How do I turn it off, or delete Google from my machine and my life?

Reasonable point. To clarify terms, what Livermore calls "ignition" measures the input energy as the laser energy delivered onto the target, not the power used to generate the laser pulse, which is what would count in operating a power plant. What makes ignition significant is that it marks the threshold where the energy generates starts increasing faster than the energy delivered, so there's hope that if we could make a laser that could deliver higher energy it could produce a lot more power. However, it took over a decade to build and start running the National Ignition Facility and another decade to improve its performance to reach ignition. The NIF design is over a quarter of a century old, and the laser technology has improved a lot, but we need to scale up the laser design and build a new generation of laser, and that's going to take time to get up and running. We're still much better off than we were in the late 2010s, when most observers thought NIF would never reach ignition.

It's not just China. DoE's Lawrence Livermore National Laboratory reported the first ignition of a laser fusion target in December 2022 and have scored six more ignitions since then. https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Flasers.llnl.gov%2Fnews%2Fb... It's the first real proof of principle for inertial confinement fusion, and it uses powerful laser pulses to achieve ignition. Inertial (laser) confinement fusion is the major competition for magnetic confinement, and what Livermore has done with the National Ignition Facility is to greatly improve the uniformity of target compression. It's taken a long time coming, and fusion reactors are not just around the corner. It may have been the final blow for First Light Fusion.

Also challenging the traditional "wisdom" of the US auto industry of pumping up the size and the price with bells and whistles, getting buyers to trade the flashy cars for even flashier ones in 2-3 years, and milking the repair market for everything it's worth afterwards. At first glance it makes a lot of sense for people with tech skills and sense.

Nobody warned you about them because they didn't exist until recently. It's a scam that grew out of colleges and universities insisting that professors publish papers to demonstrate their academic credentials, but never looking at the papers, just counting them. That, in turn, arose from "publish or perish" and legions of academic bureaucrats. I write about science and I get invitations to conferences around the world that are another academic scam; many of them don't exist, but the conference "papers" also serve to demonstrate academic credentials for the bureaucracies.

EBay works for me. I've never been on Facebook, but I've been on eBay for years and it works for me. EBay sellers are more professional and great for things like spare parts, cables, adapters, reading glasses, and components that you can't find if you don't have a hardware store handy. Try finding USB-C adapters, for example. It's also great for finding old -- or selling -- books and collectables, and odd things you'll never find. It's my go-to place for useful stuff at a cheap or at least reasonable price unless you're looking for real rarities, and it's a good place to sell things you don't need. Check it out; you'll find much of the same small stuff you see on Amazon,

The problem now is that marketing says "you gotta have new features" but the new features are crap, break something we need, or both.

Right now, Amazon has 95 ratings with 4.8 out of 5 stars 95 ratings Good Reads have 4.6 stars for 73 ratings. The more Zuck tries to shut her down, the more people will buy the book.

I'm with you. I have given up on Google because their AI answers are garbage, and DuckDuckGo is simpler and more reliable. Most answers I get from AI are wrong, and they don't understand my questions. If they don't offer a way to turn the AI off, I'm going to start looking elsewhere.

In the mid-60s my high school in Nutley, NJ sent a group of us to ITT Federal Laboratories where they had an IBM 1401(?) and somebody on their staff taught us something about Fortran that I have since forgotten. One of my classmates ended up as a computer professor. The school must have had some kind of computer to schedule classes and send out report cards.

Agreed. I hit bad ice on route 1 in the Boston 'burbs and ended up sliding sideways for a few hundred feet. You survive that by holding the wheel still and letting the car slow until it the wheels finally find a grip. That's one of those things you never forget, and after that I've always gone slow on anything that looks like ice.

If they're doing the Upper Peninsula of Michigan and upstate New York (think Buffalo and Rochester, famed for their winters), they are doing it to test the cars in some of the worst winters in the US. Bad weather is one of the big question marks for autonomous vehicles. Looks to me like these are tests for the engineers, not tests for the publicity value.

Or maybe people liable to lose a bundle if it works -- like nvidia or Microsoft -- staging a denial of service attack to save their asses?

Once they Microsoft figures that out, they will start charging customers to delete the AI.

Go for it. And the more you sell that you sell them that you generate with AI, the more it's going to screw up the AI they generate from it. It will be hallucinating all over the place.