The capacity of the government of a large jurisdiction like California, or more particularly the US, could bankrupt someone like Musk, so I say, bring it on. Within a decade Musk would have abandoned all efforts, or, even better, be stone cold broke (frankly billionaires shouldn't exist at all, and we should tax the living fuck out of them down to their last $200 million).

We're too afraid of these modern day Bond villains when we should be aiming every financial, and probably every real, cannon straight at them and putting them in a sense of mortal danger every minute of their waking lives, so that they literally piss themselves in terror at the though that "we the people" might decide to wipe them out for good.

You're assuming two things: a) that they exist, and b) that they require maintenance. Modern process design reduces manual switches and dials to be simple electronic sensors read remotely. Modern equipment is insanely reliable, and a steam raising facility is an incredibly simple process to get right from a reliability point of view. And for random faults, well that's what redundancy in design is for.

These aren't new problems by the way. We have been building things and been putting them in worse conditions for many years. Take for example from oil and gas well head control systems. They are located on the seabed in deep water. We're not just talking a sensor or two, no we're talking the entire system, sensors, valves, hydraulic motors, control system, multiple UPSes, packaged neatly and dumped a mile under the ocean surface. It's a lot of fun to design, and even more fun to see someone approach the first time, thing such as mean time to repair suddenly dominate reliability calculations when you need to schedule a submersible to go collect your safety system module at 1.5km depth. Yes I have done that, with a system that is identical to ones used in nuclear reactors (except when we design them in nuclear reactors we get a 1E certificate and slap a zero on the end of the purchase order for that piece of paper).

As for remotely accessing things being shenanigans, we've been doing that for many years already as well. All that "access" you are talking about is access during construction and design, not during operation. Fun fact I worked opposite a completely unmanned air separation facility. Not only remotely operated, but remotely operated from a completely different continent. The plant was in Australia and the control room in Indonesia. At my company we also have completely unmanned facilities in the ocean. (Again, MTTR and equipment reliability which is my bread and butter, gets interesting when we consider maintenance teams not being on site and having to provision a boat or helicopter to go fix something).

This just isn't new. Also no one here is talking about anything "big". In fact that is the biggest problem here. Not the remote operated bit, or the maintenance bit, but rather that these atom-bros are yet once again pushing tiny SMRs which don't exist yet. Forget pipes, crews, or any of that, they are talking about a reactor, steam, and rotating equipment package that is less than 1m across and about 8-10m high generating fuck all electricity. That's the big problem, not the maintenance, but the fact that this kind of tiny stuff remains in fantasy land and on concepts drawn on napkins.

No they do not. Many secondary parts of reactors may require this, but nuclear reactors cores and turbines have little maintenance overall. But even then I don't think that's where your understanding is wrong...

Yes there is. Reactor fleets being largely 40+ years old mean we are still operating them with the technology of the day, the designs of the day and the operational requirements of the day. Much like people run around old oil platforms reading gauges and dials as well.

That's however not at all related to what we are building these days and there's little to no walking around or checking anything. A large portion of modern process design is reducing the need to read anything. Sensors are cheap. Data recording is cheap. Everything is digital. For a project it now costs almost as little to install a wireless pressure gauge than it does a physical one (same for every other process measurement). For a greenfield construction the cost of wiring is borderline irrelevant too so even wired equipment costs little more.

It's a two edged sword, the problem of people no longer needing to do rounds thanks to modern designs that keep people away from equipment has lead to new an interesting products where companies produce gear that takes over what the rounds were previously for. Gas detectors, cameras, heck one company even has acoustic monitoring systems designed to "listen" to the plant since operators don't do that anymore.

The reality is in a modern facility there's very little for an operator to actually go out and read.

It's exactly what it means, it becomes a core codec that all of Windows programs can universally use, managed by the OS, native support. Just because it's it's not there by default doesn't mean it isn't native to the OS.

Or what do you claim it to be? Where do you draw the line? The kernel? Is nothing in the OS userland "native"?

So let's follow this theory, to what end? WebP was developed by Google. What do they get from its adoption? It's an open format, a free format, and has a public specification. That gives Google very little in the way of any benefit from pushing it. It offers them no advantage over any other image format. Likewise if the point was to push something developed by Google it's worth noting that Google was one of the core contributors to JPEG-XL. It's only marginally less Google than WebP is.

I suspect this was as simple and plain as expected. JPEG-XL didn't look like it was gaining any traction. It wasn't used, support was half-baked and experimental in Chrome, so they abandoned it. No surprise there, Google has killed things far more popular.

On the flip side what happened recently forced an issue, PDF's adoption of JPEG-XL. Google maintains an inbuilt PDF reader in Chrome, meaning they now *need* to actually add JPEG-XL support if they want to remain PDF compliant. It's no longer an optional addition of something no one uses, but now becomes a PDF standards issue.

I really wish them luck, but there have been so many false dawn's on fusion...

Actually this will be the easy part. They are talking about tiny modular reactors. It's not exactly complex to drill a 30" hole and it's not exactly complex to hook a gizmo to a winch capable of lowing something 1 mile. We actually have mining elevators that lift and lower workers continuously all day every day that are longer, deeper, and physically larger than this.

Well false, and covered.

Firstly no, nuclear plants do not require daily maintenance. In fact the core / steam loops are largely maintenance free outside of planned shutdowns years in advance. Maintenance is usually only carried out every 24 months.

As to how, it's not exactly rocket surgery. This proposal just lowers two components to the bottom of a hole in a water column, just shut it down, cool it off (like you would do with a normal one), and then all you've got is the extra hour or so it takes to winch the thing up to the surface. It's not in any way buried or sealed down there.

No one is burying anything. They are lowering something under a water pressure column. The same way you get it down you get it up: winch and cable.

Is this oversimplified? Could be, but that is literally in their marketing materials discussing maintenance.

Honestly their bigger problem is cost. Combine the expense of nuclear, with the added expensive of horrendously small economies of scale building small reactors, and add the expense of a geological work and you've made the most expensive form of power generation even more expensive. This project is just as dumb as all SMR projects.

Maybe we're not all plodding along life wallowing in our own misery staring out the window and have business / pleasure that involves the *destination* rather than trip.

Road trips are great for taking road trips, they fucking suck for getting somewhere.

Define better. For all but travelling to opposite extremes of the country, high speed rail is usually faster for passengers and drops you right in the central business district. You can move far more people through high speed rail, far cleaner, with more comfort.

The entire world had flights long before high speed rail came into the picture, yet there's a reason why many major economies have invested in the latter (and why that investment has paid huge dividends).

The vast majority of people don't see Windows as enshittified. That's something more unique to the minority of technical people.

Tech people: "WTF can't I install a local account anymore. ENSHITIFICATION!"
Normal people: "Oh look my documents folder is automatically backed up by the cloud!"

I struggle with your argument. You say that Ubuntu does it better than Windows but describe a situation where some of the features don't work, and then sign off with a statement that it's "good enough for you"? That doesn't sound like it's better than an alternative, it sounds like you're accepting of downsides. My own experience is Ubuntu's horrendous management of audio devices is ... charitably said, as bad as Windows, but especially for the likes of communication headsets, significantly worse.

Feature updates haven't taken 30min for about 5 years now. If they are, consider getting your SSD checked. I don't think a Windows Update (feature or otherwise) has taken a computer offline for more than a few minutes at a time. Back in the Windows 10 days I'd happily have agreed with you, they were really bad back then.

I honestly question what kind of system admin you are, or what your users are doing. I would happily wager real money that 99.9% of users have never ever seen, used, or in most cases even heard of Powershell. And the only situation I have ever come across a requirement for someone to use it, it was entirely self inflicted, i.e. the admin's own scripts were the ones that screwed something up in the first place. Corporate IT for you.

Define working? What didn't work specifically? Honestly I've not come across a Windows install that didn't "work" out of the box. Now you may want to adjust it to suit your tastes, but most people use apps on their computers and don't even bother screwing around with their OS. On the flip side I do remember spending an hour to get Ubuntu to try and use both of the connected displays on my dock at home. I actually gave up, but it turns out Wayland worked just fine out of the box. It's better now that it's not a choice anymore, but when your first user plugs a laptop into a USB C port and you need to explain to them the difference between X11 and Wayland that's not a good experience in the slightest.

FWITW, Ubuntu 24.10 did work on my machine out of the box first go, but then so did any Windows I've ever installed. These days I hope it's a wash, but seriously your complaint about Windows here makes little sense to me. And yes I use a default Microsoft image, every driver automagically downloaded, everything automagically worked. Sure maybe you have some esoteric hardware, but that problem still exists in Linux, I fuck around for an hour trying to get my printer going whenever I have a new Linux install at home and that's plug and play in Windows. I fuck around an hour trying to get my GPIB card working in Windows, that's plug and play in Windows.

The reality is for most people it's an absolute wash. Techies frequently are the source of their own problems. We fuck around with the OS (all of them) and find out.

The hardware being unusable of course made a big difference. But you're well and truly overestimating how a *normal* (that should read: no one here on Slashdot) user uses the OS. For the most part virtually no one gives a shit about the occasional popup in windows, virtually everyone expects accounts required as a fact of modern life for everyone (and outright expects it in a modern OS - thanks Android and Apple), and I'm not sure why you think there's a single person out there who considers bitlocker on by default to be a downside.

Why the sooner the better? The price isn't pegged to natural gas, the market has made natural gas the outcome of the system used to price power. That's two very different things (and natural gas doesn't set the spot price always, but it does the majority of the time).

The UK uses marginal cost pricing, just like the EU, many places in the USA, and several other countries. The system is an absolute boon to green energy as well as grid stability. Having the highest marginal cost set the price means that green energy sources become some of the best ROI projects, which is why there's so much industry investment in the first place. Also having the highest marginal cost set the price means that special purpose systems ensure grid stability (batteries / gas peakers).

Decoupling it from gas presents a problem, what's the alternative? Based purely on supply and demand there's incentive to provide grid stability. The bulk power provider gets the majority of the profit, which would probably be large gas plants. If you regulate the production of the price how do you setup power purchasing contracts with the regulator that are low enough risks that projects will go ahead? (E.g. Australia has an interesting regulatory market, and they found themselves in a strange situation recently where a battery operator was ordered to discharge at a time where it wasn't cost effective for them to do so to provide grid stability in the middle of the day, that sort of thing creates tension with people investing in power generation).

And you break the link and the price goes down, what then? How low can the price go? Pure market dynamics would set the price to the point where gas peakers will no longer offer grid stability services since pure supply and demand often reacts too late in the market (also a problem that Australia was dealing with, and one of the reasons they actually have 3 mini-markets in their main market, baseload, FCAS, and in the latter a sub-market specific for producers who can react in sub 6 second speeds).

This is a lot more complicated than you think and there are many downsides to moving away from a marginal cost pricing system on power.