dlcarrier

By the logic of banana energy potential, you have to go for all of the energy in the entire bomb, not just energy that could be easily released through fission. That's about 4*10^20 Joules vs the 6*10^13 Joules released over Hiroshima.

which, by Cave Johnson logic, gives you about 6.5 million more bombs per bomb!

Fun fact: There's about 200 Hiroshima atomic bombs worth of energy in a banana.

A growth in power generation by 50% isn't even enough to meet the anticipated increase in demand for electric vehicles and appliances. Data center usage is barely a drop in the bucket, in comparison.

Also, there's much, much more going on in data centers than AI, so data center growth isn't necessarily AI growth.

Does anyone make TVs with Opera/Vewd (rythmes with lewd) anymore? That's what mine is running, and it supports Netflix and whatnot, but doesn't show ads or spy.

Also, Android TV boxes sometimes have shenanigans installed by the manufacturer, but they are easy to customize to only run what you want, and they run fine without a Google account.

As far as I can tell, there's no way to use an Apple TV box without linking it to an Apple account and letting them spy on you.

I'm surprised, usually critical embedded equipment runs on OS/2 or NT.

AI stuff is especially buzzwordy. I guess that's what happens when a technology spends half a century being little more than a buzzword, before breaking out and becoming useful.

Usually from power plants, although they sometimes use generators when off-site power has outages. One of the Air Force bases near me recently installed a specifically designed multi-fuel generator: https://www.afmc.af.mil/News/Article-Display/Article/4421866...

It's already resulted in five injuries: https://abc30.com/post/five-high-speed-rail-workers-injured-...

TCAS has solved this for airplanes. It wouldn't take much to modify the system for trains so that, in the least, it notifies when an unexpected train is nearby. A little more logic could allow it to verify which track each train is on.

I will never cease to be amazed by the range of thought experiments that get published as math papers.

Try asking Qwen or Deepseek if anything of note happened in Tiananmen Square in 1989, and you'll see what impact politics already has on models.

Yeah, it's trivially easy to rip the alignment out of the model, but that's only because it's tacked on as an afterthought. If they implemented their censoring as part of training, instead of inferencing, it would be extremely difficult to eliminate.

Alignment isn't only used for censoring, and it doesn't have to be transparent. Considering how heavy of a hand most governments put into companies developing frontier models, it's reasonable to assume they will start restricting capabilities, e.g. there may be a domestic version of Llama and an ITAR-compliant Llama, and China will likely do the same with Qwen, Deepseek, and the like.

I think I need about a terabyte more RAM, to run that.

Open-weight: Anyone with a powerful enough computer can download the model and run it locally.

hybrid instruct-and-reasoning MoE: The model includes a subsystem that can run portions of the model at a time and assign itself tasks, greatly improving the capabilities over running the model in the original fancy-autocomplete method early LLMs used, and letting the model run on simpler hardware.

with multimodal input: The model can parse images, video, and audio, as well as text, so the user can upload files of those types and run a prompt that analyzes them.

Of course polls that show web unexpected lead for an individuals favorite candidate are more likely to be shared by that individual.

That just means they burn a lot of fuel. Generators can burn the same diesel as a boat, and most can burn Jet A, too.

It this point, the best strategy seems to be showing a captcha and only letting through the users that fail on the first try.

Having the users differentiate a scooter from a motorcycle could work too, because every captcha with scooters in it seems to ask the user to select motorcycles.

The 737 is one of the few remaining airliners that you can turn everything off in, and still have full manual primary flight controls.

Thanks for not only maintaining such a great development environment but also for demonstrating that cross-platform development isn't an insurmountable hurdle that requires massive workforce or launching separate Chrome instances for every application, to display text and images.

Considering that the development environment is called Lazarus, you're probably not the only one.

Same here. I don't know if WebGPU works on my computer, but it really isn't needed for a 2D card game.

I've also noticed a lot of early degradation on edge lit Samsung TVs. Their TVs with backlight zones seem to fair much better.

I once used a web page that prevented me from pasting in a password from the password manager. I used xdotool to type it in, out of spite.

There's already an established market for folding phones, so it's likely to have a recurring niche. It's not like the Vision Pro that tries to create an XR-headset market by doing everything it can to not have the functionality that would make it competitive against pre-existing VR headsets.

It's also the same chassis.

TCAS is a great system for stepping in after everything else fails, but I sure hop e no one is using it as their primary means of navigation.

That just makes it computer -> person -> computer -> computer -> person -> computer

We need something where a control center computer that issues clearances can radio them directly into an aircraft's flight management computer. This wouldn't disallow ATC from reviewing the routes before transmitting them, or the flight crew from reviewing them before accepting them, but without people manually copying over the information, there's a lot less potential to insert errors in the first place.

I document re-work procedures well enough that I can turn off my brain while following them, so even that can be a relaxing art project.

There's enough ceramic to absorb enough microwaves. There's other reasons the heat could ruin the microwave, and plenty of reasons not to share food with a microwave that has vapors from melted colored glass. Considering how cheap they are second-hand, I wouldn't use a kitchen microwave for it, especially so if there's colorants in the glass.

I thought CAs automatically filtered out really low keys. There's a great Suckerpinch video (https://youtu.be/M1si1y5lvkk) about trying to purposely make the least secure HTTPS page possible, and Tom initially tried to choose a laughably-low number as the key, but the CA automatically rejected it.

I don't know what you mean by 100% efficiency.

The magnetron burns out from a high standing wave ratio, which is the ratio of how much energy bounces back into it, to how much its outputting. If you aren't heating anything up, the microwaves bounce back and cause damage.

If you have something in the oven that absorbs the microwaves and gets hot, then you're fine. Liquid water does this, including the water in your food, but ceramics do it too. That is why when you cook frozen foods on a ceramic plate, the plate can heat up faster than your food. Once the food defrosts, it'll have liquid water and begin cooking at full speed. Sometimes the food only defrosts from the plate heating up, so you end up with food cooking from the bottom up, instead of the top down.