That’s a bit outdated by now, Störerhaftung doesn’t apply anymore for other people’s action on your WiFi

The y axis doesn’t start at 0, making it look like the change has been a lot more drastic than it actually was (even though it’s still very bad). I think that’s what they’re referring to

• Fluxer
  • Slightly sus vibes

Can you elaborate on this part?

A feature that will not do anything unless you explicitly press a button to start using it is quite literally opt-in, though? Opt-in doesn’t mean “I won’t even know the feature exists without hunting through the settings”. It just means that it won’t start doing things without your consent. Presenting a way to provide that consent in a more visible place than buried deeply in the settings does not make it opt-out. It might be a bit annoying to you, but it has no effect on your user choice or privacy, especially if there’s also a way to globally hide it and any other features like it, including new ones that might be added in the future.

Ah, okay, gotcha. Yeah that’s fair. Not something I’ve ever really used, so wasn’t aware of that. Your comment read to me as if Windows as a whole just didn’t support drag&drop.

Or maybe you’re overestimating the amount of people who actually used that. Spending effort on something that less than maybe 1% of users actually use and that is not load bearing to any important workflows is hard to argue for when you’re a corp that is only concerned about its own bottom line. It’s a pretty rational business decision, even if you (and I) disagree with it.

Sadly not true. Microsoft removed the Start button in a version of Windows before

They didn’t say that every version of windows since then had a start button

First of all they only talked about the start menu, which was still part of 8, even if it was annoying and full-screen. And second they only said that every Windows version that had that allowed you to move the taskbar around. Not that every Windows version so far had it.

Uh, what? Can you clarify what you mean by “drag&drop”? Because dragging and dropping files or text around within or between application windows definitely worked even when Win 11 was new, so you’re probably talking about some specific instance, I assume?

Didn’t want to be too combative from the start lol

Like it or not, words have meaning, and black and white thinking is comfortable, but doesn’t help anyone. There is still a massive difference between the gates foundation having a 0.5% stake in Kurzgesagt and them being “owned” by PE, and pretending otherwise just means you’re not actually interested in any kind of productive discussion.

Bringing this general issue up in a thread about Kurzgesagt, without also providing evidence that Kurzgesagt specifically is actually part of that issue, is at best irrelevant and at worst misleading though

Do you have actual evidence for Kurzgesagt being among PE-owned channels, or are you just extrapolating? Because the video you linked doesn’t mention them, and a quick search didn’t turn up anything about that.

I think the idea is to pressure the partners of Collective Shout, per the url in the comment. Those might not necessarily agree with what they’re doing in this case, and if they see it’s making waves, reconsider their partnership.

It doesn’t need to be the latest android version per se, but I wouldn’t want to use a phone that’s not getting security patches anymore

Well, yeah, kind of at this point. LLMs can be interpreted as natural language computers

Watch time is pretty important on YouTube afaik, initial clocks themselves don’t count for that much

What? Since when does Valve prohibit companies from redirecting customers to non-Valve purchasing flows? Because that’s what this ruling is about, it says Apple can’t prohibit apps from telling users to go buy off-platform for lower prices. Valve isn’t doing that with Steam afaik, actually I’m not aware of any other platform that does this

“The planning thing in poems blew me away,” says Batson. “Instead of at the very last minute trying to make the rhyme make sense, it knows where it’s going.”

How is this surprising, like, at all? LLMs predict only a single token at a time for their output, but to get the best results, of course it makes absolute sense to internally think ahead, come up with the full sentence you’re gonna say, and then just output the next token necessary to continue that sentence. It’s going to re-do that process for every single token which wastes a lot of energy, but for the quality of the results this is the best approach you can take, and that’s something I felt was kinda obvious these models must be doing on one level or another.

I’d be interested to see if there are massive potentials for efficiency improvements by making the model able to access and reuse the “thinking” they have already done for previous tokens