I can’t speak to Germany’s system, but there’s no need for a site to tell the verification service its identity. If it just asks “is the current session authenticated to someone over 16” and gets an answer back. Identity of both parties remains secret.
With a 70% non-compliance rate, that isn’t entirely surprising.
Platforms are even less likely to implement real reforms that the author alludes to.
The platforms aren’t complying with the law:
Of the parents who reported their child had an account on each platform prior to 10 December 2025, around 7 in 10 reported that their child still had an account on Facebook (63.6%), Instagram (69.1%), Snapchat (69.4%), and TikTok (69.3%). Around 3 in 10 reported that their child no longer had an account. One in two of these parents (48.5%) reported that their child still had an account on YouTube following the age restrictions coming into effect.
And here we are, fifteen years later.
“well known” for this, or something else?
Isn’t that Android?
Lemmy does not like this opinion
precaratized? Now that’s a word you don’t hear every day.
It’s so grey.
Are they really using the same terminology as Putin?
Satellite services are pretty amenable to hiding sensitive parts of the world: https://en.wikipedia.org/wiki/List_of_satellite_map_images_with_missing_or_unclear_data?wprov=sfla1
It seems totally on-brand for the US government to request that bits of the war zone be hidden, and it’s entirely on brand for satellite companies to hide them.
I really think removing downvotes or requiring the downvoter to explain the downvote would make communities better.
Lemmynsfw solved this by disabling downvotes. I don’t think memes were a problem there.
An alternative is strong moderation to keep off-topic comments and posts out.
Phew. I’m glad humans did better than bots.
However, according to Opper. ai, only 11/53 cloud-based Al passed the test (~20%). Worrying, about the same error rate as humans
lololol
Citing Program for International Student Assessment data taken from 15-year-olds across the world and other standardized tests, Horvath noted not only dipping test scores, but also a stark correlation in scores and time spent on computers in school, such that more screen time was related to worse scores. He blamed students having unfettered access to technology that atrophied rather than bolstered learning capabilities. The introduction of the iPhone in 2007 also didn’t help.
“This is not a debate about rejecting technology,” Horvath wrote. “It is a question of aligning educational tools with how human learning actually works. Evidence indicates that indiscriminate digital expansion has weakened learning environments rather than strengthened them.”
…
Classroom technology usage has ballooned in recent years. A 2021 EdWeek Research Center poll of 846 teachers found 55% said they are spending one to four hours per day with educational tech. Another quarter reported using the digital tools five hours per day.
Pretty much this. We have strong beliefs. Strong enough to leave a popular, functional platform for ideological reasons.
If the verification service is structured like oauth, then the request could be passed through the browser as signed plaintext. You could verify that the requesting site is only passing a minimum age request to the service. That would be as straightforward as viewing the interaction in your browser’s debug tooling.
If you say that you don’t trust the signature, and that it could be used to smuggle identifying information across, there’s a couple of ways to deal with that: open source and audited provider governed by legislation; information theory that would show personally identifying information wouldn’t fit into a field of that size; and “personal auditing” where you can try throwing data at the service to see if you can trick it into accepting invalid input (that really goes with the previous point, because the only field you can usefully vary is the signature).