Absolutely agree. I’m only talking about the fleeing the country part. Those of us who can stay are going to have to put in a lot of work, speaking out against fascism, protecting those who cannot flee, and being generally rebellious against tyranny.

If anything those of us who happen to not be directly in their crosshairs have a greater responsibility to speak out for the groups that are going to be targeted, because it could quickly get to a point where it’s dangerous for those marginalized people to be as vocal. We cannot leave the most vulnerable to fight alone for their right to exist.

Technically it’s not a full 55% of my countrymen, just 55% of the ones who bothered to vote. I’ll admit that’s not really a meaningful distinction though. Unfortunately, there’s also more of us who want to leave than the rest of the world can reasonably handle. I hope as many marginalized people can get out, because it’s going to be bad, especially for them. But those, like me, who are unlikely to be directly targeted due simply to being lucky enough to be born straight, white, men should probably leave those limited seats for those who truly need to leave.

I still use DDG as my “daily driver” (I know there are better options for privacy and avoiding big tech, but I haven’t yet found anything independent that is good enough for me to switch to full time yet). I bookmarked Stract a while back, and it proved useful a few months back when Microsoft had an outage that took down Bing and by extension, Duck Duck Go. I do like Stract, their index seems to be enough larger than MoJeek (another independent search with their own index) that it gives me better results.

Stract might not be as open as I’d like, but it’s nice to have as an option, and I’m never going to complain about having more search providers with independent indexes.

I’ve been using Syncthing-Fork (on F-Droid) for the extra features it has. I wonder if that developer will be able to continue.

While I’m not the person you replied to and don’t know what their argument would be, I’ll take a shot at giving my own answer. In many cases when people post examples of AI giving unhelpful or bad information, there’s often someone who runs off to their favorite LLM to see if it gives a better result, and it usually does, so it gets treated like user error for using the wrong LLM or not wording the prompt properly. When in other examples that person’s favorite LLM which gave the correct answer this time, is the bad example hallucinating or mixing unrelated concepts, and other people are in the comments promoting other LLMs that gave them a good reply this time. None of the LLMs are actually trustworthy consistently enough to be trusted alone, and you won’t really know what answer is trustworthy unless you ask several LLMs and then go research the topic on your own anyway to figure out which answer is the most correct. It’s a valid point that ChatGPT got the answer more right than Gemini this time, but it’s somewhat useless to know that because other times ChatGPT is the one hallucinating wildly, and Gemini has the right answer, but since they’ve all been wrong before who do you trust.

LLMs are like asking an arrogant person who thinks they know everything, who rather than admitting what they don’t know, will pull an answer out of their butt, and while it might be a logical answer, it isn’t based in reality, and may still be wildly wrong. If you already mostly know the answer, maybe asking the arrogant person works, because you already know enough to know if they are speaking from their actual knowledge or making up an answer, but if you don’t already have knowledge on a topic, you won’t know whether the arrogant person is giving useful information or not.

I have had someone tell me that they’d rather live in an economic system “like we have in America” where people have a chance at rags to riches, than a system “like Germany, where the social safety net means the average person doesn’t have a chance at making it big.”

If anyone ever tells you wealthy people are intelligent, don’t believe them.

Those are local outages due to trees causing downed lines, squirrels getting into transformers, motor vehicle accidents, etc. They aren’t the same as the blackouts like you get from an overloaded or strained power grid like they have in Texas.

Just so you’re aware, Match owns OkCupid now, as well as Tinder, Hinge, Plenty of Fish and a bunch of other smaller dating sites. I don’t think they’ve all been ruined as much as Match yet, but it’s probably only a matter of time.

Compiled shaders are unique to every GPU model and often driver revision. The console versions don’t studder because they all have identical hardware, so compiled shaders can be shipped with the game.

Steam will eventually download a shader cache specific to your hardware, otherwise if you jump straight into a new game on PC, the game is going to have to compile them during gameplay, or make you wait 30 minutes to play while they compile (similar to how a lot of emulators for modern consoles like the Switch make you wait). And since nobody wants to launch a newly downloaded game just to sit at a boring 30 minute loading screen, they do their best on the fly.

This isn’t about defending Fromsoft, they’re just another company trying to get your money. I’m just saying that’s how PCs work, and new games with complex shaders are probably pick being accused of having performance issues at launch than hitting players who are expecting to launch a game and play right away with a long loading screen (that a patent prevents them from putting a mini game on while you wait).

From what I saw the negative reviews were split between complaints about difficulty, and performance complaints. On the performance front it looked to me to mostly be shader compilation studders, which is relatively common with most new games.

Difficulty wise, yeah, it’s hard. That’s a big part of the appeal of Fromsoft games. They have made some adjustments since launch to bring the difficulty down a bit, but it’s probably better that they launched a game that is “too hard” and patching the difficulty down, than releasing something that everyone can steamroll through in a day and getting complaints that it was too easy. The game also rewards exploration, and if you just try to rush the bosses without exploring you’ll make things much harder on yourself.

I think it was an interview with Seth Meyers, but somewhere Heidi said she had seen them in costume, but Mikey’s lip prosthetic/makeup was much more extreme in the live performance than in rehearsal, and that was what caught her off guard and made her break.

Yeah, I know a lot of the smaller, independent search engines are lacking, but the people using the “udm=14” trick to remove Google’s AI results now, as if that won’t be removed as soon as Google needs to show investors the AI is more profitable.

To add to this, Scarlett Johansson took on Disney and they settled. And Disney is like the final boss of litigious companies (either them or Nintendo). If she has the same legal team for this, and they think she has a case against OpenAI, this could open the door for OpenAI to get rightfully clobbered for their tech-bro ignoring of copyright laws.

Eh, there are plenty of old quotes that don’t hold up, and the last 1500 years haven’t really been that peaceful. I think it’s fair to be critical of a philosophy that’s been around that long and has really just been better at marketing increased military spending than actually successful at building real peace.

Just to add a bit of clarification, the image wasn’t just a headshot, yes that’s the part that was originally scanned and used, but it’s a cropped in section of the centerfold, a 3-page fold-out image in the magazine. If I remember the story correctly, they needed a large image to scan, and several people brought in images to scan in, and one guy brought a Playboy.

I remember seeing an interview with the model, who at the time of the interview was in her 70s or 80s, she apparently wasn’t enthusiastic about having become a common test image. But since she had technically consented to be in Playboy (which was only a magazine at the time), there wasn’t anything she could do to stop it. I think in this case it’s probably best to stop using her image specifically, as it does kinda get into a weird messy situation of consent, and how her consent to be in a magazine morphed through technology into something more “permanent” than she originally realized. There are plenty of other models who would absolutely be down for that, and given enough time, knowing how nerds are, there will be other test images of women. But I think it’s probably for the best that this one gets retired from this use.

And yes, there are people who have tried to use this instance as a “there shouldn’t be images of attractive/implied nude women a standard test images, because it can cause body image issues for women who go into that field.” Which on one hand, I can see where they’re coming from, but also people take pictures of people, and some people do look better than most of us, having more diverse test images would be a good thing, because we don’t all look like that. But some do, and they’re probably going to get more pictures taken of them than the rest if us.

Not sure exactly how good this would work for your use case of all traffic, but I use autossh and ssh reverse tunneling to forward a few local ports/services from my local machine to my VPS, where I can then proxy those ports in nginx or apache on the VPS. It might take a bit of extra configuration to go this route, but it’s been reliable for years for me. Wireguard is probably the “newer, right way” to do what I’m doing, but personally I find using ssh tunnels a bit simpler to wrap my head around and manage.

Technically wireguard would have a touch less latency, but most of the latency will be due to the round trip distance between you and your VPS and the difference in protocols is comparatively negligible.

I figured you were being genuine, but there’s usually a few people who point at Microsoft’s “embracing” of Linux as the first step in the “embrace, extend, extinguish” trope, and see any involvement by Microsoft as nefarious. When the reality is just that Microsoft’s Azure cloud services are a much larger share of their annual revenue than Windows, and Linux is a major part of their cloud offerings.

If you browse the LKML (Linux Kernel Mailing List) for 5 minutes, you’ll probably see a bunch of microsoft.com email addresses, and it’s been that way for years. I understand why it bothers some people, but also Linus (and a couple others) approve everything that actually gets merged, whether it’s from a microsoft employee, or a redhat employee, or anyone else. Even if microsoft wanted to pay employees to submit patches that would hurt the kernel, the chance that they’d actually be approved is so low it wouldn’t be worth their time.

Maybe I’ll try and give it another go soon to see if things have improved for what I need since I last tried. I do have a couple aging servers that will probably need upgraded soon anyway, and I’m sure my python scripts that I’ve used in the past to help automate server migration will need updated anyway since I last used them.