As the article says, it's a dark pattern, and there's no actual reason for any random visitor to press 'Accept all' other than the fact that it's simpler.
I'm still waiting for the EU or whatever other court to put the hammer down on that fact and start handing out fines to people who follow this dark pattern, or just straight up ban the data collection that's the whole reason people want you to press 'Accept all' in the first place. Some courts have already ruled that you need a 'Reject all' button,[1] precisely because not giving consent needs to be just as easy as giving it. It's obvious to anyone that having 'Customise' be the alternative to 'Accept all' is naught more than an attempt to comply with the letter of the law while wholly rejecting its spirit.
Now, maybe we can expand the rejection of that dark pattern to other software. The article has some good examples like 'Get started' versus 'Remind me later', but I'd love it extended to software updates, so you can accept security updates (somewhat analogous to required cookies), and reject "feature" updates (often analogous to tracking/marketing/screw you cookies), but I doubt there's an easy way to regulate that. Even free software (both as in beer and as in speech) have done the same kinds of crimes, if only because they're following the UX language of the times, or because someone is really excited about their new feature and just really thinks you should try it.
> But when that option is hidden, I've resorted to opening my browser's developer tools and deleting the banner element from the page altogether. It’s a desperate workaround for a system that refuses to offer a legitimate "no."
... Okay, but then does the site actually drop cookies?
It shouldn't, as it shouldn't save your preferences without you actually having pressed the button to save it, but I wouldn't put it past some web-dev to save that cookie as a default 'everything enabled' on first load, and only edit it to actually reflect the user's preferences once they've gone through the motions.
Doing anything other than not dumping those cookies is a dick move in this sort of case, since the user hasn't provided clear and obvious consent.
As the article says, it's a dark pattern, and there's no actual reason for any random visitor to press 'Accept all' other than the fact that it's simpler.
I'm still waiting for the EU or whatever other court to put the hammer down on that fact and start handing out fines to people who follow this dark pattern, or just straight up ban the data collection that's the whole reason people want you to press 'Accept all' in the first place. Some courts have already ruled that you need a 'Reject all' button,[1] precisely because not giving consent needs to be just as easy as giving it. It's obvious to anyone that having 'Customise' be the alternative to 'Accept all' is naught more than an attempt to comply with the letter of the law while wholly rejecting its spirit.
Now, maybe we can expand the rejection of that dark pattern to other software. The article has some good examples like 'Get started' versus 'Remind me later', but I'd love it extended to software updates, so you can accept security updates (somewhat analogous to required cookies), and reject "feature" updates (often analogous to tracking/marketing/screw you cookies), but I doubt there's an easy way to regulate that. Even free software (both as in beer and as in speech) have done the same kinds of crimes, if only because they're following the UX language of the times, or because someone is really excited about their new feature and just really thinks you should try it.
[1] https://www.techspot.com/news/108043-german-court-takes-stan...
> But when that option is hidden, I've resorted to opening my browser's developer tools and deleting the banner element from the page altogether. It’s a desperate workaround for a system that refuses to offer a legitimate "no."
... Okay, but then does the site actually drop cookies?
It shouldn't, as it shouldn't save your preferences without you actually having pressed the button to save it, but I wouldn't put it past some web-dev to save that cookie as a default 'everything enabled' on first load, and only edit it to actually reflect the user's preferences once they've gone through the motions.
Doing anything other than not dumping those cookies is a dick move in this sort of case, since the user hasn't provided clear and obvious consent.