Solar Bear
Almost nothing Valve has worked on is only for SteamOS, other than packaging and distributing SteamOS itself. They’ve upstreamed kernel patches, RADV patches, KDE patches, etc which affects all desktops. Not to mention the open source tools like Gamescope and Fossilize, the latter of which is used automatically on all Linux PCs playing Steam games, and their contributions and funding to Wine and other projects. Even the new Steam big picture UI, which was initially only available on SteamOS, is now broadly available.
It’s no exaggeration to say that Valve is carrying Linux gaming these past few years. It has been a downright renaissance.
It has one of the same inherent inefficiencies as animal meat: they need to eat something to grow. Therefore we need to use a lot of resources to produce a lot of food for them to eat, before we can eat them.
It always seems much more efficient to use those resources to grow food for us to eat directly. Also, getting Americans to eat a vegetable is slightly easier than getting them to eat a bug.
This doesn’t really bother me because FSR is open source and platform neutral.
I will never tire of people who do not grasp linguistics trying to be pedantic about words.
In a living language, words mean whatever people currently use them to mean. This is how “literally” literally doesn’t mean “literally” anymore. Most people use Nazi to refer to both 1930s German National Socialists and modern day white supremacists. Therefore, that’s what it means. English is not a dead language like Latin, so stop trying to treat it like one with regards to this one word. It just makes you look like a Nazi.
The entire republican party is currently lockstep saying all trans folk are trying to corrupt, if not molest, children. They’re promising national legislation to oppress and punish them if they win. Before them, it was Mexicans and Muslims. They always have a group of “others” to scapegoat, and given enough time and power they always eventually act on it.
“Never again” means we’re supposed to recognize and stop this behavior before it can harm anybody, not wait until it’s too late as we disingenuously quibble over definitions.
LLMs can be really good at answering simple to moderate questions on a subject if you prime it with a bunch of relevant material first, or even better, finetune it on that material if you have the time, power, and access to do so. There’s a really good case to make for using them in search. But general models are pretty spotty at answering anything but the most basic questions on the most common topics.