Linkfest #32: Technofossils, "StickTok", and A Web Site You Update By Calling It
Hello there!
Time for "the opposite of doomscrolling” — my latest “Linkfest”, in which I carefully boil the entire Internet to produce a savory reduction containing the finest posts about science, culture and technology, just for you.
If you’re a subscriber, thank you! If not, you can sign up here — it’s a Guardian-style, pay-whatevs-you-want affair; the folks who kick in help keep it free for everyone else. And also — forward this email to anyone you know who’d like it!
Let’s begin ...
1) 🚲 Jumping robot bicycle
This stunt was achieved by researchers at the Robotics and AI Institute, an offshoot of Boston Dynamics (the company that makes those four-legged “Spot” robots). Over at IEEE Spectrum, Evan Ackerman describes how the researchers used reinforcement learning to train the bicycle to pilot itself and pull off these jumps …
There’s no independent physical stabilization system (like a gyroscope) keeping the UMV from falling over; it’s just a normal bike that can move forward and backward and turn its front wheel. As much mass as possible is then packed into the top bit, which actuators can rapidly accelerate up and down. “We’re demonstrating two things in this video,” says Marco Hutter, director of the RAI Institute’s Zurich office. “One is how reinforcement learning helps make the UMV very robust in its driving capabilities in diverse situations. And second, how understanding the robots’ dynamic capabilities allows us to do new things, like jumping on a table which is higher than the robot itself.
Another cool detail: They also used reinforcement learning to get a Spot robot to run faster than ever. The reinforcement system stumbled upon an interesting trick — it modified the robot’s trot so there is …
… an added flight phase (with all four feet off the ground at once) that technically turns it into a run. This flight phase is necessary, Farshidian says, because the robot needs that time to successively pull its feet forward fast enough to maintain its speed.
Fascinating piece! I’m really intrigued by the possibilities here. Bicycle robots would neatly blend the agility of a four-legged quadruped robot with the higher-load-bearing capabilities of a ‘bot that rolls on wheels instead of legs.
2) 🌾 Iphone app that makes you touch actual grass
Ever worry that you’re doomscrolling too much?
“Touch Grass” is an Iphone app that will lock your access to social-media apps, and unlock it only if you take a photo of yourself actually touching grass.
As the creator tells Fast Company …
“I was sick and tired of my reflex in the morning being to reach for my phone and scroll for upwards of an hour,” Kentish says. “It didn’t feel good and I wasn’t getting anything out of it.” [snip]
Currently, the app uses Google’s image-labeling Cloud Vision API to verify that the grass has, indeed, been touched. However, Kentish says, the app has gone so viral that he’s considering training his own image-detection model for cost-reduction purposes before Touch Grass makes its App Store debut.
Pretty cute! It’s probably easy to fake — you just submit a picture of yourself previously touching grass. It’d be more algorithmically solid if the app required you to show live video of your hand gently stroking the green turf of a lawn … but I suspect that would jack up the visual-recognition costs pretty high.
Relatedly, while suburbanites and rural-dwellers might have ready access to grass — me, here, in Brooklyn? Man, I’d need to travel several blocks before I could find a piece of sod. That might be feature, though, and not a bug 😅
3) 🏄 Are you a Wikipedia “busybody”, “dancer”, or “hunter”?
It’s easy to fall into a Wikipedia hole: You start reading this article, then link over to that one, then this other cool one … then, whoops, an hour vanishes.
A group of researchers wondered whether there were common patterns in the way people poked around Wikipedia. So they got 482,760 people (from 50 countries, spanning 14 languages) to give them their Wikipedia surfing data, and then analyzed it.
“The busybody loves any and all kinds of newness, they’re happy to jump from here to there, with seemingly no rhyme or reason,” says co-author Professor Dani Bassett, from the department of bioengineering at the University of Pennsylvania, USA. (Guilty as charged.)
“This is contrasted by the ‘hunter,’ which is a more goal-oriented, focused person who seeks to solve a problem, find a missing factor, or fill out a model of the world.” [snip]
“The dancer is someone who moves along a track of information but, unlike the busybody, they make leaps between ideas in a creative, choreographed way,” says co-author Professor Perry Zurn, a philosopher at American University, USA.
“They don’t jump randomly; they connect different domains to create something new.”
You can read the full paper for free here.
Now I’m wondering what type of Wikipedia surfer I am. I think use different strategies at different times: I’m a hunter when I’m trying to nail down research for a book or an article I’m writing, but if I’m just dorking around and procrastinating I do busybody-style surfing. Mayyybe I engage in dancer-like connections when I’m on the hunt for new ideas, and mashing up different domains?
4) 🐮 The linguistic practices of cows
Here’s a fun profile of the Dutch linguist Leonie Cornips, who has in recent years been studying how cows communicate. It’s surprisingly complex:
Patience … turns out to be crucial when cows communicate with each other. When a mother calls her calf, it sometimes takes 60 seconds for the calf to respond. The space between is filled with bodily gestures. Studies from Austria show that ear positioning and neck-stretching are integral to cow language. Humans think of the ability to wiggle our ears as a party trick. For a cow, it appears to be fundamental to communication. The first sentence in a conversation with a cow is likely to involve movement of the ears and a look.
Because cows are penned and live on farms, they’re surrounded by objects made by humans. It turns out they use these objects to communicate …
She observed one herd where individuals used their bodies to bang on an iron fence to communicate with the rest of the herd at feeding time, which she views as a type of language. She noticed cows responding to her differently depending on whether she entered a barn with solid walls or open sides: since cows on different farms are surrounded by different physical features, this offers distinctive opportunities for linguistic expression. Cows, she argues, develop diverse languaging practices – almost like dialects – where meaning depends on the shape of their surroundings.
It appears that cows alter their vocalizations when they’re attempting to communicate with humans …
"In becoming a dairy cow," she says, "they must have very rich communicative skills because they have to understand what the farmer wants them to do … which is not easy."
Cornips has analysed recordings to show that cows will simplify their vocalisation once a farmer recognises their need. Rather than having their intelligence bred out of them to be more compliant, Cornips thinks domestic animals are forced to develop a fuller communication repertoire than wild animals.
My final fave observation in this piece: Cornips argues that the language of cows may be even more complex than the language of whales, because cows have more ways to produce sound …
But Cornips points out that whales lack some of the capacities of cows. A whale's ears are not as moveable as a cow's, she says. They also lack hooves. "Whales cannot express themselves bodily very much," Cornips says. "In that way they may be [less complex] than cows."
Great piece, check the whole thing out!
5) ‼️ “Huh?” One in seven utterances is an interjection
For decades it was hard for linguists to study the spoken word, because very little of it was recorded. These days, though, we’re swimming in recordings, so now linguists are making intriguing new discoveries about the fabric of everyday chitchat.
One of the more interesting? The role of interjections like “huh?”, “ummm,” or “yep”.
These interjections are incredibly common in speech, as Bob Holmes writes …
“One in every seven utterances are one of these things,” says Dingemanse, who explores the use of interjections in the 2024 Annual Review of Linguistics. “You’re going to find one of those little guys flying by every 12 seconds.”
Why are they so common? Because they’re crucial signals about whether we’re hearing and understanding one another. For example, “Mmmm-hmmm” is a quick way to signal “yeah I get it”. (With the bonus that you don’t need to open your mouth, so “mmmm-hmmmm” doesn’t suggest that you’re about to start talking and interrupt the other speaker: Awesome phonemic UI design.)
Meanwhile, “huh?” is an exceptionally efficient way to convey “I’m not following you …
That need seems to be universal: In a survey of 31 languages around the world, Dingemanse and his colleagues found that all of them used a short, neutral syllable similar to huh? as a repair signal, probably because it’s quick to produce. “In that moment of difficulty, you’re going to need the simplest possible question word, and that’s what huh? is,” says Dingemanse. “We think all societies will stumble on this, for the same reason.”
I particularly dug the discussion of “grounding” — or, the art of signaling “what each participant thinks about the other’s knowledge”.
Some languages like Mandarin have built-in ways to indicate when I’m telling you something that I know you don’t know versus when I’m telling you something that I know you do know. But many other languages don’t, so they use interjections to establish these states of awareness …
One of Wiltschko’s favorite examples is the Canadian eh? “If I tell you you have a new dog, I’m usually not telling you stuff you don’t know, so it’s weird for me to tell you,” she says. But ‘You have a new dog, eh?’ eliminates the weirdness by flagging the statement as news to the speaker, not the listener.
The whole piece is damn fascinating. One of the researchers is researching whether we use emoji for some of the functions that interjections serve in F2F conversation. I can’t wait for them to write up that research …
6) 🌐 “Take me to a useless web site”
Since 2012, Tim Holman has been collecting links to “useless” webs site. Over a decade on now, he has produced, as he calls it …
… a hub for all things quirky and weird on the internet.
This includes sites that are just weird-as-heck animations, like “Pug In a Rug” (seen above), as well as oddball creative tools, like “Paint With Text”, which lets you draw images using a string of prose taken from Alice in Wonderland. I used it to paint this …
You can read through his archives sequentially; he’s done Q&As with many of the creators of these useless sites, and they’re super interesting.
But for me, the best part of Holman’s collection?
It’s his randomizer — if you go to “Take Me To A Useless Website” it will send you to one selected at random. I clicked it a couple of times and was sent to this online memory game, a maze generator, a collection of optical illusions, and the Checkbox Olympics.
(Thanks to Sebastian for pointing me to this one!)
7) 🎶 A hand-built wooden pipe organ
In 1992, Matthias Wandel decided to build a DIY pipe organ out of wood.
In this awesome blog post he breaks down the major engineering challenges. For example, he had to figure out how to generate air pressure, so he used the motor from a vacuum cleaner — only to discover it was way too loud and pushed air too forcefully, so the notes deviated from the fundamental. (The solution: Muffle the motor in a foam box, and reduce the voltage.)
Tuning the pipes without having convenient access to a woodworking shop was also more difficult. They way I had originally tuned them was to tune my Commodore 64 to the pitch the pipe was playing, then work out the frequency I had tuned the computer to, calculate the corresponding reduction in length of the pipe, marked it on the pipe, and then took it to the shop to cut the pipe off on a band saw. This worked for tuning up. For tuning down, I could place a small block of wood in the end of the pipe to reduce the pitch. For minor increases in pitch, I also ended up flaring the end of the pipe with a knife.
I also ended up finally replacing the vacuum cleaner motor with something more elegant. It turns out that running at about 40 volts for extended periods was not good for the vacuum cleaner motor. Carbon deposited between the copper contacts on the commutator, and then conducted electricity and started to smolder. My theory was that the motor's design intended for centrifugal force to clear the carbon dust from the commutator, but at the low speeds at which I was running it, this didn't happen.
Lots of pictures, and here’s a video of him playing it. It sounds like a really old organ — which makes sense, since the pipes are wooden!
8) 📱 “Technofossils”
A couple of scientists are trying to figure out what sort of fossils will be left behind by our modern industrialized society.
One big one? Aluminum cans, as they tell the Guardian …
Pure metals are exceptionally rare in the geological record, as they readily react to form new minerals, but the cans will leave a distinct impression.
“They’re going to be around in the strata for a long time and eventually you would expect little gardens of clay minerals growing in the space where the can was. It’s going to be a distinctive, new kind of fossil,” says the geologist Prof Jan Zalasiewicz, a leading proponent of the Anthropocene as a new geological epoch that reflects the impact of modern humanity on the planet, who with Gabbott has written a book on technofossils, Discarded.
Another? Chickens!
Bones are well known as fossils, but while those of modern broiler chickens are fragile – they are bred to live fast, dying fat and young – the sheer volume will ensure many survive into the geological record.
At any moment, there are about 25 billion live chickens in the world, vastly more than the world’s most abundant wild bird, say Gabbott and Zalasiewicz, making them likely to be the most abundant bird in all of Earth’s history. The sudden appearance of vast numbers of a monstrous bird five times bigger than its wild forebear will certainly strike future palaeontologists.
And then there’s synthetic clothing, which won’t decompose the way that historic clothing — made from organic materials — has:
“We are making them in ridiculous amounts,” says Gabbott – about 100bn garments a year, double the number 20 years ago. “People would be surprised just how many clothes are actually out there in the environment as well. I work to clean rivers in the city of Leicester and about a quarter of the stuff that we take out is clothing. We also stick them into landfills, which are like giant mummification tombs.” As the geologists say in their book: “It is already clear that much of modern fashion will end up being, in the deepest possible sense, truly timeless.”
I’m gonna order Discarded now.
9) 🎮 Super Nintendo systems are speeding up as they age
Recently, Alan Cecil — a guy who does “speedruns” on old Super Nintendo games (i.e. he tries to complete them as quickly as possible) — documented something fascinating: Parts of the system seem to have increased in speed over time.
The SNES includes an audio-processing unit, the SPC700, and according to its original specs, it’s supposed to run at 32,000hz. But over time in the 2000s, SNES hackers have been noticing that the unit seems to be running faster.
Cecil posted on Bluesky about this, asking SNES owners to check the speed on their SPC700s. Over 140 did, and sure enough, they were running much faster — one was going at 32,182hz.
Why? No-one’s sure, though the component has a ceramic resonator that can be affected by heat and other changes in its condition. Clearly age tweaks it in some fashion too?
What unsettles the speedrunning world is the possibility that this acceleration could be artificially making speedruns go more quickly. As Emanuel Maiberg writes for 404 Media …
In theory, if the SPC700 is running faster, it would deliver audio data to the CPU faster, and this could impact how a game runs. Let’s say you’re playing Super Metroid and you hit one of those many room-to-room transitions where you shoot to open a door, go through the door, and then the entire screen fades to black and pans over to the next room. Part of what is happening there is that the SNES is loading the data for that next room, including audio data. If the SPC700 is running faster, that data would load every so slightly faster, meaning overall the game would take less time to complete because you’re spending less time on those transitions.
Cecil himself doesn’t think the hardware speedup is enough to affect a human speedrun. It might affect a “tool-assisted speedrun”, where a software robot plays the game with inhumanly fast reflexes. But even there he’s not yet sure — he’s collecting more data on it.
Trippy nonetheless. (And thanks to Bret for pointing this one out to me!)
10) 🤏 A game where you pretend to use your phone
Pippin Barr produces video games/apps that riff wittily on the nature of digital life; back in Linkfest #12 I posted about his hilarious updating of Pong.
Now he’s back with a brilliant new web thingy: “It Is As If You Were On Your Phone.” You can play it using this link here, though it only works on a smartphone, not a laptop.
Basically, the app asks you to engage in potemkin mimickry of phone-use. For example, at one point it asks you to tap the screen in pattern that resembles typing, but in which you’re not actually typing anything. If someone were to observe me playing the game they’d think, oh, look Clive is texting or scrolling through a news site or something, but in reality I’d be just going through these random, meaningless motions. (Screenshot above.)
It’s super weird and I love it. As Barr writes in his blog post about the game …
what if we had an application on our phone that allowed us to seem to be on our phone, to go through those reassuring motions, to know what to do, to appear 100% like a human on their phone, but without having to actually be on our phone an exposed to the direness of the news, the panic of dating, the shitpile of social media, the emptiness of online video, the timesuck of games? A kind of contentless experience. For the win!
That’s the underlying speculative but also totally honest motivation behind this particular game. I’m making it because I think it’s legitimately something people might use and find helpful and because it is fundamentally funny that that is a possible design goal. To me it’s both a piece of comedy and a piece of truth and I can’t tell which is more important or if they’re even distinct. (And I like that.)
11) 🌥 The “aerobiome”
It turns out clouds contain so many thriving microbes that scientists now refer to it as a separate region of life — the “aerobiome”.
Smithsonian has a terrific excerpt from Carl Zimmer’s new book Air-Borne, which describes how the scientist Pierre Amato collected samples from cloudtop mountains …
Clouds, Amato demonstrated, are alive. Every teaspoon of mist floating over Puy de Dôme contains several thousand microbes. While many are dead airborne husks, some are still alive. They make new proteins and destroy old ones. They grow in the clouds and even divide in two. Their DNA has revealed that some belong to familiar species, but many are new to science … a single cloud, by his estimation, can contain thousands of species.
It also appears that cloud-borne bacterial lifeforms are crucial to making it rain and snow …
Even at temperatures far below the freezing point, water molecules can remain liquid. A seed of impurity is required. As water molecules stick to its surface, they bond to one another. Other water molecules then lock onto them and assemble into a crystal structure. Scientists have found that fungi, algae, pollen, lichens, insects, bacteria and viruses are especially good at encouraging water to freeze. The life that floats in clouds seeds much of the rain and snow that falls back to Earth.
Thinking about aerobiomes could also help us search for extraterrestrial life. The surface of Venus is probably too hot for life as we know it, but the clouds 30 miles up “have the same temperature and pressure as clouds on Earth” and thus could have all manner of microbial life.
I’m ordering Zimmer’s book now!
12) ⚡️ An argument for cross-border energy trading
Countries don’t sell trade energy across borders very much: In 2023, only 2.3% of energy was bought from a national neighbor.
Why? Partly because politicians are worried about being too reliant on another country for something as crucial as energy; in Europe, they’ve seen how Russia used sales of natural gas as a geopolitical weapon.
But a pair of short pieces in the Economist argue that countries would be far better off if they traded lots of power with peaceful neighbors. Historically it has meant cheaper energy that’s also more reliable — and greener …
When countries can tap imports at times of peak demand or reduced domestic generation, they can avoid building largely redundant and therefore costly spare capacity. Cables under the English Channel tend to carry power from Britain to France in the morning, since French consumers take more time over breakfast than Brits, but the other way in the afternoon, as the British turn on their kettles to make cups of tea. The one-hour time difference between the countries also spaces out peaks in demand.
Trading power can also help countries decarbonise. Neither wind nor sunshine is constant, so grids with lots of solar or wind power tend to suffer from too little generation on cloudy or still days and too much when the sun is blazing or the wind howling. Cables to other countries relieve both surfeits and shortages. Exporting countries can sell power that would otherwise go to waste and importing countries get cheap, clean energy. All told, EU regulators estimated the benefits of integrating their grids at €34bn ($35.5bn) a year in 2021. [snip]
Investors are energised, too. Britain’s first interconnector to be built under the current regulatory regime, to Belgium in 2019, was so profitable in its first five years that it hit a legally binding cap, and has had to return £185m to consumers.
This is, also, a general point about renewables: If you can stitch together several far-flung wind projects together, they can buttress each other …
Grids with lots of solar or wind power see big fluctuations in generation and prices, depending on the weather. If power can be exported when it’s abundant, instead of being wasted, investment in renewables becomes more attractive. If the wind dies, power can come from far off, where it is still blowing.
I hadn’t thought about this question of cross-border power trade as a component of decarbonization, but it’s damn interesting.
13) 🏈 Fixing the social-sciences replication crisis with “replay review”
Social science publishing has been, for a long while now, grappling with a “replication crisis”. To wit: A scientist is most likely to get published — and heavily cited — if they have a surprising, counterintutive finding. But an unsettling number of these surprising findings have turned out to be wrong, produced by statistical juking, error, and sometimes outright fraud.
What to do? As Andrew Gelman and Andrew King note in this essay in the Chronicle of Higher Education, scholars often say two things: 1) journals should be more skeptical when publishing findings in the first place, and 2) scholars should do more “replication” of existing research to check it out.
The problem with 2) is that replication is expensive. There’s no way to pay to replicate thousands of positive findings every year.
And the thing about 1) is that top journals are already pretty picky. They reject 90 percent of their submissions! And of the few that go on to become published, “most are lightly read and cited and thus have little influence.” Only a very small number become big hits.
Ah, but that latter point is interesting!
Since most published papers are ignored, they’re not where the problem lies. The real danger is in the tiny chunk of social-science papers that explode in popularity.
That’s only about 162 papers a year, as Gelman and King figure.
So they propose a fascinating idea: We should wait and see which papers begin to amass influence, and then pick those for a new, secondary level of scrutiny. Don’t replicate everything. Just replicate the stuff that’s getting really popular, to verify whether it holds up.
They compare it to a sports “replay” analysis …
publications that go on to have an outsized impact would be evaluated again, and in more detail, to confirm or refine the initial assessment. As in sports, this process could be highly effective without undue disruption or cost. [snip]
We contend that replay review provides an effective and efficient way to correct the published record. Here’s how it would work. Once a publication receives a specified number of citations, it would receive an independent review. These reviews would then be published in full, along with author responses, so that readers have additional guidance on how to interpret the initial publication. As with “booth review” in sports, replay review in science should include analysis from multiple angles, but also a clear assessment of how the “play” should be called, and whether the original conclusions were justified or not.
I’m really interested in this idea. As a practical matter, I don’t know how you’d implement it. I guess you’d need all the top social-science journals to agree to this replay regimen, which wouldn’t be easy. But it’s the first serious proposal I’ve read about how to grapple with the challenge of figuring out which buzzy social-science findings truly hold up, and which are bogus.
14) 📸 Crawling through the dusty archives of Flickr
tvwishes is the Instagram account of Evie, a college student in Missouri whose past-time is crawling through Flickr to find old photos that have some historically magnetic quality. They’re often very uneventful: A photo of a teddy bear alone on a bed, a shelf-full of athletic trophies, a cat sleeping on a CRT television.
One thing she sometimes finds? A photo that no-one else has ever viewed, as she notes in this terrific Q&A …
Flickr tells you how many views a photo has and the likes. I'd come across these photos that no one had seen before. And it felt really special to be viewing a piece of this person's life online from almost 20 years ago. Here I sit looking at someone else's memories that oddly feel like my own. Here I am in 2025 looking at a photo from 2005 that looks like it could have been taken yesterday.
Evie is drawn — and, she says, many folks her age are also drawn — to the low-fi quality of older Flickr images. Those pictures have a definite vintage feel to them, an artifact not only of the lower-resolution cameras and glitchier compression algorithms, but of older forms of behavior: They were taken before Instagram and TikTok trained us to incessantly stage our snapshots …
I’m really drawn toward photos that kind of feel like they maybe were taken by accident or that were truly candid, because that’s so close to what a real memory feels like. Especially the photos that are slightly blurry and it’s turned up to a ceiling. You’re not really sure what you're looking at, but it’s oddly beautiful. When we talk about ephemera, we’re usually talking about physical things, but there’s something about some of these photos on Flickr that are like digital ephemera.
I’ve noticed this generational shift too: My 17-year-old and his friends have recently been buying old mid-2000s Canon cybershot cameras — they really dig those aesthetics.
15) ☎️ A website you update by calling +1 (715) 999-7483
This one is super weird, folks!
Brian Moore is an technology artist, and I just stumbled across a project of his which he describes thusly …
715-999-7483 is a phone-powered multiplayer website builder. By calling the phone number, anyone at any time can update the homepage by describing the changes they'd like to make to it. What happens when you give the public the power to change one central website? Will they use the power for good, for stupidity, and will they wait on hold to use it?
I called the number just now. I was told there were eight people ahead of me, and that it’d be an eight minute wait, so I was put on hold. After about eight minutes precisely (nice estimate!) it was my turn.
So I left this message …
“I would like the main image on the site to be a large picture of a pomeranian dog.”
I refreshed the page a minute later and, le voila: There was the pooch. (Screenshot above.)
How is Moore doing this? Judging by the AI-generated style of that dog I asked for, he’s doing voice-recognition on each phone message, then sending the requests to one of the major large language models and asking for it to generate the appropriate HTML and, where required, an AI image.
Give it a whirl, it’s pretty nutty.
16) 🍸 A final, sudden-death round of reading material
Ice batteries. 🍸 Steganographic encryption of data inside audio of cats purring. 🍸 Time reflections. 🍸 “I left private equity to work on shrimp welfare.” 🍸 F/F fan fiction is on the rise at Archive Of Our Own. 🍸 A wooden jacket. 🍸 FreeTube. 🍸 Dataviz of UFO sightings collected by the US government. 🍸 Geothermal’s “Captain Bubbles”. 🍸 “Terminating pandemics with smartwatches”. 🍸 Excellent history of misophonia. 🍸 DIY paper tape reader. 🍸 Mathematical model suggests that plants lie to each other, and eavesdrop. 🍸 Mondrian-style CSS toggle shifters. 🍸 Saturn has 128 more moons than we thought. 🍸 Claude analyzes 27-year-old Visual Basic EXE and reimplements it in Python. 🍸 “Just give me a normal door handle.” 🍸 Anime 4chan comment provokes 14-year-long mathematical quest. 🍸 Exceptionally deep reading. 🍸 Carving a pumpkin with a waterjet. 🍸 NASA uses GPS on the moon. 🍸 When Jimmy Page played a prom in Ohio. 🍸 3D-printable VHS-rewinder drill attachment. 🍸 "Kummerspeck", "aspaldiko", "ageotori". 🍸 Free95, an open-source operating system that runs Windows 95 programs. 🍸 Commodore PET emulator. 🍸 Wind-turbine blades made from wood. 🍸 Open-source random-number generator using avalanche noise. 🍸 Poetry, DNA and LLMs. 🍸 On REM’s unusual career. 🍸 The first electric cars were for women. 🍸 StickTok. 🍸 Estimating the percentage of medieval manuscript-copiers who were women. 🍸 Electric car with a roof-mounted drone launcher. 🍸 Looks like those Majorana particles don’t check out. 🍸 Xenobots. 🍸 Nigerian mini-grids. 🍸 Russian poem from 1901 about electricity. Virtual punch-card creator. 🍸 The importance of repetition in the workplace. 🍸 How a taco truck boosted the safety of an LA street. 🍸 The Capitanian extinction. 🍸 Coffee reduces risk of Type 2 diabetes. 🍸 “Tommy John” surgeries in baseball are up. 🍸 Planetary-gears fidget spinner. 🍸 Scottish magnet fishing. 🍸 Thermorphs.
CODA ON SOURCING: I read a ton of blogs and sites every week to find this material. A few I relied on this week include Strange Company, Hackaday, Andrew Drucker’s Interesting Links, Numlock News, the Awesomer, the Morning News, and “When The Going Gets Weird”; check ‘em out!