Sir Arthur Conan Doyle created one of the most beloved fictional detectives of all time in Sherlock Holmes, spawning countless novels, films, TV series, and so forth featuring the sleuth. Far fewer people know about the real-life inspiration for Holmes: a renowned surgeon at Scotland's University of Edinburgh named Joseph Bell, who taught Doyle when the author was still a young medical student. Bell is the subject of a new, eminently readable biography, Sleuth-Hound: The Case of the Real Sherlock Holmes, by medical historian Lindsey Fitzharris.

A noted science communicator with a fondness for the medically macabre, Fitzharris published a biography of Scottish surgical pioneer Joseph Lister, The Butchering Art, in 2017—a great, if occasionally grisly, read. She followed up with 2022’s  The Facemaker: A Visionary Surgeon’s Battle to Mend the Disfigured Soldiers of World War I, about a WWI surgeon named Harold Gillies who rebuilt the faces of injured soldiers.

In 2020, Fitzharris hosted a documentary for the Smithsonian Channel, The Curious Life and Death Of…, exploring famous deaths, ranging from drug lord Pablo Escobar to magician Harry Houdini. Fitzharris performed virtual autopsies, experimented with blood samples, interviewed witnesses, and conducted real-time demonstrations in hopes of gleaning fresh insights. And last year she and her husband, caricaturist Adrian Teal, collaborated on an illustrated children's book, Dead Ends! Flukes, Flops, and Failures that Sparked Medical Marvels.

With Sleuth-Hound, Fitzharris returns to the Edinburgh of The Butchering Art, of body snatchers and surgeons' discovery of ether, which was also the Edinburgh of Arthur Conan Doyle and Joseph Bell. Bell was notorious for his peculiar method of diagnosis: he was able to infer a patient's occupation and personal habits, as well as any ailments, just by looking at them. For instance, Bell once figured out the route a patient had taken walking to the infirmary by identifying the type of clay on the patient's boots—an incident Doyle used in The Sign of the Four. Bell was also something of a medical detective, working with a local police surgeon on some of the most famous murder cases of his day.

Ars caught up with Fitzharris to learn more.

Ars Technica: What drew you to writing an entire book about Joseph Bell?

Lindsey Fitzharris: I've known about Bell because I'm a medical historian and I knew that he was an inspiration for this character, but nobody had written a big commercial book on him. There was a book in 1982 by a guy named Ely Liebow. So people in the Sherlockian world, they know a bit about Bell through him, but I really wanted to bring a medical history lens to it.

When I sold the Sleuth-Hound proposal, I had never read the original Sherlock Holmes tales. I decided to put off reading them initially because I knew enough about Bell to know that it became an irritating comparison; he found it rather reductive at the end of his life. I really wanted to bring the real man behind the myth to life in Sleuth-Hound, not just as the inspiration for Sherlock Holmes, not just as this parlor trick, but as this complex character. I've been talking to a lot of Sherlockians who have been really surprised at how much Bell is Sherlock, but also how much he isn't Sherlock. That's an interesting discussion as well: why Conan Doyle focused on this one aspect of him.

Dr. Joseph Bell Credit: Public domain

Ars Technica: What are some of the ways Bell is and isn't Sherlock?

Lindsey Fitzharris: He had this Sherlockian way about him where he could tell everything about a person before they ever spoke a word: a callous, a stain, a limp, all were clues to Bell if you only knew how to see it. But in a lot of ways he wasn't Holmes. For instance, he was a deeply empathetic character. He was very feeling for his patients. And he also was a family man. His wife dies when he's in his 30s, and he doesn't remarry his entire life. He holds the torch for her. She dies in childbirth, and he raises his three children alone.

My husband is a caricaturist, and we have this discussion about how he can only caricature people he doesn't know well. That's because caricature is about those first impressions. I think Conan Doyle knew Bell in his capacity as a student, but he didn't know Bell the man. So he took the most recognizable flashy trait of Bell and transposed that onto Sherlock Holmes. The question I pose at the end of the book is, would Holmes exist if Conan Doyle had known Bell better, or would he exist in the same format?

Ars Technica: Did Conan Doyle's contemporaries recognize the similarities?

Lindsey Fitzharris: Those who knew Bell recognized it immediately. Robert Louis Stevenson reads Sherlock Holmes. He writes to Conan Doyle and he says at the end of the letter, "One thing bothers me, could this be our old friend Joe Bell?"

But Bell was a lot more complex, and I really wanted to bring that medical history lens, because this isn't just a story about his relationship with Conan Doyle and the inspiration for Sherlock Holmes. It's about the development of forensic medicine. It's about the moment that medicine enters the courtroom for the first time. Bell wasn't just the inspiration for Holmes; he was actually doing what Holmes was doing in practice. He was working officially with a police surgeon named Henry Duncan Littlejohn, who was fascinating in and of himself. He was a pioneer of fingerprinting and courtroom photography. They end up partnering and working on some of the biggest high-profile cases in Scotland in the 19th century.

Ars Technica: I was frankly shocked about the baby farming case; I had no idea that was a common practice.

Lindsey Fitzharris: It's such a crazy story. The thing that I want people to contextualize here is that there was no birth control. So there were a lot of women having a lot of children, and they were deeply impoverished, and they didn't have enough money to feed these children. So they turned to the baby farmer. In theory, the baby farmer took the child and then took a fee from you, the mother, with the promise that they would farm out the child to somebody who could take care of the child. That was how it was supposed to work. There were no legal routes for adoption in the 19th century, so they turned to the baby farmer.

The case in Sleuth-Hound, the baby farmer, Jessie King, ends up taking the fee and then just killing the infants. This was a horrific case. The chapter opens with these boys finding this bundle of rags and using it as a football and kicking it around. And then they open it, and it's the decomposing remains of a child. Littlejohn and Bell work on the case. They perform the postmortems; they testify in the trial.

One of the dark themes of this book is that a lot of early forensics was built on the brutalized bodies of women in particular. Jessie King was killing infants, but this is a crime that is particular to women—women trying to get their children into families that could afford to feed them. A lot of the crimes are committed on women who never receive any justice, but the science outlives what their bodies taught these early forensic investigators like Bell and Littlejohn.

Ars Technica: Case in point: the infamous Chantrelle murder case and how it coincides with the rise of toxicology.

Lindsey Fitzharris: The Chantrelle case is so fascinating to me. Eugene Chantrelle was this respectable gentleman who lived in Edinburgh. He was a French tutor, and his wife Lizzie is seemingly poisoned by a gas leak, a broken pipe in the flat, and she's removed from the property and later dies. But it turns out that Chantrelle was abusing her the entire marriage.

We think Littlejohn brought Bell onto the case. Bell and Littlejohn discover that it was opium poisoning, and they test the vomit on her nightgown, and they're able to tie him to this crime. Bear in mind, Bell worked like Sherlock Holmes. He worked in the shadows. So in these earlier cases, it was very difficult to know if Bell really did work on them.

But there's a really intriguing clue at the end of this story. Littlejohn investigates, [Eugene Chantrelle] is convicted, and he is hanged for this murder. On the scaffold, he allegedly looked around at the audience, locked eyes with Littlejohn, and shouted, "Give my regards to Joe Bell. He did a good job putting the noose around my neck." This was such a Sherlockian moment. How did Chantrelle know that Bell had worked on this case? We don't have any records that actually confirm this except that he allegedly said this on the scaffold.

The other extraordinary aspect of this story is that Chantrelle was drinking buddies with Robert Louis Stevenson, so he becomes one of the inspirations for Dr. Jekyll and Mr. Hyde: this respectable person on the outside and very sinister and evil on the inside. The prosecutor told Stevenson that this was not Chantrelle's first victim and that he had killed four or five other people with his opium and cheese parties.

Ars Technica: Your book launch should feature an opium and cheese party.

Lindsey Fitzharris: (laughs) And ether drinks. You're invited. The Victorian period was wild because there were a lot of poisons that you could just get over the counter from your pharmacist. Chantrelle had trained as a doctor at one point and had bought a huge amount of opium just before the murder. Even the chemist thought that this was strange, but he assumed that Chantrelle was a doctor. So poison was a big problem in the Victorian period. Some of it was legal, it was hard to detect. Forensic medicine was just developing. I call it the golden era of murdering because you probably had more of a chance in the 19th century of getting away with it.

Ars Technica: Sleuth-Hound is written in a deceptively novelistic style, but you've clearly done very extensive research on this. What sources did you find most helpful? 

Lindsey Fitzharris: One of the things that happened was that Bell's diary became available, that had been locked away for over a century by the Bell family. Perhaps they were waiting for enough time to pass. I almost called the book Sleuth-Hounds because I do also think Littlejohn taught Conan Doyle as well. Bell gave Conan Doyle the method, and Littlejohn gave him the world of forensics. He was a professor of forensics in Edinburgh at the time. We can flesh out what Bell was doing because we know that they worked together often on these cases.

You do get dialogue through the murder trials, because there's so many documents and court records that go along with these trials. There are a lot of letters, diaries. Nothing's made up, but sometimes I had to take educated guesses. For instance, Conan Doyle is a terrible narrator of his own life. He constantly is contradicting himself, even in his own memoirs. It was difficult figuring out when he came into contact with Bell for the first time. I believe it's in the winter of 1878, and I provide a lot of context for that in the back notes of the book.

I'm a historian. It's got to be armed to the teeth with sources. But I want people to be able to pick up the book, not know anything about Sherlock Holmes, not know anything about medical history or forensics, and just be able to enjoy it. Medical history is the most universal way into the past. I make the past relatable. I tell you what that was like a hundred years ago.

My superpower in 2026 is that I'm one of the few people who can make you feel good about living today. Progress is not a straight line necessarily, but if you had that medical crisis a hundred years ago, it would've been a lot worse. I've been through cancer myself. Today is the best day to be alive and to get that diagnosis and tomorrow will be a little bit better, because I'm an optimist when it comes to medicine. I'm hoping that chemotherapy or radiation treatments will become obsolete. They're brutal. And there's a cancer vaccine on the horizon.

Ars Technica: In the US, sadly, we're going backwards on vaccines for preventable diseases like measles.

Lindsey Fitzharris: I know. We have short memories. We don't remember. It's weird to me as a medical historian because I remember it all. I remember the times when parents lost their children to preventable diseases. It's coming back, and it's sad to see that repeat. If I could just show people the diaries and the letters of brokenhearted parents who lost child after child to measles or polio and all of these diseases we can now treat and prevent. I put up pictures of smallpox on social media, and people think it's AI. They can't believe that something like that existed. It's the only human virus we've ever eradicated. It should be celebrated. So I'm always out there banging the drum. I hope that we steady the ship a bit and people learn from the past again.

Read full article

Comments

The Environmental Protection Agency late last week announced it would end a 20-plus-year effort to develop methods to monitor air pollution from factory farms, effectively shuttering the agency’s stop-and-start attempts to regulate the potentially pollutant-spewing facilities under the Clean Air Act.

The announcement came just days before the midterm elections and was characterized by the agency as a long-awaited reprieve for farmers, a historically important constituency for Trump and the Republican Party.

When tens of thousands of animals are packed into one place, pollution from their waste is blown from ventilation fans or emanates from giant lagoons of liquid effluent. Livestock facilities emit fine particulate matter, ammonia, nitrogen oxides, sulfur dioxide, and volatile organic compounds, all contributing to a toxic mix. Research has shown that emissions from these operations are a major cause of air pollution, causing nearly 12,700 deaths in the United States a year.

Strictly speaking, concentrated animal feeding operations, or CAFOs, are required to get permits under the Clean Air Act and report emissions of these pollutants under the Comprehensive Environmental Response, Compensation and Liability and Act and the Emergency Planning and Community Right-to-Know Act. But through a deal crafted in tandem with the EPA and the livestock industry, they’ve been given amnesty from doing so for more than 20 years.

The arrangement goes back to the early 2000s when the agency attempted to prosecute CAFOs for exceeding limits on some of these pollutants. The industry pushed back, arguing there was no adequate way of measuring them. So the agency worked with industry groups to develop what was dubbed the Air Consent Agreement, through which the EPA granted thousands of CAFOs safe harbor from legal action in exchange for their participation in a study that was funded by the industry.

The goal of the study, the agency said, was to develop tools and models that CAFOs could use to “determine whether they emit pollutants at levels that require them to apply for permits under the CAA or submit notifications under CERCLA or EPCRA” and “produce a scientifically sound basis for measuring and estimating air emissions” from these facilities.

But since the adoption of the agreement in 2005, the agency has failed to produce any air “emissions estimating methodologies,” even prompting a 2017 report from the agency’s Office of Inspector General that faulted it for slow-walking the process.

Meanwhile, the Trump administration and Congress have also gutted reporting requirements for factory farms to report air pollutants under CERCLA, also known as the Superfund Act, and EPCRA.

“As EPA has been saying it’s proceeding with this plan, it’s been dismantling all of its own tools to regulate emissions,” said Tarah Heinzen, the legal director for the advocacy group Food & Water Watch, which has tracked the proceedings from the beginning and filed dozens of lawsuits attempting to push regulators to control pollution from CAFOs. “Now it has announced it has no intention, basically, of ever holding factory farms accountable under the Clean Air Act, either.”

In its October 30 announcement, the agency noted that in the two decades since the study, known as the National Air Emissions Monitoring Study, was conducted, many farms have modernized their barns and manure systems.

“Good science requires good data, and these data are too old and too limited to support methods that farms could rely on,” said Abigale Tardif, the acting assistant administrator for the agency’s Office of Air and Radiation, in a statement. “We would rather tell farmers that plainly than issue numbers we cannot stand behind.”

Rob Brenneman, president of the National Pork Producers Council, which has long pushed for the decision, said in a statement: “This move confirms what farmers have long known: Every farm is different, and there is no one-size-fits-all model to apply to them.”

Heinzen noted that the agency’s decision simply means that this administration has no intention of regulating air emissions from CAFOs, but for now, the Clean Air Act still requires permits for facilities that exceed emissions thresholds. The amnesty granted to these facilities will no longer be in effect now that the agreement has concluded.

“There’s no EPA-issued methodology to estimate emissions,” she said. “That doesn’t mean that there aren’t good studies out there of emissions from different types of animal feeding operations that can be used.”

Heinzen pointed to California, which has models for livestock emissions that it uses to enforce its own air regulations. “We just don’t have one set of national emissions factors from EPA,” Heinzen said. “That does not absolve these factory farms from their obligation to comply with federal law.”

Still, under this administration, it’s clear air emissions from factory farms won’t be measured and monitored any time soon. “No one’s going to estimate anything,” Heinzen noted. “As EPA said in its press release, we’re certainly not about to try to enforce the law.”

The Air Compliance Agreement did not extend to methane. The US livestock industry is the single-biggest source of methane in the country, but Congress effectively blocks livestock facilities from reporting methane emissions in annual spending bills.

Georgina Gustin covers agriculture for Inside Climate News, and has reported on the intersections of farming, food systems and the environment for much of her journalism career. Her work has won numerous awards, including the John B. Oakes Award for Distinguished Environmental Journalism, and she was twice named the Glenn Cunningham Agricultural Journalist of the Year, once with ICN colleagues. She has worked as a reporter for The Day in New London, Conn., the St. Louis Post-Dispatch and CQ Roll Call, and her stories have appeared in The New York Times, Washington Post and National Geographic’s The Plate, among others. She is a graduate of the Columbia University Graduate School of Journalism and the University of Colorado at Boulder.

This story originally appeared on Inside Climate News.

Read full article

Comments

An anonymous reader quotes a report from VentureBeat: Researchers from the Massachusetts Institute of Technology (MIT) Computer Science & Artificial Intelligence Laboratory (CSAIL) reported today that they used OpenAI's GPT-6 Astra language model to raise a small, open source Qwen model's math score roughly sixfold by directly editing its weights -- the numerical parameters that govern its behavior. Rikiya Takehi, Jayden Teoh, Alex Zhang and Omar Khattab described their work in a research blog post published November 2. They call the approach "weight-space programming" and tested Astra's ability to interpret and alter networks without using conventional training procedures to generate the edits. In the strongest language-model example, Astra's intervention raised Qwen3.5-0.8B's MATH-500 score from 7.9% to 48.1% while touching approximately 0.016% of its weights. The improvement came primarily from correcting repetitive reasoning—a distinction that shaped both the technical findings and the debate over their implications. Their experiments explored a broader possibility: AI systems that inspect, repair and construct other neural networks by working directly on their internal machinery. "This is cool as heck!" wrote Yuchen Jin, a technical staffer at Databricks and former co-founder and CTO of Hyperbolic Labs who was unaffiliated with the research, on X. "Imagine using a large LLM to directly edit a smaller model's weights and dramatically improve its capabilities. It's like gene editing for neural networks. Even cooler than distillation!" Voyage AI staff researcher Chandan Singh, formerly of Microsoft AI, also reacted to the findings: "For anyone questioning whether this post is overhyped, I think they are actually *underselling* the approach."

Read more of this story at Slashdot.

This is why you get intimate with code, line by line. Artemiy at Sinevibes obsessed over details, breaking a filter model by rewiring its internal connections -- and now, with v3, giving it precise modulation in a wide variety of shapes. Now, like kids kicking over a sandcastle, we can take that handiwork and apply it to total, absolute destruction.

The post Mangle everything: modulated mayhem in circuit-bent Sinevibes Malfunction v3 appeared first on CDM Create Digital Music.

Vibe-coded clones of Adobe software are shaking things up.

Sam Altman, in an interview, said it's 'terrible' for people to have 'a bunch of the world's limited supply of DRAM' that they barely use.