Quoting Sabine Hossenfelder as anything just outs you as anti-science. She has pivoted years ago to making videos that stir the pot of controversy, aiming at those who don't like "academics". If you want any proof of that claim, you can check out her video saying that Physicists are afraid of Eric Weinstein. Eric Weinstein has a bulshit "geometric unity" theories thoroughly debunked, where it's not even that he can't get published: he doesn't *want* to get published, claims in his copyright notice for that paper that it's for "entertainment only" and requires his express permission to write any papers that build upon what he has written: so you technically can't publish a paper debunking without violating his copyright, if anyone actually *cared* to, because physicists that have looked at it like Sean Carroll pointed out his paper isn't rigorous enough to say anything testable.

She allies herself with quacks to get views, and you're falling for her bs.

So...there's really a shed that people build to store bicycles?

Again, is this a European thing? I've never heard of a shed dedicated to storing bikes....I can see this is brewing used as an analogy for something else, but the shed/bike concept itself is foreign to me.

Most people rarely ever change a time zone....a good number never leave the one they are born into....

So much this..

Yes except, we are not talking about keeping ourselves warm in the winter, cool in the summer, the lights on in the ER etc... We are talking about power for AI data centers; not even data centers for "traditional" computer applications and software.

I'd be the last person here to argue we don't to develop energy infrastructure. I would agree that could maybe even should include nuclear fission at least in the form of modern much safer designs with real consideration for who bears the cost for the 'common defense' when a commercial entity is permitted to paint a giant target on a community.

Never the less this is a societal cross roads, where it comes to AI. The choice isn't massive unrestricted AI build out or mud hutts. We options spanning a range of carrying on like it 2016...damn the torpedoes restart EOL reactors and generating plants to feed the things.

Nuclear power plants (particularly 3rd gen and older) represent an enormous socialization of risk. Their really isn't any getting around that. Historically a rising tide has also lifted *most* ships. Even if it is not immediately clear and their are short term employment impacts etc, AI driven productivity probably will make most American's better off and a small subset with a lot if direct financial interested in things like restarting aging nuclear reactors more fabulously wealthy than they already are. So the question is does this "Promote the General Welfare" in net or not in terms of should regulators really go along with it.

Like many stories...the answer is:

"Well...it seemed like a good idea at the time...."

What the fuck is a bike shed?

Never heard of one or seen one before...is this a European thing?

What is left to change besides oven and microwave times?

One...the most IMPORTANT one....my internal clock....

That really takes close to a week to get fixed every time.....it throws all of us humans off to some degree.

Oh...and my dog....I have to adjust her schedule twice a year too....

It probably makes more sense for ST....but as far as I'm concerned....

JUST PICK ONE AND STICK WITH IT....

I want to quit the damned changing in my lifetime.

The problem with Nuclear power is that it looks really good in the abstract sense but where it encounters reality it does not suit society so well.

Someone on here a few days ago pointed out that with something like 400 power stations and large industrial reactors there have only been three serious containment incidents.

I would say that isnt actually that good a track record given the cost of such calamity. Which really is unbound as far we know because it remains unclear what dealing with Chernobyl may still require as there is still fissile activity producing enough heat it could destabilze the surroundings, and all kinds of weird fungus that might be attacking the structure of the buildings, and Fukushima while it looks like we have handle on the ongoing clean up requirements, there can still be surprises. If scale the numbers of 2017-19 Chevbolt fires to 30 years you still have a rate of about 2 tenths of percent where spontaneous ignition would occur. GM still did a a recall and that only burns down your garage...

The other problem is the size and acute nature of catastrophe when it does happen. Yes you can prove with statistics coal plants release more radiation and kill more people one way or another. That's fine but they tend do it over decades at a rate society can absorb those costs. Even when a tail pond lets go or something we are usually looking at the loss of one town not a 1000+ sq miles. That is difference between one these 100 year floods and 500 year floods right there, in terms of social cost.

The picture for modern reactors does not change as much as Bill gates wants us to think because humans. Sure maybe it can't meltdown, maybe it can't produce hydrogen and create explosive conditions that will scatter materials on their own. But that does not mean some nut job shouting "Death to America" won't crash a jetliner into it, or execute some other kind of attack potentially spreading contaminates over 100s of miles. Remember you don't need a blast its the fire and the smoke that pump stuff high into the atmosphere where it can rain down later.

Exactly. Base the time zones on GMT and adjust your daily schedule accordingly.

The bigger you are, the harder it is to turn the ship in an emergency. See Titanic.

At some point the inertia of change isn't worth it, as you suggested. It is possible, but is it worth it?

The last way to look at this is a way to start the process of diversifying our Tech holdings. Windows, Office, Adobe, VMWare etc. But here is my take, STEAM is showing the way out. We're not there yet, but the tide appears to be turning (to keep the nautical terms going).

I don't disagree with any of that. I see a lot of LLM and agents being thrown at problems we used to have much more tailored tools for that did at least as good a job, while consuming a tiny and by some comparisons almost immeasurably small amount of compute resources.

You're not wrong, Word 6 on a 486 with less than handful of megabytes of memory was able to perform grammar checking on word doc. Now you need 10s of gigs of primary storage, perhaps as much as 50 gigs of secondary storage between the model and the office suite. How much better a job it actually does is disputable. The experience is different, and a no-knowledge user can get started immediately. There is value there in both aspects. I don't see people who get paid access thru and employer etc to claude, gpt, etc stop using it or ignore it.

You're asking a deeper question though, how does that no-knowledge user become someone who understands the system when the system is getting bigger all the time and the short term incentives to learn anything about it get smaller smaller. I don't know. There is a lot of scifi on the subject, dated from before 2020, so thinking without being colored by recent economic interests, probably worth reading.

Broken clocks are still broken, even when they occasionally tell the time correctly.

I love when the jackals start fighting each other..