Yesterday, Seamus noted here the remarkable act of policing in Toronto this week: A man had just killed ten people, and injured many more, by driving into them with a van -- and he was apprehended by a single police officer who never fired a shot.

Today, the New York Times has a detailed, step-by-step reconstruction of how Toronto constable Ken Lam did this. It turns out he executed, nearly perfectly, the de-escalation techniques that Toronto police training has in recent years emphasized -- training designed specifically to prevent the jittery, screaming, confrontational police encounters we've seen too often in the US, which end in dead-and-unarmed suspects.

As the Times describes it:

After Mr. Minassian stopped the van, he was encountered by Constable Lam. Video recordings of that high-stress confrontation, captured by a number of bystanders and stitched together by The New York Times, tell a story that can be dissected, step by step.

First, Constable Lam turned off the siren blaring from his car. This immediately lowered the temperature, experts said, making it easier for him to communicate with the suspect. Also, by leaning into the car, the officer is indicating that he is not in a rush.

“It is about slowing things down, using time and distance to de-escalate the situation,” said Chuck Wexler, the executive director of the Washington-based Police Executive Research Forum, which develops best policing practices. “The most important thing at a time like this is time and communications.”

Constable Lam continues to shout, loudly but calmly, “Get down.” The suspect replies, “Kill me.”

“No, get down,” Constable Lam repeats.

A fun little tale of interspecies aid here at Youtube:

We spent our Holidays at the red sea. While walking along a lonely beach we saw a stranded Octopus in the sand. We were not sure if he is already dead. So we pushed him back into the water. He needed some minutes to recover, then he swam away.

Next day we went to the same beach for a walk. While Walking along the beach we saw a shadow in the water coming very quickly to us. It was our Octopus we rescued yesterday! He recognized us! He accompanied whith us a long time while we walked along the beach, all the time tried to touch our feet.

We are sure that this Octopus came back to thank us for saving his life. It's amazing how intelligent animals are.

I can't be sure of former point here -- it's hard to know what an octopus is thinking about, so it's tricky to say it was thanking them -- but the latter point is indisputably true: Octopuses are mad smart and curious.

(Thanks to Harry Allen for finding this one!) Read the rest

Apple has unveiled Daisy, a robot that can disassemble nine different types of iPhones, as part of their program to try and recover and recycle more materials that go into their devices. It can apparently handle 200 iPhones an hour.

It's pretty mesmerizing to watch at work, I must say. As Eyewitness News writes:

According to the Apple report, Daisy can recover 1,900 kilograms of aluminum (used in enclosures), 770kg of cobalt (a material crucial for making batteries), 710kg of copper (crucial in circuit boards) and 11kg of rare earth elements (which play a role in controlling magnets, and also are used in the cameras and haptics devices). That’s just a partial list.

The company plans to install Daisy in multiple locations, starting in the United States (she’s already in Austin, Texas) and Europe. Here’s the list of products the system can handle: iPhone 5; iPhone 5s; iPhone SE; iPhone 6; iPhone 6 Plus; iPhone 6s; iPhone 6s; iPhone 7; and iPhone 7 Plus.

And here’s the caveat: Right now, Apple can only work with end-of-life devices that are returned to it directly. So the company is putting considerable muscle behind the Apple GiveBack trade-in and recycling program. That includes offering credit toward a future purchase.

I'm glad Apple's working on upping its recycling game, but they could have an even bigger impact by better designing their devices to be repaired by third parties, or even customers themselves, in the first place.

As the folks at iFixit have discovered during their wonderfully forensic teardowns of high-tech devices, Apple products can sometimes be a bear to unbuild and fix. Read the rest

Begun, the deepfake wars have.

As usage grows of FakeApp -- the software that makes it comparatively easy to create "deepfaked" face-swapped videos -- a couple of researchers have decided to fight fire with fire. So they trained a deep-learning neural net on tons of examples of deepfaked videos, and produced a model that's better than any previous automated technique at spotting hoaxery. (Their paper documenting the work is here.)

This is good, obviously, though as you might imagine the very techniques they're using here could themselves be employed to produce better deepfakes. Technology!

As MIT Tech Review reports ...

The results are impressive. XceptionNet clearly outperforms other techniques in spotting videos that have been manipulated, even when the videos have been compressed, which makes the task significantly harder. “We set a strong baseline of results for detecting a facial manipulation with modern deep-learning architectures,” say Rossler and co.

That should make it easier to spot forged videos as they are uploaded to the web. But the team is well aware of the cat-and-mouse nature of forgery detection: as soon as a new detection technique emerges, the race begins to find a way to fool it.

Rossler and co have a natural head start since they developed XceptionNet. So they use it to spot the telltale signs that a video has been manipulated and then use this information to refine the forgery, making it even harder to detect.

It turns out that this process improves the visual quality of the forgery but does not have much effect on XceptionNet’s ability to detect it.

Get Out director Jordan Peele and Buzzfeed CEO Jonah Peretti got together to make ObamaPeele -- a faked video of Obama, using Peele's voice.

It's a PSA aimed at getting people to be more wary of possibly-faked videos online. They created this one using FakeApp, a relatively easy-to-use iteration of "DeepFakes" technology that lets you swap faces in videos (previously here, here and here). Thus far, it's most infamously been used to put the faces of famous actors into porn videos, but it's painfully easy to see how this will get used in producing political disinfo in the months and years to come, too.

So, in addition to shooting the video, Buzzfeed composed a five-step guide offering pointers on how to spot a deepfaked video. It's a great list, including suggestions like these:

3. Check where else it is (and isn’t) online Do an online search with the video’s title or other relevant keywords to see if multiple credible news outlets and the person or entity in question have weighed in on it. Again, no need to rush to a conclusion until you can consult multiple courses.

4. Inspect the mouth Farid said that today’s deepfake videos often have telltale signs. One thing to look closely at is the mouth of the person speaking, because these tools often struggle “to accurately render the teeth, tongue, and mouth interior. So, keep a watchful eye on the mouth for [visual anomalies].”

5. Slow it down Farid also encourages people to slow down and freeze parts of the video to watch it more closely.

Is it possible that modern humans aren't the first civilization on Earth?

This is the insanely interesting question probed by "The Silurian Hypothesis", a new paper authored by Gavin A. Schmidt and Adam Frank, two NASA scientists.

As they point out, if an industrialized civilization existed in the deep past, it's not clear there'd be easily recognizable traces of it. Our geologic record doesn't go back any further than the Quaternary period of about 2.6 million years ago. "Go back much farther than the Quaternary," as Frank writes in an essay about the paper in the Atlantic, "and everything has been turned over and crushed to dust."

It's not even clear we'd find fossilized remains of a previous civilization, because while museumgoers might think that fossils are reasonably common, they're actually incredibly rare. A near-zero percentage of life on earth has ever been fossilized. A civilization could last what seems -- to us -- like a super-long time and still not produce any fossils, as Frank notes:

So, could researchers find clear evidence that an ancient species built a relatively short-lived industrial civilization long before our own? Perhaps, for example, some early mammal rose briefly to civilization building during the Paleocene epoch about 60 million years ago. There are fossils, of course. But the fraction of life that gets fossilized is always minuscule and varies a lot depending on time and habitat. It would be easy, therefore, to miss an industrial civilization that only lasted 100,000 years—which would be 500 times longer than our industrial civilization has made it so far.

Stéphane Pigeon has created Purrli, a web site that generates audio of a cat purring.

It's customizable; I found setting it for "sleepy" and "relaxed" produced my particular fave timbre of cat-purr.

As Pigeon notes:

The sound of a purring cat is one of the most comforting sounds available and can help soothe and calm you down when you're feeling stressed. Naturally, it's not just the sound that is important, but it's also the presence of the warm cuddly cat. Purrli tries to recreate both the sound and the presence of your very own virtual cat through a custom sound engine modelled after real purrs.

With a purr that delicately changes over time, Purrli aims at making the experience as real and lively as possible. Just like a real cat, Purrli will call for your attention. Just be careful when adjusting the last slider, if you don't want to be nagged in the middle of your work.

Scroll down the page and read the testimonials -- it's quite interesting to see the varied reasons people enjoy hearing a virtual cat purr, including:

My cat that grew up with me during childhood died two or three years ago, and whenever I was upset, she would come and lay next to me and purr to calm me down. She would nap with me, and her purring would help put me to sleep. I really miss that. This cat purr generator sounds just like her, and it really helps with my anxiety, especially during large projects.

Behold the Mary River Turtle (previously!), which has, alas, recently joined the endangered-species list maintained by the Zoological Society of London.

Apart from sporting a totally rad green mohawk, the turtle breathes through its genitals. As CNN reports:

The Mary River turtle, native to Queensland, Australia, has the unusual ability to breathe underwater through specialized glands in its cloaca -- a posterior opening for excretion and reproduction.

This biological function allows the turtle -- referred to as a "butt breather" -- to stay underwater for up to three days. That ability also usually provides these turtles with a vibrant green mohawk, the result of algae growing on their heads because of the extended time spent submerged.

Rikki Gumbs, a reptile biologist at Zoological Society London (ZSL), told CNN that because of the exotic pet trade in the 1960s and '70s, the turtles were often kept as pets and were already at risk of being endangered when they were first recognized as a species in the 1990s.

"The turtle takes a long time to reach sexual maturity, taking up to 25 to 30 years," he said. "As their vulnerability was discovered late, we lost a whole generation due to the pet trade and now their population has become very small."

The one fragile upside here is that because the turtle is both endangered and incredibly weird, it has some utility in calling attention to the biodiversity that human activity is eradicating from the planet.

Such is the conclusion of the Zoological Society of London, which has for years published lists of endangered species -- but this year began publishing lists specifically of "evolutionarily distinct" species that are in danger of vanishing. Read the rest

"Forced rhubarb" is rhubarb grown to maturity in complete darkness -- during which it grows so rapidly it produces incredibly cool, weird sounds.

If you listen to that video above you'll get an earful: Rapidfire clicks and pops that sound like somebody irregularly whacking a couple of wet sticks together.

The process of "forced rhubarb" -- a term I am using for my next band name obviously -- is outlined nicely in this story on Atlas Obscura:

The method of growing forced rhubarb dates back to the early 1800s, and continues in much the same way today. Farmers let the rhubarb grow out in the open for two years, as the roots collect and store calories. Then the plants are transplanted to lightless growing sheds around November, where they continue to grow—warm, but out of season and in the dark. The rhubarb grows without photosynthesis, which normally makes the plant tough and fibrous. “You get a lot more tender, less tart rhubarb. There’s not too many strings. Outside rhubarb is quite stringy. When you’re cooking with [forced rhubarb] you use around 40 percent less sugar,” says French. The process also results in deep, red stalks, without the normal green shading.

As the stalks burst up out of their initial buds, they create a distinct popping sound, and as they get larger, the stalks rub together and create squeaks and creaks. “It’s growing over an inch a day. It’s not like your field or garden where things are growing two to three feet apart,” says French.

Behold a photo of a massive lava bubble, taken by the United States Geological Survey in 1969.

It's from the long eruption of the Kīlauea volcano in Hawaii, which began on May 24, 1969 and went on for a remarkable 1,774 days.

That bubble above is about 65 feet high, but apparently other bubbles were as huge as 246 feet!

As Smithsonian, via Live Science, reports:

Lava fountains erupt either from isolated vents in lava lakes, or from lava tubes that are penetrated by water, according to the USGS. The formation and expansion of gas bubbles in molten rock pushes powerful streams of lava into the air—typically in a haphazard fashion, with the fountains spurting every which way. It is rare, the USGS notes, for the fountain to take the shape of a dome, like the one seen at Mauna Ulu.

In a 1979 report, the USGS wrote that the dome fountain appeared frequently during the October event, which “lasted for 74 hours, nearly twice as long as any other fountaining episode of the eruption.” The report also notes that the dome had a mottled surface, caused by solidified crust getting mixed with liquid lava. As part of the dome slid away, experts could see a lava core inside, indicating that the dome was “not simply a large bubble.

NASA just released this fascinating and gorgeous video of Jupiter's north pole -- and its wild, swirling storms.

They're not some CGI mockup; these are actual images of the real planet. They were taken by Juno probe that's currently orbiting Jupiter; Juno has an infrared mapping device that's able to probe up to 45 miles below Jupiter's surface, capturing this lush 3D detail.

Jupiter's north pole is dominated by a cluster of cyclones: One large central one, surrounded by eight smaller ones – though on Jupiter, "smaller" means these ones are up to 2,900 miles in size. Jupiter is biiiiig man!

This second video zooms out a bit further and lets you see how the main cyclone and the eight smaller ones are positioned ...

These videos are so trippy, I could stare at them all day.

Figuring out how Jupiter works means figuring out what's happening beneath the surface, so this new phase of infrared imaging is crucial. The planet is still awfully mysterious, because as NASA notes, we're still not sure why Jupiter's atmosphere is divided the way it is:

The map Connerney's team made of the dynamo source region revealed unexpected irregularities, regions of surprising magnetic field intensity, and that Jupiter's magnetic field is more complex in the northern hemisphere than in the southern hemisphere. About halfway between the equator and the north pole lies an area where the magnetic field is intense and positive. It is flanked by areas that are less intense and negative. In the southern hemisphere, however, the magnetic field is consistently negative, becoming more and more intense from the equator to the pole.

There are 500,000 pieces of space junk orbiting the earth, many of which significantly endanger satellites and crewed space missions.

Now there's a fascinating experiment underway to test various techniques of collecting and disposing of space junk: Orbital garbage collection!

A European consortium launched the RemoveDEBRIS satellite last week, and it's already in space ready to be deployed for its tests in May. The satellite will test a couple of ways of capturing space junk, including firing a net around a junk satellite (producing enough drag so the junk begins spiraling towards Earthbound destruction), and -- more dramatically -- firing a harpoon at a target, to test whether space junk could be collected using space harpoons.

To repeat: Space harpoons.

Some more details from Smithsonian's Air and Space magazine:

In the first test, a CubeSat released from the main spacecraft will maneuver to a distance of more than 20 feet, where it will unfurl a balloon (to provide a bigger target). A net similar to the kind used in commercial fishing will then be deployed from the main spacecraft to capture the CubeSat. Atmospheric drag should cause the netted CubeSat to burn up in the atmosphere in a matter of weeks, according to Guglielmo Aglietti, Director of the Surrey Space Centre at the University of Surrey, and the principal investigator for RemoveDEBRIS.

For the second test, another CubeSat will be released to fly at a distance from the main spacecraft, where a camera and LIDAR will observe it to assess how well future trash-collecting satellites could judge the position and speed of a piece of debris.

Almost ten years ago, I interviewed Mike Brickley, an inventor who'd designed a new architecture for an internal combustion engine -- one that he predicted would produce 35 per cent less friction than a typical ICE, and 20 per cent better fuel mileage. (Mark blogged about it for Boing Boing back then too.)

Pretty rad specs, if he could pull it off! As I described the design back then ...

In the Brickley Engine — as he calls it — there are no piston skirts and several fewer crankshaft and crankpin bearings. What’s more, the cylinders are connected through a pinned joint that rotates a comparatively small amount. The resulting device has a curious, flattened appearance: two sets of pistons face away from each other and punch in opposite directions, joining in the center to drive the engine’s shaft.

At any rate, ten years later Brickley has a prototype built, and he's put some videos of it online. That gif above is from this video showing the engine's guts, and the physics of the newfangled pistons; the one below shows it in action ...

Brickley told me that preliminary tests show his predicted efficiencies seem to have been borne out, and I hope he publishes some data on that; it'd be interesting to see where he can take this. Read the rest

Back in 2011, Sara Hendren -- a design and disability-studies researcher -- teamed up with the graffitti artist and philosophy professor Brian Glenney to redesign the International Symbol of Access. Often known as the "Wheelchair Symbol", it had been around for almost 50 years, but Glenney and Hendren thought the symbol -- a blocky, immobile figure -- was too passive.

So they designed the icon you see above, one that makes the figure much more active. Then they engaged in a street-art campaign, printing up 1,000 of the icons as transparent stickers that were pasted onto old-school Wheelchair-Symbol signs around Boston. Since you could still see the old sign through the transparent overlay, Glenney and Hendren's goal was to make passersby think about the meaning of that old symbol.

The new "Accessible Icon" -- as it's been called -- grew so famous that it's been informally adopted in locations around the world, employed by a US Department of Treasury sign, and included in MOMA's permanent collection. (The final version, above, was tweaked by graphic designer Tim Ferguson Sauder to make the icon comply with professional standards.)

As it turns out, though, people with disabilities disagree over whether the new symbol is better than the old one. Atlas Obscura has a fascinating piece outlining the various views:

Mike Mort, who runs the blog Disabled Identity, also favors the new icon. “I don’t mind the older symbol,” he says, “but I definitely think this is a step, roll if you will, in the right direction.

Stanley Kubrick had a lot of trouble getting the right voice for HAL in 2001: A Space Odyssey. He initially used the Oscar-winning actor Martin Balsam, but his Bronx accent wound up making HAL sound "a little bit too colloquially American," as Kubrick later decided.

So Kubrick turned instead to Douglas Rain, a Canadian stage actor who was a veteran of dozens of plays. It worked -- Kubrick got the weird, calm, eerie affect that made HAL so unsettling.

Part of what made things work? Apparently it was Rain's Canadian tonalities:

Kubrick was attracted to Mr. Rain for the role partly because the actor “had the kind of bland mid-Atlantic accent we felt was right for the part,” he said in the 1969 interview with Mr. Gelmis. But Mr. Rain’s accent isn’t mid-Atlantic at all; it’s Standard Canadian English.

As the University of Toronto linguistics professor Jack Chambers explained: “You have to have a computer that sounds like he’s from nowhere, or, rather, from no specific place. Standard Canadian English sounds ‘normal’ — that’s why Canadians are well received in the United States as anchormen and reporters, because the vowels don’t give away the region they come from.”

As a Canadian myself, this tickles me to no end.

There are plenteous other trippy details in that piece, including the fact that since HAL's performance was recorded after all the other scenes were in the can, Kubrick got Stefanie Powers to read HAL's lines during the shoots! And then this ... Read the rest

Now that is not a headline I thought I'd type when I got up this morning.

Nonetheless, science! So, here are the grisly details: Black-footed and Laysan albatrosses have thrived for decades on Midway Atoll in the north Pacific. Then in 2015, scientists began noticing birds with open, bloody wounds on their heads, necks and backs. What was going on? Were predatory owls or hawks attacking them?

Nope. Video footage uncovered the culprits – common house mice that were climbing onto the albatrosses and "eating them alive."

House mice aren't native to the island, but they'd been introduced 75 years ago and had coexisted without incident. The scientists eventually figured out what was going on: A drought on the island had made the mice desperately thirsty.

As the Washington Post reports:

"The mice were looking for sources of liquid, moisture, so actually drinking blood,” Keitt said.

He added that it is possible the attacks spread because mice learned from one another, discovering a new way to survive. Mice are omnivores, meaning they will eat any source of food they can find.

But how is it possible that large seabirds — standing almost three feet tall with a wingspan of more than six feet — are falling prey to a rodent?

Nesting albatrosses are particularly vulnerable to attacks because of an intense biological instinct to guard their young, Keitt said.

“The risk that they know is to their egg or chick that they’re sitting on,” he said. “They want to stay there and protect it.

A 19-year-old named Paul Williams discovered someone named "LisaJames419419" had started following him on Instagram. He idly checked her account and found ... she was following dozens and dozens of other people named Paul Williams. And only following people named Paul Williams! (Or ones with that name as a stem, as with "Williamsen".)

A bot, right? Except the original Williams messaged "LisaJames419419" and the account blocked him a few minutes later, which seems like unbotlike, human behavior.

Buzzfeed reports on the story and the theories that are now raging as to what the heck is going on, including:

looking for the father of her child looking for a long-lost relative stalking someone FBI agent "Lisa is the Terminator sent back in time to kill all Paul Williams(es)."

The actual photo used in the "LisaJames419419" account is the adult film star Briana Lee, "whose photos are widely used for catfishing and scamming online." Read the rest