Yeah, condensed matter is full of fraud stories. And it's almost always interesting, if disheartening, to read about them.

It's a shame showing a lack of successful measurement isn't rewarded or even encouraged. I end up hearing stuff through the grapevine how the idea I thought worth revisiting was already tried by some small team back in the '90s, and it's only mentioned at the back of the supplemental materials. There's a binder (and database) on my desk (laptop) that catalogues excerpts and mentions of such misses that may end up being my biggest contribution to the field.

So many grad students will run the labs for like 80 hours a week, gather the data sets they were told to, and then have no idea what any of it means.

To be fair, grad students span gamut from 'wait, why isn't B a constant?' out-of-their-depth beginners to the likes of you, who probably shake their head at visiting professors' inexperience with methodology. Not really trying to defend how some people run labs, but I know in my heart there were times when prof wasted his breath on explaining my role in the grand scheme of things.

If you're also paywalled.

Funny enough, I thought it's going to be about something a lot more #kleinbl00batshittery-y.

https://ossuaryrecords.bandcamp.com/album/popio-y-wiar

So, did the organizers try to scam some art subsidy? 'Cause, going blind, this gave me major The Producers vibe. Poor everyone, though.

For sure, the word of mouth is as crucial as earning a good reputation. That said, the course is gonna have only 15 seats by design. I'd rather be having 4-6 people who want to come rather than deal with a hundred who just want a pass.

Meetup reminder: March the 7th, 06:00 PM ETS, so about week from now.

After getting positive poll responses, I'm drafting an elective course. One of those 'proof writing for not just math majors' you have in the US. So much for doing less paperwork, but this one at least doesn't feel purposeless.

I wasn't in a hurry, it was nice to have front row seats for such a prescient demonstration.

When every foodhole in Warsaw connected with delivery service overnight, outgoing orders had much much higher priority. So, during pandemic, you had a crowd of deliverers, normal line that moved at snail's pace, and a nearby crowd of people who placed their orders in an app to game the system. This lead to a situation where people from the last group placed order to <restaurant's address> and added comments like "I'm the one wearing a brown hat with a gigantic pompom" or "I'm already behind you."

Insert something about follies of idiots with access technology. I don't know, I barely slept since Friday.

Correct me if I'm wrong

Dunno, probably not, but I think you could instantiate one that can when they can and freeze its learned ability, so the whole hoping it doesn't forget might go away.

But I have no idea. Don't write that much code or work with raw data these days, so bibliographic aid is just about all it can do for me in an hour of need. Otherwise, it's about as tangential to my goings-on as it can get.

When I tried that 'explain paper' site, it left enough of a distaste for me to roll eyes and move past. Between absolutely fucking insisting that some unrelated mathematical concept[0] is absolutely crucial to explain my question and rephrasing a circular argument until I got bored and left, I probably won't bother again for quite a while.

Unfortunately, the above experience mean I'm unlikely to trust LLMs with stuff I don't know a lot about. Also, I kinda regret writing anything in this thread and will probably just add more tags to my ignored list. Fun company notwithstanding - too much hassle, too few fucks left.

[0] - I wrote and deleted 900 word footnote of jargon about orbits of the coadjoint representation groups and operators in de Sitter space, so let's pretend I said Tits index and wiggled my eyebrows in an amusing way.

Philosophy deserves every burn. Sorry. But only a little.

Eh, I'm being my usual exaggerated dismissive, but it's sad that the two most visible to me camps are essentially "it's only so unbiasedly rational of us to consider how many AGI could dance on the needle's head" and "mathless/IFLS quantum vibes" types. It's not even that I don't see the merits of those two, let alone philosophy at large, but that I have absolutely no fucking interest in either yet they keep talking at me like I'm a lobotomite for not caring.

And no, wasoxygen, I'm not calling you out specifically, it's just how you Yudkowsky-ites communicate. We're cool, I hope.

Not sure why you'd want a middleman, either.

Well, ML/whatever excels at finding patterns, even if it can't/won't explain them. Having a tool that goes "exploring these parameter spaces is most likely worthless" or even "isn't it funny how second order solitons only form when this parameter is divisible by 17?" may be invaluable to a right person who can find context to those observations. That's the "(or something)" in my previous comment.

Tying this to "making sure conclusions are correct-ish is going to be hard to replicate." <- that's the bottleneck as far as I can see. First you have to separate seeds from chaff, and then make sure those seeds aren't blighty or cleverly disguised angry bears. I wouldn't mind science becoming (even more) akin to computer-assisted chess, though. Tools are tools, experts use tools better, so that checks out too.

Losing shits and meetups

Wasn't singling you out here, though I hope you take care of yourself and wife. And it's not like I don't understand or lack the presence of mind to understand why people are so agitated. I simply can't keep dealing with it. It's been two goddamned years, and I can't even force myself to go to Ukraine anymore. I haven't seen the worst, and it's too much. Focusing on what I can affect has to be enough for me right now.

As to meetups: no worries, I can make another one in April or May. They're about as informal as flip-flops anyway.

Meetup? Meetup. The 7th of March, 06:00 PM ETS, so two weeks from now-ish.

- But you have to fill in those forms yourself! By the day after tomorrow!

- Oh, not a problem. They follow the same schema as the ones I had completed last year, and only need my signature at the end, once the dates are updated.

- They still need to be filled. By hand.

- Say no more, you've been hereby deputized!

And thus I took about a half of papers back to my desk, since that's not my ass on the line if they're late. As someone who managed to out-pedant local civil servants for so long they gave up, I feel the thrill of battle once more.

I'm not arguing those problems won't go away, or that it's any more or less than a tool. You can give me that much I hope.

And you're right that I wouldn't pay a human for those, at least unless those would be recurring NPCs or something like that. I do commission background sets regularly because 1) the free/cheap/generated ones are usually on par with what I can make, 2) what I can make suffers a severe pizazz deficiency. Lotsa bang for a buck, too.

We'll tell our grandchildren "we used to make our own handsome faces".

Cyrodiil's Jesus! No, I tried generating something a touch less 4chan-does-Amnesia and more Balkan Romani without the perpetually disappointed look.

Since I'm self-righteous, I'd like to think one of the last things it'll come for is physics and math.

Theory is much less about hand-waving connections between deeply understood parts and more about doing the math with as little preconceived ideas as possible. Don't imagine what atom/potential/sun is, calculate and interpret what comes out, see if anyone tested something similar / calculated it in a similar regime. Propose an experiment, try to make a feedback loop with someone (or something) that'd bounce ideas back. It's everything else that ought to be automated, 'cause the amount of paperwork they try (underline: try) to pile on me is just fucking ludicrous.

The problem is that models aren't better at determining they're wrong than humans, and are unlikely to learn it since their very nature is numerical bias. And, frankly, LLM/models/AI/whatever should have less of a problem replacing philosophy, because doing proper math requires pencils, paper and a wastepaper basket for wrong ideas... whereas philosophers seem to only ever need the first two.

Otherwise, I kinda stopped paying attention to anything that isn't directly related to my interests tbh. Seems like everyone is losing their shit over anything and everything in the news/work/word holes, while I'm tackling the deeper mysteries of is it better to keep seeing someone with a 3-year-old and see where it leads or cut it loose before things get difficult for the kid moreso than us.

Anyway, I hope you are well. :)

Same to you. We gotta do some meetup. I wanted to organize one in January, but my health took a dip, maybe it's time to try again.

It's visual and obvious, dude.

I'd go with 'trivial' or 'left as an exercise for the reader'.

You're right, though. Generators seem to be less able to remove 'turbulence' from the output, but rather move it someplace else within it and hope for the best. Like, I tried to make some character art for my game, and it can pull off some handsome faces, for sure more detailed than I'd have patience to draw, but the clavicle-to-armpit areas look inexplicably like Munch's melted cheese period.

Maybe they launch from a milk silo into another one? Scary times to live in Wisconsin.

An avian flu panzootic — a pandemic among animals — has struck some 320 bird and mammal species, including elephant seals.

(Illustration by Emily Sabens/The Washington Post; Luis Robayo/AFP/Getty; iStock)

It felt like watching one of those blockbusters about the end of the world. Like witnessing an apocalypse, but in real life. Which, in a way, it was.

The beaches of Valdes Peninsula in Argentina, normally so packed with elephant seals that time of year it is impossible to stroll along the shore, were desolate except for hundreds of dark, rotting carcasses — nearly a whole season of seal pups dead, with gulls pecking at the remains.

Instead of the usual cacophony of guttural honks that drowns out the waves during breeding season, the eerie silence was only broken by the sound of a few remaining elephant seals shaking their heads, snot running down their protruding, namesake noses.

“You felt like a bomb had exploded,” said Martín Méndez, recalling the scene he witnessed in October during an annual survey of southern elephant seals in that stretch of coastal Patagonia.

“It is catastrophic,” added the marine biologist at the Wildlife Conservation Society, a New York-based nonprofit group. “This is the largest die-off for the species, period.”

Altogether, an estimated 17,000 elephant seal pups seals died there last year from avian influenza, victims of an unprecedented panzootic — a pandemic among animals — that has struck around 320 types of birds and mammals.

Over the past few years, a potent strain of H5N1 avian influenza has jumped between species and raged through domestic and wild animal populations on every continent except Australia and Antarctica, crisscrossing the world along birds’ migratory routes.

So far, cases of humans getting seriously sick from this strain of flu are rare. But the possibility of the quickly evolving influenza virus gaining the ability to be transmitted between from one mammal to another — and eventually, to humans — has scientists concerned about the pathogen turning into another pandemic. “Every year that this doesn’t happen,” Méndez said, “we’re being lucky.”

For poultry farmers, the outbreak has already come at a significant economic cost, striking tens of millions of birds in the United States. For wildlife, it threatens to disrupt ecosystems and push endangered animals closer to the brink of extinction.

“We’ve never seen such a massive spread of virus in wild birds, and we’ve never seen such massive infections of wild mammals,” said Ron Fouchier, a virologist at the Erasmus University Medical Center in the Netherlands.

The trajectory of the virus, unprecedented both in its global spread and in the number of species infected, shows how dependent and connected humanity is to the natural world — and how farming practices that facilitate the flu can disrupt it.

‘I was not prepared’

Scientists first spotted the precursor to this flu strain in commercial geese in southern China in 1996. Over the next several years, it spread to poultry farms in the region, popping up occasionally in wild birds within flight range of infected farms.

Then, in 2005, some 1,500 wild geese and gulls dropped dead in a protected nature reserve in northwestern China, far from any poultry coops. The virus had found its way into the wild bird population.

From there, the virus rapidly spread across Asia, Africa and Europe, popping up in seasonal outbreaks. But it was only after a genetic change that allowed the pathogen to spread in migrating birds around 2021 that the numbers of outbreaks exploded and the virus went global, establishing itself in North America that year and in South America the following.

Since then, the death toll has been staggering.

About 5,200 common cranes in Israel. More than 2,200 Dalmatian pelicans in Greece, about 40 percent of the species in southeastern Europe, and roughly 20,000 Sandwich terns in Europe, 17 percent of the northwest European breeding population. More than 18,000 dead barnacle geese in Scotland. And tens of thousands of gannets in Canada.

Last year, ornithologists found about 12,000 dead black-legged kittiwakes in Norway. By July, more than 500,000 birds died in South America, including about 41 percent of all Peruvian pelicans.

Flu symptoms depend on the bird, according to Jonas Waldenström, a disease ecology professor at Linnaeus University in Sweden. Telltale signs include tilting heads, struggling to stand and tumbling when trying to take flight — all signs the virus has hit the nervous system.

“In some species, it’s pretty ugly,” he said.

The virus is literally reaching the ends of the Earth, killing brown skuas on islands near Antarctica and, for the first time just this winter, a polar bear in Alaska. Scientists worry it is only a matter of time before it reaches penguins and other vulnerable populations on Antarctica itself.

“The Antarctic situation is at a precipice,” said Michelle Wille, a virus ecologist at the University of Melbourne. “Many of the animal species that live there are found no where else in the world, and many are already facing substantial pressures due to things like fisheries and climate change.”

Scientists say the disease can jump between species when a wild animal eats a dead or dying bird, or when bird poop plops into a farm animal’s feed. So far, bird flu has infected coyotes, lions, tigers, grizzlies, raccoons, red foxes and other terrestrial mammals, but only sporadically. And it has also struck farms for mink and fox fur in Spain and Finland.

But a series of mass die-offs of marine mammals that congregate on beaches has biologists worried about the possibility of the virus evolving to spread directly from one mammal to another.

Last year, more than 5,000 sea lions were found dead in Peru. And more than 17,000 elephant seal pups succumbed to the virus in Argentina, representing at least 96 percent of the juvenile population. Nursing seal pups likely would not have caught the flu by eating birds. They instead may have gotten it from contaminated milk, water, aerosol or feces, Fouchier said.

Valeria Falabella, a Wildlife Conservation Society marine biologist who has studied Argentina’s elephant seals for decades, had only ever seen a dead adult once before this flu outbreak. Seeing beaches of dozens dead was “absolutely devastating.”

“I was not prepared to see that,” Falabella said.

The risk to people

If a bird flu virus evolves to spread between seals, it is more likely to spread within other mammal populations, such as humans. “It’s quite a distance from a gannet to human,” Waldenström said. “But from a seal to a human, we’re pretty much alike.”

Tracing how viruses hop between species is hard. In the past, humans have caught disease from animals after working on farms (the 1918 flu pandemic likely started in Kansas farm country) or after encroaching into wilderness (HIV likely jumped from chimpanzees when a hunter killed one for meat).

In the case of the latest flu strain, it is “very difficult to prove mammal-to-mammal transmission,” Fouchier said, adding that a genetic analysis of the virus found in the dead seals could provide clues about transmission.

So far, what we know about this strain of avian flu suggests an outbreak among humans is not imminent. The adaptation that the virus made to replicate in migratory birds — and thus spread across the globe — seems to have inhibited its ability to infect humans, Fouchier said.

“It is slightly less concerning to human health at present,” he said, compared to past flu viruses. But given the unprecedented nature of the ongoing outbreak it’s hard to predict how it will evolve. “A typical bird flu virus never makes it into mammals at this scale,” he added.

The longer the virus persists in the environment, the more opportunities it has to fuel a new pandemic. Influenza viruses evolve much faster than the coronavirus behind the covid pandemic and other pathogens.

Right now, there is little to do to stop the spread of this flu virus among wildlife. The U.S. Fish and Wildlife Service, for instance, is working with zoos to vaccinate critically endangered California condors.

“It’s hard to contain a virus that’s now on more or less all the continents,” Waldenström said. “There’s no putting the lid on that. It will run its course.”

But there are steps the poultry industry can take — monitoring for viruses at farms, disinfecting equipment, having workers wear protective gear — to reduce the risk of future outbreaks, according to Jonathan Sleeman, a science adviser for wildlife health at the U.S. Geological Survey. “One recognizes that this is a lot of effort,” he said.

For the surviving elephant seal population in Argentina, it may take years to come back. “They may never recover to the same level of population as they were before,” Sleeman said. Biologists are working to understand how many adult seals died.

Méndez, the biologist who likened the seal die-off to a disaster movie, worries how the recent outbreak may undo decades of conservation work.

Now, the virus has transformed their work. It’s about more than just aiding faraway animals — it’s about protecting people, too.

“We’re really trying to protect the very functioning of our planet,” Méndez said. “We’re looking at this disruption through the lens of wildlife, but this is obviously very serious for humans as well.

This article is part of Animalia, a column exploring the strange and fascinating world of animals and the ways in which we appreciate, imperil and depend on them.

Hah. Neeeerd.

It's the gist, yes. My problem is that if the idea is to scale the process to something that's more than a 100ml jar on a bench, this experiment is almost a spherical cow in a vacuum. For starters, you have to take steps to keep it under 80C, so it'll either need to be current-limited (takes more time) or be actively cooled (takes more energy, likely skews carbon balance). As it is, they probably just put a moist paper towel on the vessel and call it a day. Your product and reagents are strongly alkaline, so not all materials will stand the reaction conditions for whatever amounts of time. Those aren't insurmountable problems, sure, but stack enough of them and this might not be as viable as it looks on paper.

That said, if there's a will there's a way. I'd like it to be me making mountains out of anthills.

Oh, I didn't mean to underplay Fifteen Hours. My approach was cautious, but I thoroughly enjoyed the read. Kinda makes me wish I had time and folks to play Only War with, but that's gotta wait. I'd be also interested in your takes on Horus Heresy books and 40k in general. Got any other IG recommendations?

I dropped reading GoT around the middle of book 1, in the chapter where old-ass lecher betroths his whichever daughter to one of the Starks where we get name-drops of no fewer than thirty characters and I knew this is just author trying to waste my time by putting in hooks that'll never pay off. It annoyed me enough to ignore the show by association, but maybe it'd be fair to give audiobook a go.

That friend interested in enlistment decided to postpone it, but volunteered for the month-long civil defence training. Then, the recruiter (different, more seasoned) looked at me like a wolf on sheep, which was funny. He didn't stop even after seeing my service category evaluated at D (A is best, E means you're either blind or on a wheelchair), so I gave him the equivalent of a girl just wanting to be friends by showing dual citizenship and institute ID. There's protection from conscription, and then there's having recruiter give up on trying. There's gotta be an achievement for that.

I have a lot of work to do, though. Found my stride at teaching and made repeated lessons less tedious by forcing myself not to re-use examples or analogies (but still gave all in downloadable slides), but overall don't think this is a good fit for me. There are also three papers I contribute/coordinate, and it's been a pain. I don't mind juggling multiple different topics if it's just me, but when others are involved shit becomes unmanageable. Then there's some bullshit commitee... no me gusta, as ancient memers said.

Huxley is interesting, with a lot of ideas thrown from the get go, and it's more likely my current workload and family bullshit at the distance preventing me from enjoying the story. I rarely don't give a book multiple chances, so we'll see. Right now, it's flatter than my color perception, but the language isn't a problem.

re 40k: I usually recommend people start with Sandy MItchell's Ciaphas Cain books, who's pretty much Blackadder given a role of a commissar (political/morale officer) and a gun to execute morale problems. Comedy of errors/accidental hero is a good framing for grimdark.

Adding my own two cents:

Fifteen Hours by Mitchel Scanlon - a story of a guardsman on his first deployment. It's 40k, so you know it's a tale of merrymaking and sunshine. I was afraid people oversold this book, but it turned out to be pretty dang good. TL;DR: a Vietnam war movie with an arc from FNGs to body bags, but with laser guns and space-orcs.

Chivalry and Courtesy: Medieval Manners for a Modern World by Danièle Cybulskie - it's the same author who wrote How to Live Like a Monk -- mentioned in the last thread -- who this time takes on manners, grooming, and upbringing of children. It's... hm. I didn't find it to be as good, but probably because I didn't learn all that much new stuff from it? Wholeheartedly recommend it, the substance is there, it was just a tad too introductory to someone who read some of the sources beforehand.

A Short History of Nearly Everything by Bill Bryson. You know what? It's the same problem as with Chivalry and Courtesy: I'd love this book 10-ish years ago, but now it felt like marathoning 12 straight hours of something like SciShow on youtube (neither a channel recommendation nor lifestyle idea). There's a lot of trivia on history of science that I'll probably struggle to recall in three months, and a more than a couple things that are no longer correct, but I don't regret picking it up or recommending if that's your jam or didn't do well in the science classes.

De Generibus Dicendi (On the types/kinds of speeches) by 'Iacobus Gorski' (Jakub Górski, or Jacob Mountain-like if you're into silly translations) - It's one of the first Loeb-inspired (Latin text on the left, Polish on the right) books published here, and a real treat. It's an overview of ancient and contemporary (XVI century) rhetoric, and a bit of a instructor's handbook on the topic.

I've also tried to go through Brave New World by Aldous Huxley, but it's been a fucking slog even in audiobook form.

Hey, glad to see you back!

What made you take the plunge on Expanse? Did you watch the show?

2. Blockchain is totally just a database with interesting security characteristics

I mean, database is essentially a list with fields, so it's not exactly a unique structure or data representation/storage.

Always suspected that everything beyond plier-wrench is not only man-made, but unstable.

Thank you for reaching out, I might get back to you on it. But you're from the USA, though, right? I'm not sure how much overlap is there with Polish forces.

Feels kinda gauche to comment under my own thread, but fuck it. I'm tired as hell and didn't get much sleep in three days, so let's roll.

Despite being shit with recognising faces, I caught an 'outsider' cheating on a mid-term for one of my students. It gotta suck to (probably? hopefully?) get expelled from a school whose motto may as well be "Cs get degrees[0]," but it ain't exactly my problem.

My ex-interns came back to work on that paper. They also had me invited/requested on a student project making a radiotelescope array with stuff like old sat-TV antennae. Not gonna lie; it's weird to wrangle scrap that'll ultimately be synchronised with a signal from our new atomic clock, but what the hell. It's super, and it's neat to construct something that's largely outside my scope. Like, I get the underlying principles, but my day-job puts me dangerously close to failed-phil dumbasses who ''''''prove'''''' how multiverses must exist because meta-predicate logics are constructable, so it's necessary to recalibrate and touch grass now and again. The whole thing reminds me of hackathons before recruiters and dickheads displaced real people. Also, labs/experiments are about a billion times more enjoyable when you know you won't have burn a weekend to write a 30+-page report afterwards.

I've been asked to come along to a military recruiter as an all-purpose witness/assurance/nit-picker, and (for me) it was fun in an unintentional way. Almost feel bad for the guy 'cause between minute-long waffles to 'answer' yes-no questions and getting thrown off his game amusingly easy, I may have had my haircut longer than he's been in service. Still, my mission was accomplished. We also got some of the least informative literature since my last encounter with the Jehovah's Witnesses, but that was to be expected.

Despite this experience giving my friend a pause, he remains interested, so I'm gonna do what we probably should have started with, i.e. contact soldiers I met through paintball. I know, there are many bullshitters there, but those get off on LARPing as special forces, not say they "mostly do admin" or "fight the entropy in the motor pool."

[0] - I think Ut mediocris fuerunt, ita vicerint tracks.

While looking for one of the sources to Ghost Wars, I had happened upon a book at the university library which, roughly 100 pages in and leafing through the rest, think you may enjoy: Poisoned Peace by Gregor Dallas.