.... create a cookie. But I guess we can argue that's essential for the user experience - to keep the banner out of the way on follow-up page views. Not a lawyer or privacy expert though. Edit: above is behaviour I found with at least 2 cookie banners. I'm interested in suggestions if this isn't standard!
- jefftk 3y ago
  Good point; you shouldn't see exactly zero cookies. But all cookies should be "strictly necessary" for fulfilling an explicit user request.
2rsf 3y ago
Yes, I tried that with mostly inconclusive results, maybe I chose the wrong site, except for one or two sites that were either really impressively good or bad

by jefftk 3y ago

dndn1 3y ago

The truth about cookie behaviour can be found, as it's a browser-driven feature (unlike fingerprinting). But at least some analysis of the cookies will be needed, as I think essential cookies under some definition/interpretation are permitted? Now what exactly that means I'm not sure. But the first thing my cookie banner does when someone declines the banner is.... create a cookie. But I guess we can argue that's essential for the user experience - to keep the banner out of the way on follow-up page views. Not a lawyer or privacy expert though. Edit: above is behaviour I found with at least 2 cookie banners. I'm interested in suggestions if this isn't standard!

jefftk 3y ago
Good point; you shouldn't see exactly zero cookies. But all cookies should be "strictly necessary" for fulfilling an explicit user request.

2rsf 3y ago

Yes, I tried that with mostly inconclusive results, maybe I chose the wrong site, except for one or two sites that were either really impressively good or bad