I was in a different camp back then. Our CRT TV was high quality and produced a very sharp image, especially with the 3D consoles of the '90s hooked up to it through SCART. Similarly, the first CRT monitor I ever owned was an excellent Sony Trinitron with a flat image, no blur, no perceivable scanlines (I used it for a decade, because I was unable to find flat screen displays that came close). That’s why I felt absolutely no love for those scanline filters and didn’t get their appeal until many years later, when I realized that the art of most '80s and '90s games was intended for highly flawed CRTs. By that point, those simple filters had evolved into complex shaders that are much more accurate too.

A couple of years ago, I configured PS1 emulator DuckStation into what a fictional (and entirely impossible) “PS1 Pro”. Extremely high rendering resolution in the 6K range to remove any hint of jagged edges, with a scanline shader and some carefully tuned bloom on top to simulate the phosphor glow. I kept textures unfiltered, but enabled settings that fix the console’s unstable geometry and texture distortion. I then got a modified version of Gran Turismo 2 with enhanced draw distance (and some bug fixes). The effect is remarkable: The original art is preserved, but enhanced, there’s remarkable clarity, yet the scanlines and bloom still create the illusion of a high res CRT. It looks amazing.

Exception: Old pixel art that was meant to be smoothed out by blurry CRT monitors and TVs. Yes, I know those are not algorithms, but still.

This could end up working better than foldable devices. No creases to worry about.

It’s a so-called “soft” paywall that doesn’t always trigger. It’s common for these to engage after a certain number of articles have been read by the same user. Another method is for a paywall to only engage after an article has gained traction, but depending on how users got there (people coming from Google search results are often exempt).

Teslas are among the safest cars on the road by all metrics. It’s just that they get the most press out of all EVs, because they are 1) sort of a poster child for electric vehicles due to how influential the Model S was and 2) due to that idiot at the helm of the company receiving constant attention from the press.

As such, it raises concerns that AI systems deployed in a real-world situation, say in a driverless car, could malfunction when presented with dynamic environments or tasks.

This is currently happening with driverless cars that use machine learning - so this goes beyond LLMs and is a general machine learning issue. Last time I checked, Waymo cars needed human intervention every six miles. These cars often times block each other, are confused by the simplest of obstacles, can’t reliably detect pedestrians, etc.

If you think the type of person who signs up for a Facebook product will flock over to the “real” fediverse the moment they are seeing ads (which they are most likely seeing everywhere else on the Internet, since this kind of low-information user is usually not even aware of the possibility of blocking ads), then I got a bridge to sell to you.

Also wars, future pandemics, any kind of global cooperation that depends on the White House not being a madhouse, which is a lot.

In the real world, communism also suffered from the mandated growth problem, as well as a long list of other issues that some people still like to pretend solely exist under capitalism and some serious problems that are exclusive to this system. Yes, it is actually bad, with and without Cold War propaganda making it sound both worse and better than it actually was. It failed everywhere for a reason.

This doesn’t mean that there aren’t real issues with capitalism as well. So far, the best system we’ve come up with as a species is heavily regulated capitalism with strong social safety nets. Not perfect, but nothing is on this rock.

This nonsense reminds me of how many phones in the olden days had a dedicated Internet/WAP button on them from the phone companies that primarily existed so that people would be charged for accidentally pressing it.

With this new iteration of the same idea (they could have easily chosen a spot where it would never get hit accidentally, but didn’t), I suspect that Microsoft banks on people accidentally pressing the button in the hopes that at least some will be converted to using their dubious “AI” assistant more than once. Like the author of the article, I have my doubts this will happen. On laptop keyboards in particular, it’ll be pressed when people are actually trying to hit the left arrow key and cause more annoyance and confusion than anything else. I can already imagine IT departments disabling these on all new devices just to save them the extra headache.

I’m not usually one to defend Facebook, but it might have something to do with (and is possibly being done to counteract) this massive state-sponsored disinformation campaign:

https://www.nytimes.com/2023/11/03/technology/israel-hamas-information-war.html

In case of paywall:

https://web.archive.org/web/20231201152318/https://www.nytimes.com/2023/11/03/technology/israel-hamas-information-war.html

That seems like a much better way of doing things.

only 754,633 office workers

Did you actually think about this for a second? Do you really think only 0.2% of the American population or 0.4% of American workers are working in the office? This should be a lesson on not to pick the first Google result.

Wine is not a practical solution for your average user.

I’ve seen this done. Store lasted for a bout a year, which is longer than I would have expected given the obsolete e-waste they were selling for extortionate prices. This was only a few years ago, but most of the laptops they were offering still had 4:3 displays and disc drives, that’s how ancient they were. Hell, one of them had a floppy drive.

Sure, but I’ve experienced hiccups that would never occur in Windows with things as mundane as hooking up a printer, which is well within the realm of what a normal person is using their computer for.

Also, you can fault Apple for many things, but a lack of polish and a poor user experience aren’t among them. I’ve used Apple devices five times in the last ten years and each time I was, with no prior knowledge nor the need to look anything up, able to help people with their issues and quickly. Linux is the polar opposite of that.

I’m a moderately technical person and every single time I’ve tried Linux in the past 20+ years it went like this: Huh, this isn’t so bad, I might use it more of- oh wait, never mind, a cryptic error message just appeared, because I had the audacity to plug some device in or download some generic application so I had to use the terminal again for some incredibly mundane thing and it only worked after I tried three different approaches from forum posts so old I needed to use the Wayback Machine to be able to read the guides they linked to. Those guides naturally omitted vital details that I only noticed, because I’ve been trying to use Linux for over 20 years and actually read a book or two on this mess. It doesn’t matter which distro, which device, which use case, it’s always like this.

The very best “Linux for the masses” I’ve used so far (outside of Android) is SteamOS on the Steam Deck, but even it falls apart the moment you venture outside of the user-friendly walled garden that is the Steam application.

Consider keeping school the one place in a child’s life where they aren’t bombarded with AI-generated content.

In Hongkong. We all know what happened to that place.

I noticed that this isn’t just an issue with this particular tool. I’ve been experimenting with GPT4All (alternative that runs locally on your machine - results are worse (still impressive), but there is complete privacy) and the models available for it are doing the exact same thing.