where anyone thinks it’s ok or normal to recommend suicide to people

Except that’s already happening even without it being normalized, there have always been assholes that are gonna tell people to kill themselves, especially if they’ve never seen the person they’re talking to before. I don’t see how this is any different.

Literally the whole thing would not have happened without the policy.

It also wouldn’t have happened if a fucked up system wasn’t withholding actual, reasonable alternatives that the person was clearly asking for. That’s my point. Let’s fix the actual problems, rather than try to silence the symptoms.

…and did you notice how everyone was outraged by that? That incident was not an issue with assisted suicide being available, that was an issue with fucked up systems withholding existing alternatives and a tone-deaf case worker (who is not a doctor) handling impersonal communications. Maybe it’s also an issue with this kind of thing being able to be decided by a government worker instead of medical and psychological professionals. But definitely nothing about this would have been made better by assisted suicide not being generally available for people who legitimately want it, except the actual problem wouldn’t have been put into the spotlight like this.

I don’t want to create a future where, “I’ve tried everything I can to fix myself and I still feel like shit,” is met with a polite and friendly, “Oh, well have you considered killing yourself?”

Are you for real? This kind of thing is a last resort that nobody is going to just outright suggest unprompted to a suffering person, unless that person asks for it themselves. No matter how “normalized” suicide might become, it’s never gonna be something doctors will want to recommend. That’s just… Why would you even think that’s what’s gonna happen

Seems like clients vary wildly in how they interpret this markup. This is how it shows on Sync:

I was thinking of an approach based on cryptographic signatures. If all images that come from a certain AI model are signed with a digital certificate, you can tamper with metadata all you want, you’re not gonna be able to produce the correct signature to add to an image unless you have access to the certificate’s private key. This technology has been around for ages and is used in every web browser and would be pretty simple to implement.

The only weak point with this approach would be that it relies on the private key not being publicly accessible, which makes this a lot harder or maybe even impossible to implement for open source models that anyone can run on their own hardware. But then again, at least for what we’re talking about here, the goal wouldn’t need to be a system covering every model, just one that makes at least a couple models safe to use for this specific purpose.

I guess the more practical question is whether this would be helpful for any other use case. Because if not, I hardly doubt it’s gonna be implemented. Nobody is gonna want the PR nightmare of building a feature with no other purpose than to help pedophiles generate stuff to get off to “safely”, no matter how well intentioned

Yeah but the point is you can’t easily add it to any picture you want (if it’s implemented well), thus providing a way to prove that the pictures were created using AI and no harm has been done to children in their creation. It would be a valid solution to the “easy to hide actual CSAM between AI generated pictures” problem.

Yeah, but by generating with AI you’re incentivized to skip that initial research stage into your own code base, leading you to completely miss opportunities for consolidation or reuse

Huh, you’re right. But it’s even just within different Lemmy apps, the link doesn’t work in Sync, but it works in Boost.

Your link seems broken, here’s the correct one: https://en.wikipedia.org/wiki/Meteor_(missile)

Never heard of that, thanks for bringing it to my attention!

Tf are you talking about, unless being gay involves raping men, being pedo also doesn’t involve raping children. Even as a cishet non-pedo you will often encounter situations where acting on some attraction you feel would be anywhere from morally questionable to straight up illegal, and most of us manage to deal with that just fine. Of course that’s going to be tougher for someone whose entire experience consists of that, rather than just part of it, but nothing about being pedo forces you to become a child-raping piece of shit.

Of course psychiatric care is important, but the point the other commenter was making is that it’s currently impossible to change anyone’s attraction, so it’s not a pathology that can be “cured” in this way. Any psychiatric care currently has to be aimed at helping people deal with being pedo without acting on it and also not developing any other psychological afflictions because of suppressing their attraction. Trying to “cure” the attraction itself would indeed be akin to gay conversion therapy: there’s no scientific evidence it works, and it’s going to do more harm than good.

Not really self-hosted in the typical sense, but Obsidian with the Tasks and/or Kanban plugin synced through a (self-hosted) solution of your choice could work?

Haven’t tried the whiteboard tool in Google keep (didn’t even know there was one), but the Excalidraw plugin for Obsidian should cover almost any whiteboard use case I can think of. A bit more limited but also good is the native Canvas plugin in Obsidian.

That’s how literally all language change happens? People just start using words differently or use new words, it slowly spreads, until a majority is using it. You can either embrace it and be happy you get new tools to express yourself with, or reenact the “old man yells at clouds” meme and be grumpy. I know which one I’ll choose.

Check out this one: https://thegradient.pub/othello/

In it, researchers built a custom LLM trained to play a board game just by predicting the next move in a series of moves, with no input at all about the game state. They found evidence of an internal representation of the current game state, although the model had never been told what that game state looks like.

No no no, you see, if the employee isn’t there, they could rent out that space instead, but they don’t. By getting the employee back into the office, they’re eliminating those opportunity costs! /s

On a more serious note, saving costs could be a reasonable argument if the company were compensating the employee for their increased cost of living when working from home - electricity, heating, water, internet etc. at home also have to be paid somehow. However, I kind of doubt that a significant number of the companies we’re talking about here actually does that in the first place.

Well, as long as your devices are not TVs and are online from your “household” location at least once every 30 days, it shouldn’t complain as far as I understand it

Windows: “Time for updates! Stop everything you’re doing and please wait…please wait…please wait…please wait…”

How am I hearing about this all the time, but it has never happened for me? Every windows update for me so far has always gone the same, unintrusive way - when it’s time to shut down the PC in the evening, I notice there’s an “Install Updates and Shutdown” option next to the normal “shutdown” option, which I use if I’m not in a terrible hurry right now. Takes a little longer to shut down, next boot will also take a little longer, but that’s it. I’ve literally never had these unwelcome interruptions I hear so frequently about.

I don’t understand your negativity. How else would you write a proposal for a completely new system to be talked about, if not in an idealistic and prescriptive manner? That’s the first step to then start a discussion about it and find and fix the aspects that people expect to not work in practice.

Only for three months, though, because:

The Irish regulator oversees Meta under the General Data Protection Regulation (GDPR) for the whole of Europe because the tech company has its regional headquarters there. Other European countries such as Norway are able to issue national decisions for a time limit of three months in a “case of urgency” under the GDPR.