To reference my own signature, which is meant entirely sarcastically: why do you assume I haven't researched it? Why do you assume my opinion on it was uneducated? Perhaps you formed your opinion on insufficient information?

But as for your specific example, that's completely unlike anything I do. Database migrations are the sort of mechanical, simply defined task that AI can handle. My work involves things that are much more open ended, and often involve inventing new methods or even new algorithms to do things no one else has ever done before.

I accept there could be fields where that's useful. Not for the sort of work I do. In my field, doing even basic work requires specialized knowledge, high level problem solving, and strong design skills. When you take on a junior developer, you have no expectation it will make you more productive right away. Initially it will probably make you less productive. You're doing it to mentor them, so they can grow into a senior developer.

So I can have my own permanent junior developer who isn't learning from working with me and will remain a junior developer no matter how long I mentor them? That doesn't sound useful.

All those things decrease the need for energy. EVs use much less energy than ICEs. Heat pumps use much less energy than gas furnaces. Electric stoves use much less energy than gas stoves.

But that's unrelated either to the story or to my post. We were talking about the energy used by data centers to run AI models. My point is that the amount of energy used is unrelated to the needs of any specific real world application. It's driven by companies competing for business. No amount of computing power is enough. Each of them needs to have more than their competitors, no matter how much that is.

And yet the demand for more power has never gone away. It keeps increasing. The more computers can do, the more people ask of them. However powerful they get, people will always want more so they can do even more.

Increasing efficiency won't solve this problem. Right now, the economic imperative is to build bigger companies running bigger models on bigger data centers. Bigger than what? Bigger than their competitors, of course. Everything else is secondary, including whether they destroy the planet in the process.

We need to declare that not destroying the planet is the new economic imperative. Everything else is secondary to that. But that's not happening, because the super-rich are the ones who get to make the decision. They would have to put humanity's needs above their own. You don't become super-rich by thinking like that.

If you want to play VR games but avoid Facebook, the main options are PC VR (Steam has a huge library of games) or PSVR2 (which can also be used with a PC with an optional adapter). Those are headsets that you hook up to a PC or console, not a self contained system like Quest. They can provide much better graphics than Quest, but the games are mostly the same.

Comfort is an issue with most VR headsets. They're big and heavy enough, you don't want to wear them for hours at a time. The main exception is Bigscreen Beyond, which has gotten it down to a remarkably small and light package. You pay extra for that, of course. I think it's about $1000 for the basic version, a little more for the version with eye tracking.

Yes, there's no question of that. Humanity is seriously screwed at this point. But there's still a lot of wiggle room for just how bad it's going to get. All the futures are bad, but some are a lot worse than others. The choices we make today will determine which one we get.

I mean, it's a tough choice. The German car industry or the future of humanity? They're both really important. Which should we put first? I'm sure there are good arguments on both sides...

China's per-capita emissions are 3.5x higher than India's. I really do not want India to emulate China on this!

I would say they need to be more ambitious. They plan to hit peak emissions in 2045??? The world needs to be aiming for net zero by then, not peak.

$21 trillion is a lot of money, but that's what they plan to spend over decades, not all at once. To both develop and decarbonize the world's most populous country, it's not really that much. And it will pay for itself and then some.

I do want to be fair to them, though. India and China have similar populations, but China's emissions are 3.5 times higher. India is not the main problem right now. If their emissions keep growing, though, they could become the main problem.

The name "GNU/Linux" is a shameless attempt by FSF to take credit for other people's work. In any modern Linux-based OS, almost all the major components come from sources other than GNU. That includes the kernel, the desktop environment, the graphics stack it's built on, the drivers to support your hardware, the package manager, the bundled applications (web browser, file browser, terminal emulator, etc.), and much more. None of these come from GNU. They think their relatively small contribution means they should get top billing on every OS that uses it.

The entire GNU project is only 387 packages, most of which are either obsolete or incredibly obscure. For comparison, Debian includes over 63,000 packages.

If you're looking for a cheaper laptop, the MacBook Air starts at $999. The MacBook Pro line is the more expensive models for higher end uses. The least expensive model in the more expensive line is not a low end laptop.

They probably believe climate change is real, and it's important to do something about it. The same way you do. It's just that other things are even more important, like their need to stay ahead of their competitors and your need to chat with your virtual girlfriend/boyfriend. They also probably believe it when they claim AI will magically find a solution to climate change, because believing that makes their lives easier and the human capacity for self-deception is incredible.

It's comforting to blame someone else, but morally speaking, the CEO who builds massive data centers is no worse than the ordinary person who eats lots of meat or drives an SUV because they like it better than a small car. They both choose to put their own desires above the welfare of humanity. The CEO's choices have a lot more impact, but there's lots of blame to go around. We just saw a few days ago that ChatGPT has 800 million weekly users. If those people weren't using it, they wouldn't be building data centers.

In this case, very little. They're using waste rock from quarries. From the article:

But you can't scale this very far using only waste from existing projects. To get this to the point that could make a difference to the climate, you would have to start quarrying and transporting rock on a massive scale just to crush it and spread it on fields. The tradeoffs become more complicated at that point.

It depends what your reference is. If the comparison is to having electricity in cities, but it's powered by fossil fuels instead of solar, then the cities aren't hotter.

They are. It isn't just Norway where EV sales are exploding. Globally they're up 25% in the last year. And before you try to claim that's starting from a tiny number, it isn't. 22% of all cars sold in 2024 were electric. 2025 is on track to be a lot higher.