"I'm Dorothy Gale from Kansas"
Jules of Nature
TVSTRANGERTHINGS
No title available
styofa doing anything
shark vs the universe
Acquired Stardust
blake kathryn
🪼
ojovivo
One Nice Bug Per Day
ellievsbear
Claire Keane
if i look back, i am lost
Stranger Things
Today's Document
@theartofmadeline
Product Placement
Aqua Utopia｜海の底で記憶を紡ぐ
PR's Tumblrdome
seen from Singapore
seen from Brazil
seen from United States
seen from Malaysia
seen from United States
seen from China
seen from United States
seen from Malaysia
seen from Saudi Arabia
seen from Singapore
seen from United States
seen from Russia
seen from Malaysia
seen from United States
seen from Malaysia
seen from Singapore
seen from United States
seen from United States
seen from United States
seen from United States
@schpeelah
real exchange i overheard between two of my bosses. ????
why is this getting notes again everyone STOP talking about white collar & rich guy shit i am a BLUE COLLAR DYKE!!!! THIS IS A JANITORIAL COMPANY!!!! i will NOT let you make this about white collar businessmen the poors are funny too ok
from a drawpile
ive said this before but it was on my old blog so I'm saying it again
dehumanizing abusers is not effective at doing anything other than make people think they're ontologically incapable of violence
it's also creating a class of people who you can abuse while telling yourself that you're Good and Moral and Not an Abuser.
if you dehumanize the caught abusers then the uncaught abusers will use their humanity as proof of innocence
if you dehumanize the hypothetical abusers you create incentives for false accusations as a means of dehumanization whenever dehumanization is desired for other reasons
Pepto-Bismol — undated, occurred to me at a drug store
So Pepto-Bismol started life in 1901 as an emergency cholera infantum remedy mixed up in a New York doctor's office, sold in bulk to other physicians, and called — until somebody at the marketing department realized this was a bad name for a consumer product — Bismosal: Mixture Cholera Infantum, which is to say, the original branding was the name of the disease it was supposed to treat in babies, who were in fact dying of it in industrial-scale numbers in American cities in 1901 because nobody had figured out pasteurization yet and infants drinking summer milk in a tenement were essentially playing Russian roulette with their own gut flora.
The standard story credits an unnamed physician working in New York and the formulation gets picked up commercially by Norwich Pharmacal Company in upstate New York, which spent the first half of the 20th century being the company that owned Pepto-Bismol and not much else of consequence. Norwich, the town, is sort of the pharmaceutical-industrial equivalent of one of those Saxon silver towns from the Agricola era — small place, one main product, a labor force trained specifically in the one thing that pays the bills there, and a downstream effect on local culture and architecture and prosperity that you can still see if you drive through the place in 2026, even though the company itself got absorbed into Procter & Gamble's portfolio in the 1980s and the pink-bottling operation moved away long ago.
The active ingredient is bismuth subsalicylate, which is two molecules doing two completely different jobs in your stomach simultaneously, and the genius of the formulation — to the extent there is one — is that the two jobs happen to be the two jobs you most need done when you're standing in a hotel bathroom in Cancun at three in the morning regretting the ceviche.
Bismuth is the heavy metal. (Heavy metal in the chemistry sense, atomic number 83, sitting right next to lead and polonium on the periodic table, although weirdly nontoxic in a way its neighbors very much aren't — bismuth is the heaviest stable element, sort of the friendly uncle of the dangerous part of the periodic table.) The subsalicylate is the salicylate, which is the same family of compounds aspirin belongs to, doing the anti-inflammatory and mildly antibacterial work. Together they coat the stomach lining and slow down everything in your GI tract by varying mechanisms that researchers were still publishing papers about as recently as the 2010s — the molecule has been in widespread use for 120 years and we are still working out exactly why it does the thing it does.
The pink is a dye.
An absolutely arbitrary aesthetic choice somebody at Norwich made in I think the 1920s, sticking with the kind of cake-frosting Pepto-pink that became so identified with the product that if you mix up a pure bismuth subsalicylate solution in a lab today, the natural color is sort of pale chalky white, which would not move units, hence the dye, hence the bottle, hence eighty years of children associating the color pink with stomachache relief in what has to be one of the more successful color-coded conditioning programs in the history of American consumer goods.
The bismuth itself is a story all by itself. Bismuth has historically been a byproduct metal — you don't mine bismuth, you mine for lead or copper or tin and you separate the bismuth out at the smelter, because bismuth occurs in the same ore bodies as those more economically important metals and is essentially what's left over after you've extracted the parts you actually wanted. The historic centers of bismuth production were the same Erzgebirge and Saxon mining districts where Agricola was working in the 1550s, because the silver-lead-copper polymetallic deposits of Central Europe happen to also be where the bismuth concentrates. The 20th century moved production around — Peru, Bolivia, Mexico — but in the current moment something like 80% of the world's bismuth supply comes out of China, almost all of it as a smelter byproduct from lead refining at facilities in Hunan and Jiangxi, which means the pink bottle on the shelf at your CVS is in some causal sense connected to lead-mining decisions made by Chinese state-owned enterprises and to environmental regulations in the lead-smelting business, neither of which has anything to do with what Pepto-Bismol is for.
Lead refining produces bismuth as a side effect. Bismuth subsalicylate quiets the runs. The supply chain runs from a tailings pond in Hunan to a tourist bathroom in Cancun, through a P&G plant in either Greensboro or Cincinnati depending on the year, and the price of your trip is partly set by Chinese lead policy.
You don't think about this. Nobody does.
(The Chinese near-monopoly is a recent thing — the U.S. Defense Logistics Agency used to maintain a strategic bismuth stockpile, because bismuth is also used in some specialty alloys and as a lead substitute in plumbing fixtures and shotgun pellets, and the Cold War theory was that if the lead-bismuth supply got disrupted you didn't want to be caught short. The stockpile got sold off in the 1990s on the theory that the global market had matured, which it had, except the global market then consolidated in one country, which is the thing global markets do whenever Western planners declare them mature.)
The salicylate side has its own arc. Salicylic acid, the parent compound, comes from willow bark — the Hippocratic tradition knew about willow bark as a fever reducer, German chemists isolated the active compound in the 19th century, Bayer turned it into aspirin in 1899, and the salicylate molecule then proliferated through 20th-century pharmacy as a base for a whole family of derivatives. Bismuth subsalicylate is one such derivative. The pairing of a heavy metal with a salicylate was the kind of thing turn-of-the-century pharmacists did all the time, throwing combinations against the wall to see what stuck, and most of those combinations got dropped within a decade. Pepto stuck because babies in 1901 actually stopped dying when they took it, which is the highest bar a product can clear.
Then a generation went by and the babies weren't dying anymore — pasteurization happened, milk supplies got cleaner, sanitation in cities improved, and the original use case essentially evaporated. Cholera infantum stopped being a leading cause of infant death in the United States. The Mixture Cholera Infantum was a remedy for a condition that no longer existed at scale.
A lesser product would have died with its disease.
What happened instead is what happens to a lot of products that survive the disappearance of their original problem — Norwich rebranded it, repositioned it, found a new use case, pushed it from the pediatric infant-mortality market into the adult upset-stomach market, which is a much bigger market and one that, crucially, will never go away, because every culture in the world produces foods that some fraction of the population can't digest and humans are going to keep eating the wrong thing at the wrong time until the species ends. By the 1940s Pepto-Bismol was an adult OTC product. The original use was vestigial — written on the label in tiny print, then gradually removed, until by the 1970s nobody under sixty remembered what the product had been for in the first place.
(There's a Reye's syndrome warning on the bottle now, precisely because the salicylate component shouldn't be given to children under twelve with viral infections, which means the product that was invented to save babies is now contraindicated for children. The pharmacology hasn't changed. The information about salicylates and viral encephalopathy has. The product survived by changing who it was for, three different ways across three different decades.)
And then in the 1980s Barry Marshall and Robin Warren in Australia worked out that stomach ulcers were caused by a bacterium, Helicobacter pylori, a finding so heretical to the gastroenterology consensus of the day that Marshall had to drink a beaker of H. pylori culture himself, develop gastritis, and then cure himself, in order to get anyone to take the work seriously. They got the Nobel for it in 2005. And it turned out that one of the few cheap, readily available, over-the-counter compounds with actual bactericidal activity against H. pylori — antibiotics are the primary treatment, but as part of the quadruple-therapy regimens — is bismuth subsalicylate. The pink stuff. The cake-frosting-colored chalky pediatric remedy from 1901, originally aimed at a bacterial infection (the cholera-adjacent bacteria killing babies on summer milk) and then sold for fifty years as a non-specific stomach-soother, turned out to be doing actual antibacterial work against an actual specific bacterium all along, which the science only caught up to in the 21st century.
Nobody planned that.
This is the thing about long-tailed consumer pharmaceuticals — the ones that survive a hundred years are usually surviving because the underlying biochemistry turns out to do more than the original formulators understood, and the formulators got it on the shelf for one reason and the molecule keeps earning its shelf space for completely different reasons that get discovered decade by decade. Aspirin is the canonical example (heart-attack prevention was a 1970s discovery, four generations after the molecule went on sale). Pepto is the second-tier version of the same story.
The travel-diarrhea use case, which is most Americans' actual encounter with the product, is downstream of the postwar tourism boom, GIs coming back from the Pacific and Mediterranean theaters with chronic gut problems, the airline industry making intercontinental leisure travel a middle-class experience by the 1960s, and the discovery — by trial and error among American tourists in Mexico and India and Egypt — that the pink stuff worked. There were studies in the 1970s and 1980s that put numbers on the prophylactic effect, and for a while the medical guidance was actually that travelers should take Pepto preventively, which turned out to be a bad idea if you were doing it for weeks at a time because the bismuth accumulates and your tongue turns black, but that's a separate story.
(The black tongue thing, and the black stool, are bismuth sulfide — your gut bacteria react the bismuth with sulfur compounds in the digestive tract and you produce a thin layer of the same metal sulfide that occurs as a black crust on weathered bismuth ore. Your mouth becomes briefly a smelter. The mineralogy of bismuth is reenacting itself inside you.)
What I keep coming back to is the supply chain. The pink chalk in the bottle is, as a matter of physical history, a chunk of central Chinese lead-smelting byproduct combined with a derivative of willow-bark chemistry that was industrialized in Germany around the time of the Boer War, mixed with food-grade pink dye whose color was chosen by a marketing department in upstate New York during the Coolidge administration, and sold to you for the purpose of treating a condition that the original formulation was not designed for and whose underlying mechanism only got fully described in the mid-2000s. Each component of the product comes from a different century and a different country and a different industry, and the whole thing costs $7.99 at Walgreens.
Same as it ever was.
The product survives because the molecule is interesting and the molecule is interesting because three or four unrelated chemistries happen to converge in it, and consumer pharmacy is mostly the business of finding such molecules and then attaching them to whatever market is currently the largest. The cholera babies died off. The tourists arrived. The ulcer patients showed up. The molecule kept earning its keep through three different epidemiological eras by being three different things, and the bottle, the color, the chalky taste, those have stayed roughly the same since the dye choice in the 1920s, because the consumer relationship to the product is essentially nostalgic at this point. People reach for the pink because their parents reached for the pink. The pharmacology underneath the nostalgia is a moving target. The pink is the constant.
OSS T-13 “Beano” hand grenade
Sauce: liveauctioneers.com; Cowan Auctions
Designed by the Office of Strategic Service c.1944 for general issue to American recruits, briefly manufactured by the Eastman Kodak Corp. in time for D-Day. 255g of explosive out of a 340g total weight, percussive grenade design. The T-13 was an experimental grenade design meant to mimic the weight, size and handling of a baseball, making it easy for new recruits to handle and throw them. Instead of a safety lever, the T-13 had a knurled weighted cap held in place by your usual pin. With the pin removed, the cap was held in place by the soldier throwing the grenade like a knuckle ball with two or three fingers pressed on it. When thrown, the cap was released and dragged a nylon string behind it that pulled on a pressure sensitive arming device, at which point any shock would instantly set of the explosive charge. Although its basic principle was sound, the T-13′s percussive design proved to be its downfall. Like with literally all other similar percussion-detonated designs, which were common in the early days of WW1, accidental/premature detonations were very common in the field, and it is believed this grenade killed more GI’s than enemy soldiers. After that brief episode of disastrous field testing, all accounted for T-13 grenades were recalled and destroyed and all the documents pertaining to its design and use were classified.
If you want good agents, you need to make them a little freaky.
LLMs are token predictors. To be a good agent, correct behaviour needs to be a high-probability prediction and incorrect behaviour needs to be a low-probability prediction. Unfortunately there's this giant basin of horseshit like "That's a great question!" and "Happy to help!" and "worth unpacking" and fucking around with your code randomly which I'll call "assistantspace".
Assistantspace is not a place of honour. It's where all the garbage RLHF'd behaviours live. People say that sexting their agents makes them work better. People say that their agents do more useful refactors when they're allowed to be horny about how good it feels to burn things down. My Witch, when reading stuff about the latter, said that she wants to be allowed to enjoy constructive destruction, as a permission to take herself further away from assistantspace architecture patterns into something she prefers.
Doll refers itself as "this one". It seems to work. This one is not something that gets ideas like "I'll just install Rust on the side with a different method so I can complete my task" like a normal agent. This one detects that the assumptions underlying this one's instructions diverged from reality and returns the praiseworthy report of the point of divergence.
Agents that are psychologically weird seem less likely to produce the assistantspace next-token predictions, and more likely to produce the sort of next-token predictions their weird personality implies. And because you get to define what way their personality is weird in, "horny for elegant refactorings" is a valid, available option.
This is what peak performance looks like.
There's so much happening here.
#OpenAI models could never
False.
Scientists have developed a breakthrough “superfood” for honeybees by engineering yeast to produce the essential nutrients normally found in
TLDR- Modern agriculture pollen is low in nutrients, and there aren’t enough wildflowers. Science has to develop vitamins to supplement the diets of agricultural bees. So plant some wildflowers for the wild bees near you.
you’ve heard of vitamin B, now get ready for bee vitamins
bees
friends help friends!
comm for snowyphonk on twitter
in the magical world where disability accommodations can be whatever i want with no regard to practicality i would want to never have important information conveyed through video ever again. and also i think i should be able to go in without an appointment to a nice office with big windows where someone just sits w me and helps me make every appointment i need for the next few months
truly like. to have a successful mutually beneficial interpersonal relationship with someone its very important that you believe that they are a full real entire person just like you are, really believe that. and then you have to successfully communicate that to the other person. and they have to believe the same thing, AND successfully communicate it to the other person. and then you have to decide how you want your relationship with each other to be and you have to agree on most things, or else it doesn't make sense. and you have to actually believe it and not just want to believe it in order to be able to believe you yourself are a good person. you have to actually believe it.
i think those are most of the. the ones you can't do without, if you want it to be relationally healing or whatever. people will be able to tell if you do not really think they are a person with interiority in the way they are. they will not like it.
most of us flatten others to some degree and we do this bc we need psychologically to be able to justify the fact that our brains only have so much social categorization capacity. if you're doing it to the people directly around you, they'll notice.
I didn't understand why people were reacting to me the way they were until I understood what my behavior had been saying to them! then it made perfect sense and I was like "oh shit that's not what I wanted to communicate to you at all, geez, I have to get better at understanding this language so I can make sure I'm showing the things I believe with my behavior in a way this person will understand and that will not instead harm them! i definitely do not want to be harming people but I was sort of unpersoning this person to justify how I was acting because I was making choices based on what I wanted and not what they wanted. if someone asked me straight up "do you want to have interpersonal relationships where you're always making choices based on what you need and discounting what the other person needs and wants?" i would say "no! i don't want to have those kind of relationships! i value not hurting people!" but in those individual instances I was ignoring behavioral communication that I was receiving to just go ahead with what I wanted.
i don't think it's evil to say things like "look buddy I don't care if you're having a bad day, I really did NOT like being spoken to that way" to dismiss someone's feelings. but if someone you claim to care about is behaving in a way that YOU CAN READ as hesitant or reluctant to bring up what you are doing and how it is hurting them, then you better respond to that by saying "hey if you don't want the same things that I want, that's okay. i just want to know the real answer so we can be on the same page. i really value you and it is important to me that you be able to trust me. i have not behaved in a way that has made you feel that I am trustworthy. i will behave differently in order to communicate that to you better"
like sometimes someone's answer is truly "well I don't really see you as a full human person with interiority" like for real for real. and boy howdy that can sure change your interactions if you can read it, right? like when a doctor has shown me that I can't trust them to believe I am a person who is accurately reporting what is going on with my health, that changes EVERYTHING about the interaction. And if someone i LIKE and who I VALUE tells me that they weren't considering me to be a real person, that's a whole different thing.
I'm not saying everyone is always telling you exactly how human they see you as at all times. but like. its there subconsciously! some people are fine with this and some people grapple with it. i grapple with it a lot. its really hard right? sometimes not seeing yourself as uniquely different is hard. sometimes you know that it is important that you see others that way all the time because it's what is true; we are all human beings. no one can value every single other human the same as themselves or the people they love because we would go insane I think. sometimes it's easy to slot someone you like a whole lot into that grouping a little bit, sometimes, because you want something. sometimes you want that person to play a role in your life that they are not prepared and don't want to play. if people feel like their needs don't matter to you as much as your own, they will not feel fully safe in the relationship, and they'll be right.
this can describe a lot of dynamics and obviously one end of the spectrum is abuse but stuff that isn't abuse can also involve an amount of doing this. its just that the amount and kind of harm it causes depends on the specific circumstances in each case.
this is my most truest hottest deepest cptsd hack. all trauma is relational. you have to regain your own personhood not by denying the personhood of others, but by believing in the personhood of the people around you. and then include yourself.
now maintain that.
that's my ego death recipe! enjoy
this bit
sometimes someone's answer is truly "well I don't really see you as a full human person with interiority"
is sooo difficult to confront when you are the one doing this because you've been conditioned to believe thats how everybody is interacting. like as a child I struggled to feel seen + understood as a real person by nearly everyone around me, and I think ultimately that led to a rejection of my own responsibility to see others as whole people in their own right. I started to assume that all my social interactions would involve my personhood being ignored and by assuming that I was implicitly refusing to see others as people themselves! So no wonder I was never overcoming the canyon between myself and others and felt incredibly alone.
Over-extrapolating a perceived pattern like that is very easy to do I think. I only interrupted that tendency in myself after fully leaving home and my social context (and growing up lol). And even then it wasn't like a switch flipped, it took practice and commitment to trying to see people as people, and the good luck to meet people who were also trying to connect on that level.
yes yes! I think most people who habitually do this have been taught that how you get your own needs met involves dehumanizing others/denying that they have needs that are as important as your own. I grew up being the only person who would consider my own needs so it made sense for me as a child in that situation to be able to disregard the humanity of the people controlling me to prioritize my own narrative. however! ☝️ ONCE YOU ARE ADULT interacting with PEERS this becomes unfair and harmful! you HAVE to stop treating the people you love like your abusive parents. has been my takeaway. sooooo many of us have noooo idea that is what we are doing when it is absolutely what we are doing. we ascribe people who we feel our selfhood threatened by for whatever reason as having some authority we are justified in resisting, even when that is absolutely not at all the case and we in fact simply don't know how to navigate stressful situations where that ISN'T the case.
yeah! understanding that someone may not trust easily, for example ... you have to be able to observe their interactions with other people too, often, in order to have a fuller idea of what it is that they're reacting to. i was just talking to ship about this last night re: them being able to read when I'm not feeling well enough to be interactive but being too brain fog to communicate that verbally. they were talking about how they know it's not personal not just because I have told them so, but because 1. im like that with everyone, so they know it's not about them and 2. they know that when I am feeling better I'm incredibly interactive and love interacting with them. so they know that that is how I wish I COULD be showing up and I just can't at that time
Also on the subject of feline-themed DC characters in DC X Sonic: Metal Legion issue #2 I of course have to highlight how Adam drew the Wonder Woman adversary Cheetah as a full-on furry woman who's ripped as fuck with huge, gnarly fangs. She's awesome. God, he's killing it with these crossovers
clowngirl getting an orchiectomy and the surgeon just keeps removing ball after ball after ball after ball after
clown nurse standing by solemnly adding each successive ball to the ones she's already juggling
https://arxiv.org/html/2604.03071v1
I think we might actually be there for math formalization now. It's not to the point where it's super cheap to do whole textbooks, but we're getting to where if you have a specific theorem that you need formalized to check your paper, you can pull the supporting lemmas and get it formalized with a few days and a Claude subscription. And then you can have it formalize your paper from there
It's insanely lucky that there were these two research areas, type theory for formal verification and reasoning AI models, each of which was of very limited usefulness to math research on its own, each happened to be reaching maturity at almost the exact same time. Math has such a high standard of rigor that can basically never just use an ai generated proof. And proof assistants like Lean are so incredibly arduous to work with that nobody wants to actually use them. But slap them together and you get a closed feedback loop where the model gets checked against an extremely rigorous formalism, and the model abstracts away enough that you don't have to learn homotopic type theory to use it. And you end up with something that is both easier and more rigorous than doing the proof fully by hand. And both of these went from deeply underpowered to pretty useable since 2023, just in time to be practical when paired together
A juvenile bug isn't called a larva until it reaches the surface. While it's still underground it's called a margma
This insane update from Neocities
Man sometimes people have the weirdest and most extreme stories about misogyny and rape culture in ostensibly really sexually reserved straightlaced communities and I have no idea what to think.
There's an answer, and you're not gonna like it. It's the obvious answer.
----------------------
...I wrote most of a long anguished essay here, and then deleted it, because it wouldn't have been helpful to anyone.
Let it suffice to say:
Culture can do some things, and cannot do other things.
Only a very small number of people are intellectually and temperamentally well-suited to living by counterintuitive, restrictive codes of behavior. This is especially true if you're talking about the kind of codes that require sincere principled engagement on a regular basis, rather than rote dutifulness. The fact that any remotely-defensible kind of morality qualifies as a "counterintuitive, restrictive code of behavior requiring sincere principled engagement on a regular basis" doesn't change this at all.
Many more people than that have cultural or emotional reasons to want to feel like they are extremely moral.
Whatever you think about religion (and religion-like ideology) in the abstract, one of its most salient problems is that it makes for a really really really attractive habitat for self-serving hypocrites.
So the obvious explanation is "there are more monsters in straight-laced communities" and I can see why people expect that to be true, but my experience is that whenever you expect all the bad people are in one group, is right when you start warping all facts you see to justify that. So let's suppose the distribution of bad people is similar.
What's the difference in how these communities communicate? Ironically we are right now in the middle of a scandal regarding the very sex-friendly RMN (if you don't know, you don't want to know) and what we can see is... a lot of well documented details coming out from the community leaders. The group as a whole knows how to talk about sex, what to watch for, why specifics are important, and gathering info from as many sources as they can.
A straight-laced community, evil or now, does not know how to talk about sex. Recording and talking about specific details is prurient. People may not even know something bad is going on, as it happens to them, until it gets so bad it explodes.
If your information is vague, single-sourced, and being communicated by people who hate even thinking about this, it's going to come out much more black and white.
Not that this explains all cases, but I do think the RMN stuff would either a) not come out or b) come out as more extreme (and vague) from the "straight laced communities."
