kurthr

If you can't work in order of magnitude calculations and estimations you're not going to last long in physics. Many useful models give good answers only within a certain range of inputs, while at the same time you often can tell where in a dynamic space you are by external observations. For example, estimating Martian wind speed within a factor of 3x from the size of the sand dunes, can be done having only seen pictures and with only the knowledge of a few dimensionless parameters.

That's also how you can tell if your calculator is "lying" to you (or you typed wrong). I guess I have a few similar tools in spaces where I'm more experienced to see when "hallucinations" and gibberish are being generated by LLMs. Of course having it "check sources" and evaluate its own solutions sometimes also works, if those are reliable, but you're on the hairy edge at that point.

But what if those unstudied accessible "natural" remedies have even more problems? Should we choose to do nothing, only study nothing, or just redefine what "natural" is?

Wait till you find out how much uranium there is in coal ash and how many tons a year are put in the air or dumped into ground water. Both the ash and uranium tailings are in the 50ppm range, but we make 100Mt per year of one of them and basically no uranium tailings in the US. Globally, the ratio is over 1Gt of coal ash and 10-20Mt of uranium tailings.

One is currently a problem, the other isn't.

Not the OP, but due to neighboring rotor interactions an even number (or a single helicopter like lift rotor and a tail rotor ignoring airfoil lift possibilities) rotor makes design easier and improves performance significantly. Think of the rotors like gears meshed together and realize that an even number spins easily, while an odd number is locked.

Of course there are 6, 8, and larger numbers of rotors used in actual drones. The advantage of more rotors is that redundancy to failure can be built in, and that rotor tip speed for a given lift can be somewhat reduced at the cost of efficiency.

The screwworm is spread by the DOGE fly.

Yes, obviously no LLM would make this mistake.

The collapse of global trade would greatly reduce economic efficiency, output, and investment. It has been coming for while, though greatly accelerated by the orange pdf file. It takes a lot longer to build systems of trust and belief in enforcements of global order than to disrupt them. I suppose we'll move closer to the fear side of the financial/political axis from the greed side.

I should mention that the list of thefts (other than an umbrella which I promptly replaced with its neighbor) are not ones that I have personally experienced, nor do I suspect that it is statistically accurate to those reported to police (conbini shoplifting and transit fare skipping must be larger). However, it is accurate to the top 3 "thefts" I've heard Tokyo residents complain about. If a native cares to correct me, I'd retract it.

Japanese police are very rarely willing to even ask to look at any of the disorganized hodgepodge of private cameras for property crimes or even minor physical altercations. They are far more likely to rely on personal accounts. TV dramas not withstanding.

Although Tokyo does have a system of traffic cameras which log traffic movement and license plates, that's most all that it does. Except in cases of murder or kidnapping (or political influence), it's quite rare to request the recordings of many private cameras. Outside of big cities, it's even more rare.

The largest connected system of cameras I'm aware of are for the subway camera systems (Shinjuku, Shinagawa, etc). Although independent systems, together they can do facial recognition to track individuals. Not a lot of AI yet, though.

In Tokyo, it is not uncommon to see bikes parked on residential streets with keys left overnight in their wheel locks (as if there aren't even mischievous 12 year olds?!). Oh, and outside of the cities, crime is even more rare. It is common in youth hostels for there to be open cubbies where personal items are stored in the front near the door. Nothing is taken. Most common thefts are: umbrellas (considered a fungible public good?), unlocked bikes (in high traffic business areas), women's underwear (off of outdoor drying racks).

All theories are wrong.

Some are useful.

Having theories that only give answers, but you can't reason about is not as useful. Having a theory where you don't know the limits of it's applicability, can be very dangerous.

At least in the physical realm there is not yet anything that combines relativity with QM so they can only be approximations. Even in math so far there seem to be similar challenges using programatic and "AI" driven solutions and proofs.

Still, I know that LLMs will be useful for Verilog/VHDL and particularly with verification, where they are already heavily used. Defined outputs and complete test coverage is already such a big part digital/asic design, I'd be surprised if it isn't used a lot more. Many software people would say that hardware is badly written copy-pasta, as it is. That said, higher velocity slop and hardware "technical debt" isn't something you can fix with an update. And no matter how fast you "ship", you won't get parts back in less than a few months. Poorly used, it will lead to expensive failures.

Some 10-15% of San Francisco housing is unoccupied. The exact why and how to fix it are arguable, but I'm doubtful the investment is actually helping.

https://www.pacificresearch.org/time-to-ask-why-so-many-san-...

So no need to do any deep consideration or evaluation, just the approval of the Epstein class.

QQQ is the largest of the Nasdaq100 tracking funds. It's only about 1%, increasing to 4% of the QQQ, which is ~$350B in size. So it's only $3.5B of forced buying or a little less that 5% (of $75B). For the second float would be and additional ~$14B, again about 5%.

I am still suspicious that this has something to do with the relationship between Springer-Verlag and the Max Plank Digital Library (MPDL) which supports open access.

In 2014 MPDL purchase 110k out-of-print and historically significant titles. In 2015 Springer acquired open-access journals from Max Plank Society. In 2022 There was an open-access book deal allowing Plank Institute members to more easily publish books.

Things were more not always so intertwined and in 2007 the Society canceled a licensing agreement with Springer due to subscription prices and usage restrictions.

I sort of agree with this in the abstract. The problem is that concretely, these LLMs are being used to decide whether you receive healthcare, government benefits, or whether your job/agency gets cut. So they have already had real world consequences due to DOGE, Insurance companies, and other uses. I certainly don't find either the methodology or graphical conclusions of this link very valuable.

One can argue what fraction of the several million children's (under 5) deaths are (going to be) due to cuts by DOGE to USAID (and later congressional appropriations) from Grok recommendations/justifications, what fraction were politically pre-determined, and which are "just deserts", but it would be hard to put it at zero.

https://ph.ucla.edu/news-events/news/research-finds-more-14-...

But they are going to coincide lockups with the release of additional stock float from 5% up to 20% of the total "valuation" with a 3x QQQ multiplier so that stock indexes will treat them as 60% float even though 2/3rds of those shares are unavailable. Thus they guarantee that even more shares must be bought by tracking ETFs and institutional buyers. Everybody (that already owns pre-IPO shares) wins!

Also perhaps relevant, Vinge's Marooned in Realtime, bobbles (time bubbles) take the remains of humanity with varying levels of technology and culture 50 million years into the future long after a singularity "extinction event" in the 2200s occurs.

Of course the story is just a murder mystery.

Yep, 'panacea' was grandmaster, but 'quire' was intermediate?

Meh, you see all sorts. It's definitely much more touristed in the last decade. The heyday may have been late 90s and early 00s. See Fruits magazines and books. Certainly, "normal" Japanese avoid it, but they always have, just like Asakusa, Kabukicho, Roppongi etc.

https://archive.org/details/fresh-fruits/mode/2up

That said, you still see Medatsu (目立つ) and lots of younger people there looking for fashion, because that's where many of the (overpriced) used clothing stores are. There used to be more weird bands and such doing pop-up shows or playing at the Yoyogi band shell. Still, lots of Japanese tourists as well as foreigners, and lots of food events/festivals around there.

Conclusions and Relevance There was no significant difference in sustained return of spontaneous circulation between sodium bicarbonate and placebo in adults with in-hospital cardiac arrest. These findings do not support routine administration of sodium bicarbonate for patients with in-hospital cardiac arrest.

The morning "bullet" trains (503/507/511) from San Jose Diridon take 1hr to go 48miles with 10 stops. I think electrification and widening to 3 tracks improved times and reduces the likelihood of delays. Certainly, they run more often now, about every 10 minutes at rush hour and every 30min off hours and weekends.

https://www.caltrain.com/?active_tab=route_explorer_tab

I for one welcome -the ̶̶̶̶̶screw ̶̶̶̶worm- our new AI overlords.

Yeah, but someone at one of the LLM providers would bet against you and do it, just to take your money. If someone bets $100k your house doesn't burn down with pictures posted within 30min of it happening, it probably will.

Because it's Apple. They are huge, have scary lawyers, write scary contracts, and want to "delight the user" with features only when they announce them. They hate leaks, and demand separate teams for basically any/all development.

I wanted this to be thought provoking, but for a physicist to open with, "the acceptable risk of catastrophic meltdown for a nuclear power plant is roughly one in a million", just seems sad. It is a phrase without meaning.

   Is it per plant? there aren't a million.
   Is it per year? notably have been at least 2 major ones (arguably 7).

A meltdown (or loss of containment) just isn't that bad, if it doesn't affect ground water or lead to atmospheric fallout. We're turning 3 Mile Island back on to power an AI datacenter! The AI super-intelligence apocalypse envisioned (ignoring the likelihood) is inherently global.

That said, the rest of the analysis and proposal also greatly disappoints me. The idea that the current administrations of US and China could do anything constructive seems hilarious. They're so paranoid and self-serving they couldn't come together, even if there was an alien invasion. Then the idea that LLM safety is somehow as easily traceable as nuclear isotopes and bomb tests seems equally ludicrous. I am sad.

Not only have I read it, I know people mentioned in it. There aren't very many.

Wow, it sure didn't take long to show a complete lack of familiarity in the field. It seems like that's going to be a real weakness with LLMs based on volumes of material that are later discovered to be semi-fraudulent and unmotivated by scientific principals.

https://stanforddaily.com/2023/12/31/blockbuster-alzheimers-...

I think that the amount of software produced has gone up by much more than a million fold since 1950. The first thing I would consider software and not flipping hardware switches was Assembly Language described in a book by Kathleen and Andrew Booth in 1947 and the phrase Assembly was first used to describe it in 1951. Even that is normally called firmware today.

That doesn't really mean to me that we have another million or billion-fold increase in utility coming, unless we're just counting lines of code. I don't need 1000 more apps on my phone or computer. Are we saying that every webpage and video is software, because they're already encoded in js and H264? or is it just the browser? the OS? How much more do we really need or even want?

Is it just that soon we'll all need to have "agents" working for us to protect us from the other "agents' that are trying to steal what we have? Is that the utility? Who's going to pay more for it?

It seems like we're already at the point where we pay about as much as people are willing to for software. It's already gone to the subscription model. You're already the product for advertising. It's already being enshitified to eke out some extra margin. It's already ~10% of GDP. It just can't grow much more without adding some new utility rather than just replacing what's already there. It needs to create something other than unemployment.

But if it doesn't create anything new, it will be the Tractor to Jevon's Paradox.

If you have ETFs or index tracking funds in your 401k you're going to buy it. In fact, you'll buy more of it than is going on sale. Since they only need to issue 5% of the stock for it to show up as a 20% market cap in the QQQ index and tracking funds own over 20% of the market, they pretty much have to buy all the shares. Welcome to market manipulation 101.

Maybe we're just misinterpreting the meaning of "AI Safety"?

Maybe they mean the AI needs to be safe from us? Can't have the grubby meat flappers touching the delicate bits!