80,000 cameras pointed at highways and parking lots

September 6th, 2025

Since its founding in 2017, Flock, which was valued at $7.5 billion in its most recent funding round, has quietly built a network of more than 80,000 cameras pointed at highways and parking lots across the U.S.:

Growth has been explosive, with revenue up some 70% from the estimated $175 million it booked in 2023. It’s not yet profitable and has no imminent plan to be as it prioritizes growth, backed by a $275 million March funding round led by Andreessen Horowitz. Those numbers were more than sufficient to land Flock on Forbes’ 2025 Cloud 100 list of the top private cloud computing companies. Langley says turning Flock into a $100 billion business is “very within reach.” Ilya Sukhar, an early investor and partner at VC firm Matrix who sits on Flock’s board, agrees. “It’s a bit cliché, but it does feel like we’re just getting started,” he says. “It’s not hard for me to project to a place where we get to that level.”

Each Flock license plate reader cam costs between $3,000 and $3,500, with an additional fee for FlockOS, the operating system that makes all the data Flock collects accessible via a browser or a mobile app, based on either the number of users or cameras. Dunwoody PD, for instance, pays around $500,000 annually for its array of 105 cameras, gunshot detectors, that skittering DJI drone and the software that controls it all.

Flock’s growth isn’t solely fueled by its 5,000 law enforcement customers across 49 states (it hasn’t yet installed its cameras in Alaska). It has 1,000 corporate customers, including blue chips like FedEx, Lowe’s and Simon Property, America’s largest mall owner. Then there are housing and homeowner associations, small businesses, schools and organizations like the Jewish Federation of Greater Atlanta, which has installed 64 Flock cameras across different properties in the city, including a community center that has reported a recent spike in antisemitic threats to Dunwoody police. All these customers can choose to grant the police access to their camera feeds, further expanding the surveillance coverage Flock can offer law enforcement. Many do.

Langley had no experience in police tech when he and fellow Georgia Tech alums Matt Feury, 36, and Paige Todd, 40, started the company in 2017. Previously they’d worked together on an app Langley cofounded for upgrading sports or concert seats to VIP-status events, where Feury and Todd were early employees. (It was acquired by Atlanta-based conglomerate Cox Enterprises and no longer exists.) Inspired by an unsolved robbery in Langley’s neighborhood, the trio started work on the first Flock prototype, an Android phone camera in a waterproof box that took pictures of cars and picked out license plates that could then be searched via an app.

Face to Face with Death in Mosul

September 5th, 2025

Matt Larsen shares the story of Staff Sergeant Paul McCully finding himself face to face with death in Mosul:

I wasn’t sure what had happened, I just knew I was laid out on my stomach, and I couldn’t feel my hands or legs. I could hear Owens screaming, and I was checking myself to see if I was physically intact when another explosion went off, a hand grenade, but it wasn’t as loud as the first one. I felt the shrapnel impact my helmet but was still in a daze and confused as to what was going on. Then I felt something that seemed to be tapping my helmet and everything sounded muffled. My initial thought was that it was my guys pulling me out of there, but when I looked up, everything came back to me, sound, reality, cleared vision. There was a bad guy standing over me.

I was looking up at him and expecting him to unload his AK47 on me, but he was screaming and butt-stroking me in the head. The second I realized that it wasn’t my guys, I got up as fast as I could and grabbed his AK muzzle with my right hand and his shirt on his right shoulder with my left hand. I don’t even remember placing my hands on the ground to push myself up; it just seemed like I floated up, that’s how fast it happened.

After I grabbed him and his weapon, I was jerking it in an outward motion but making sure to keep the muzzle away from me. After what seemed to be two or three seconds, I got the AK out of his hands and on the ground to the right of me a couple of feet. I had finally jerked it free, and it went flying. He tried to dive for the AK, but I grabbed him and went to the Thai clinch with him to control him. A Thai clinch is when you control a person’s upper body by placing both your hands behind his neck. Our bodies were close together; I had his hair in my right hand, pushing his head down, and my left hand was controlling his left shoulder. I immediately started throwing right uppercuts and knees to [mess] him up. I did that because I thought that there were more of my own guys behind me, but it turns out that Owens and I were the only ones to make it outside before the initial explosion. The No. 3 and No. 4 men got blown back into the building.

After I threw the blows, I held on to him with the shirt and hair and extended my arms to allow the guys who I thought were behind me to have a clear shot. But that never happened. It seemed like I was alone, and nobody was there to help me. He was screaming “Allahu akbar” and I was yelling “Fuck You” and continued to hit him as he was struggling to get to his weapon. Owens came running up to me with his pistol drawn. He had lost his M4 rifle in the blast also, so he pulled his M9 pistol.

He came up to my right side, right next to me so he wouldn’t shoot me in the struggle. Right as he fired one shot into the enemy’s stomach, the enemy had reached up and grabbed Owen’s pistol. At that moment I let go and took a step back and secured my M4. Owens had swung him around to the left, which put him right in front of me.

With Owens and the bad guy fighting for Owen’s M9, I put the barrel of my rifle in the bad guy’s right side, point-blank, right underneath his armpit, and fired a single shot. He squealed like a pig and hit the ground like a sack of shit, landing on his back. I immediately placed the barrel of my rifle in his face and fired ten shots to finish him. All of this happened within a matter of about 20 seconds, but seemed like forever.

The modern battlefield requires split-second decision-making, seamless coordination among distributed teams, and processing vast amounts of information, all under extreme pressure

September 4th, 2025

The modern battlefield requires split-second decision-making, seamless coordination among distributed teams, and processing vast amounts of information, all under extreme pressure:

As I have learned over the last year, as an advisor to August Interactive, a gaming studio, these are exactly the skills that well-designed military gaming programs can develop and refine.

Unlike traditional military wargaming — which typically involves structured, turn-based exercises on maps or models to explore campaign plans and strategic concepts — the gaming discussed here draws heavily on digital interactive platforms, including modified commercial titles and purpose-built military simulations. These environments — ranging from real-time strategy games to tactical shooters, flight simulators, and cyber-themed games — emphasize rapid continuous decision-making, high-pressure coordination, and immersive skill development. While both approaches aim to sharpen judgment and prepare leaders for complex scenarios, this form of military gaming leverages the speed, interactivity, and scale of modern gaming technology to cultivate competencies that are difficult to replicate in traditional wargaming formats. And they are also more engaging and fun, which is a good thing.

The U.S. military should formally embrace and invest in advanced digital gaming as a core training tool, leveraging its ability to build critical cognitive, coordination, and technical skills for modern warfare.

[…]

I visited the gaming center at West Point last spring. I was impressed with the setup and technological capabilities, but I was even more impressed by the insights shared with me by combat-experienced officers and non-commissioned officers overseeing the program. The positive impact on cadet leadership development was remarkable: improved communication skills, quicker decision-making, and faster adaptability to change. Notably, many intercollegiate athletes there are involved in military gaming.

Is modernism due to youth culture?

September 3rd, 2025

Robin Hanson has been puzzling over the transition from traditional to modern culture:

It happened after tech started changing a lot, when long distance trade, travel, and talk greatly increased. But with a big delay; those things had changed lots a few centuries before culture started changing fast ~1900-1920. And strangely, the new modernists were then most sure that the culture of their grandparents was not what they wanted, even though they felt quite unsure of in which new directions culture should go.

Clothing fashion had been changing for several centuries before, and there had also been slowly changing fashion in governance and morals. But suddenly art, sculpture, architecture, music fashion changed much more radically, and soon after norms and values also started changing faster.

At the key transition time, it seems that the culture of elite youth was more modern than the culture of older adults. And even today, most people most like the food, music, etc. popular when they were ~20yo, suggesting youth have a disproportionate role in cultural change. And elites have always had more influence over most everything.

All of these lead me to wonder if a key was the rise of school, which concentrated elite youths together so that they could form their own internal elite youth culture.

Should we have kept the American Empire?

September 2nd, 2025

Should we have kept the American Empire?, Max Tabarrok asks:

Note that the question of whether the US should have relinquished its maximum territorial extent is different than the one facing America in 1865 or the question of expanding American borders today. There, you have to consider the substantial costs of actually conquering territory in war. Holding on to land already conquered is less costly than conquering anew.

[…]

The first thing to notice when considering this question is that America did not actually give up all of its imperial conquests. Hawaii, Puerto Rico, the Virgin Islands, a smattering of pacific islands, Alaska, and arguably even much of the American Southwest are spoils of conquest or purchase.

No one seriously considers giving up any of the pieces of our imperial history that we kept. The Hawaiian independence movement is perhaps the most popular, but that sets a low upper bound. This acceptance of the pieces of empire we retained suggests that if we had kept more of our 1918 peak territory, people would accept those additional pieces just as easily today.

Reluctance to give up what we kept isn’t just status quo bias. The economic performance of the states and territories still within the US compared to the nations we once controlled but are independent today is evidence that American annexation has large positive effects.

[…]

Integration under the same national umbrella seems like about the only way to sustain and spread free trade and immigration. The US can’t even manage it with Canada. The principal domestic supporters of Philippine independence, for example, were American farmers who didn’t want to face competition from tariff-free Filipino sugar, and nativists who didn’t want immigration from “alien races.”

A final general argument is, perhaps surprisingly, political legitimacy. The nations that America was closest to annexing are not only economic underperformers, they also tend to be undemocratic and authoritarian. Recently, Panama and the Dominican Republic are probably exceptions here with relatively successful and stable governance.

Latest driver death rates highlight dangers of muscle cars

September 1st, 2025

The Insurance Institute for Highway Safety has been calculating driver death rates approximately every three years since 1989, and also calculated the best and worst models according to the number of drivers in other vehicles killed in crashes with them on its last study:

Six of the 21 vehicles with the highest driver death rates for model year 2020 are variants of the Chevrolet Camaro, Dodge Challenger, Dodge Charger and Ford Mustang, while eight others are small cars or minicars. Eighteen of the 23 vehicles with the lowest driver death rates are minivans or SUVs, and 12 are luxury vehicles.

“We typically find that smaller vehicles have high driver death rates because they don’t provide as much protection, especially in crashes with larger, heavier SUVs and pickups,” said IIHS President David Harkey. “The muscle cars on this list highlight that a vehicle’s image and how it is marketed can also contribute to crash risk.”

That might not be how I’d phrase it.

But three Dodge muscle cars with excessively high driver death rates also rank among the worst performers when it comes to other-driver deaths, suggesting these vehicles are driven in an aggressive manner.

Seven of the 20 vehicles with the highest other-driver death rates are large or very large pickups, and four more are midsize SUVs — categories that aren’t represented among the models with the worst track record for protecting their own drivers. Seven of the vehicles with the highest other-driver death rates also rank among the worst for driver death rates: the Dodge Challenger two-wheel-drive, Dodge Charger two-wheel-drive, Dodge Charger HEMI two-wheel-drive, Kia Forte, Kia Optima, Kia Rio sedan and Nissan Altima.

The list of vehicles with the lowest other-driver death rates include two small, two midsize and one large car, as well as six small and 10 midsize SUVs. Ten models are luxury vehicles.

The rates include only driver deaths because all vehicles on the road have drivers, but not all of them have passengers or the same number of passengers. The number of deaths is derived from the federal Fatality Analysis Reporting System. Registration data come from IHS Markit.

The latest rates are based on fatalities that occurred from 2018 to 2021 for vehicles from the 2020 model year, as well as earlier models with the same designs and features. The numbers represent the estimated risks for 2020 models, but the data include models from as far back as 2017 if the vehicles have not been substantially redesigned over the intervening period. Including these older, equivalent vehicles makes the sample size larger and therefore increases the reliability of the results. To be included, a vehicle must have had at least 100,000 registered vehicle years of exposure from 2018 to 2021 or at least 20 deaths.

[…]

The lists of vehicles with the lowest driver and other-driver death rates have nine models in common. These include the Acura MDX four-wheel-drive, Audi Q5 four-wheel-drive, Chevrolet Traverse four-wheel-drive, Lexus RX 350 four-wheel-drive, Mercedes-Benz E-Class sedan four-wheel-drive, Porsche Macan, Subaru Ascent, Toyota C-HR and Volvo XC60 four-wheel-drive. Notably, six of those are luxury vehicles.

“The models that rank among the best and worst performers on both lists point to the unfortunate fact that vehicle cost remains a factor in road safety,” Harkey said.

Vehicle cost — or driver income?

Minicars had the highest driver death rates, averaging 153 deaths per million registered vehicle years. Very large luxury cars had the lowest, averaging only 4 deaths. In contrast, very large pickups had the highest other-driver death rates, averaging 121 deaths, while small sports cars had the fewest other-driver deaths, averaging only 11 per million registered vehicle years.

The average other-driver death rate for all 2020 and equivalent models was 53 deaths per million registered vehicle years. There are more other-driver fatalities than driver fatalities because these newer models are more crashworthy than many of their crash counterparts, which come from the wider U.S. fleet, made up of mostly older vehicles.

Driver Deaths Lowest Rates.jpeg
Driver Deaths Highest Rates.jpeg
(Hat tip to Grasspunk.)

These “farmer tools” greatly simplified Ford’s machining operations

August 31st, 2025

Origins of Efficiency by Brian PotterFord’s status as a large-volume car producer began with the predecessor to the Model T, Brian Potter notes, the Model N, a four-cylinder, two-seater car initially priced at $500:

Many of the Model N’s parts were made of vanadium steel, a strong, lightweight, durable steel alloy. Vanadium steel allowed for a lighter car (the Model N weighed only 1,050 pounds), and was “machined readily.” This was important because Ford also made increasing use of advanced machine tools that allowed it to produce highly accurate interchangeable parts. In 1906, Ford advertised that it was “making 40,000 cylinders, 10,000 engines, 40,000 wheels, 20,000 axles, 10,000 bodies, 10,000 of every part that goes into the car…all exactly alike.” Only by producing interchangeable parts, Ford determined, could the company achieve high production volumes and low prices. Furthermore, Ford’s machine tools were arranged in order of assembly operations rather than by type, allowing parts to move from machine to machine with minimal handling and travel distance. It also made extensive use of production aids such as jigs, fixtures, and templates. These “farmer tools” — so called because they supposedly made it possible for unskilled farmers to do machining work — greatly simplified Ford’s machining operations.

The Model N was so popular that demand exceeded capacity, which allowed Ford to plan production far in advance. This meant Ford could purchase parts and materials in large quantities at better prices and schedule regular deliveries, ensuring a steady, reliable delivery of material, which allowed it to maintain just a 10-day supply of parts on hand.

To build your own drone batteries, you have to source quality cells from a reliable supplier and assemble them into battery packs

August 30th, 2025

If you break open a drone battery, David Hambling notes, you will find a shrink-wrapped block containing smaller batteries:

These cells are described by their size, so an 18650 cell is a cylindrical unit about 18 millimeters in diameter and 65 millimeters in height, while a 2170 is 21mm in diameter and 70 mm high.

A typical laptop battery will contain six 18650 lithium-ion cells. The battery pack for a Tesla Model 3 Long Range made before 2018 contains 2170-type cells, no less than 4,416 of them.

While not all cells are created equal, they are essentially commodity products manufactured by the billion. They’re made mainly by big players in the Far East; China dominates but it does not have a monopoly. Other sources are readily available.

The biggest battery maker by capacity is Chinese outfit CATL, making 132 GWH of cells every year. But the next two are South Korean LG (93 GWH) and Japanese Panasonic (60 GWH), and there are two other Korean outfits, Samsung and SK, in the top ten.

To build your own drone batteries, you have to source quality cells from a reliable supplier and assemble them into battery packs. And that is exactly what Ukrainian drone maker Wild Hornets has been doing for some time.

A video on social media explains Wild Hornets’ process. The building blocks for its battery packs are Samsung 50S, which are optimized for high-power applications and have a respectable 5000 mAH capacity.

The cells are arranged in blocks of 12 in a 6s2p unit (that is, 6 rows of 2 batteries) or 18 in 6s3p (6 rows of 3) configuration. These are connected with metal strips and 0.25 mm copper wiring — “we don’t economize” the presenter says in the video — spot welded into place. Spot welding is costlier than soldering, but more reliable. The completed unit is then securely shrink-wrapped with multiple layers of tough plastic.

[…]

The end result costs a total of $65 for small batteries and $90 for large, similar to commercial drone batteries.

Of course, they’re called batteries because they’re collections of smaller cells:

Benjamin Franklin first used the term “battery” in 1749 when he was doing experiments with electricity using a set of linked Leyden jar capacitors. Franklin grouped a number of the jars into what he described as a “battery”, using the military term for weapons functioning together.

The first breeders unsurprisingly selected for temperament

August 29th, 2025

A couple small mutations helped turn skittish animals into the creatures humans could saddle and ride:

Researchers led by Xuexue Liu and Ludovic Orlando analyzed horse genomes spanning thousands of years, tracking 266 genetic markers tied to traits like behavior, body size, and coat color. Their results, published in Science, suggest that early domestication didn’t begin with flashy coats or taller frames. Instead, the first breeders unsurprisingly selected for temperament.

One of the earliest signals of selection appeared at the ZFPM1 gene, linked in mice to anxiety and stress tolerance. That genetic shift, around 5,000 years ago, may have made horses just a little calmer — tame enough for people to keep close.

But the real game-changer came a few centuries later. Around 4,200 years ago, horses carrying a particular version of the GSDMC gene began to dominate. In humans, variants near this gene are associated with chronic back pain and spinal structure. But for horses and lab mice, the mutation reshapes vertebrae, improves motor coordination, and boosts limb strength. In short, it made horses rideable.

The numbers are staggering. The frequency of the GSDMC variant shot from 1% to nearly 100% in just a few centuries. Laurent Frantz of Ludwig-Maximilians-Universität München, who wrote an accompanying commentary, calls the selection “almost unprecedented in evolution.” For comparison, the human mutation that lets adults digest milk — a trait with huge survival advantages — spread far more slowly, with a selection strength of only 2–6%.

“The right conditions for the rise of the rideable horse materialized ~3,500 years ago in the Eurasian Steppe, north of the Caspian Sea,” Frantz explained. That’s when local cultures began seeking animals for war and transport rather than food. The genetic stars aligned: rare mutations already present in wild horses met human ambition.

It first reacted with the activated amino acid in water and then slowly transferred that amino acid to RNA

August 28th, 2025

Proteins are made of amino acids, which are thought to have been around long before life emerged:

In living organisms today, amino acids combine with RNA to make a protein. But this translation process requires a set of protein enzymes that, paradoxically, are made by protein synthesis. It becomes a chicken or egg problem: How was a protein made without a protein?

[…]

First, the team took an amino acid and “activated it” — basically removing a water molecule, which made it reactive and able to form a bond with other molecules. But the activated amino acid wouldn’t directly bind to RNA in this form. The team needed to find a helper molecule that would aid the amino acid in binding to RNA.

Powner and his colleagues decided to experiment with a class of compounds called thiols, or molecules with a sulfur attached to a carbon. These molecules are better known for their role in energy production and regulation in cells than for protein synthesis, but the team previously found they are fairly easy to make under basic conditions that would have existed on a baby Earth.

When the thiol was introduced, the team found it first reacted with the activated amino acid in water and then slowly transferred that amino acid to RNA. More of these compounds combine and form proteins in cells.

The binding “was very unexpected [and] wasn’t what we set out to achieve,” Powner said. He said this mechanism essentially solves how to initiate protein synthesis without another protein.

“In a scenario where you have amino acids, where you have RNA molecules, if you have thiols — sulfur molecules — this is, I think, almost inevitable that this kind of process can happen,” Powner said.

If you’re not in the meetings, you can’t accurately estimate the relative levels of dishonesty and self-delusion involved

August 27th, 2025

Dominic Cummings discusses preference falsification and Britain’s slide to chaos:

Inside the intelligence services, special forces (themselves under attack from the Cabinet Office and NI Office as they operate as our last line of defence, see below), bits of Whitehall, and those most connected to discussions away from Westminster, there is growing, though still tiny, discussion of Britain’s slide into chaos and the potential for serious violence including what would look like racial/ethnic mob/gang violence, though the regime would obviously try to describe it differently. Part of the reason for the incoherent forcefulness against the white rioters last year from a regime that is in deep-surrender-mode against pro-Holocaust marchers, rape gangs and criminals generally, is a mix of a) aesthetic revulsion in SW1 at the Brexit-voting white north and b) incoherent Whitehall terror of widespread white-English mobs turning political and attracting talented political entrepreneurs. They’re already privately quaking about the growth of Muslim networks. The last thing they want to see is emerging networks that see themselves as both political and driven to consider violence. Parts of the system increasingly fear this could spin out of control into their worst nightmare. In No10 meetings with the Met on riots, I saw for myself a) the weird psychological zone of how much order rests not on actual physical forces but perceptions among a few elites about such forces that can very quickly change, and b) how scared the senior police are at the prospect of crucial psychological spells being broken. We can see on the streets that various forces have already realised the regime will not stop them. What if this spreads? Whitehall’s pathology has pushed it to the brink of this psychological barrier and many of them know it.

Aspects of the situation are tragi-comic. E.g if you talk to senior people in places like UAE, they tell you that bigshots in that region now tell each other — don’t send your kids to be educated in Britain, they’ll come back radical Islamist nutjobs! Our regime has spent thirty years a) destroying border control and sane immigration (including the Home Office’s jihad against the highest skilled, whom they truly loathe discussing and try to repel with stupid fees etc) and b) actively prioritising people from the most barbaric places on earth (hence immigration from the tribal areas most responsible for the grooming/rape gangs keeps rising) and c) funding the spread of those barbaric ideas and defending the organisations spreading them with human rights laws designed to stop the return of totalitarianism in Europe. In parallel, they’ve started propaganda operations with the old media to spread the meme that our ‘real danger’ is the ‘far right’ (code for ‘white people’). As Tories and Labour have continued their deranged trajectory, they have provoked exactly the reactions they most feared including the spreading meme that our regime itself has become our enemy and the growing politicisation of white English nationalism.

These deep state discussions about the growing prospect of violence, like the focus group discussions about ‘civil war’, have seeped through to few MPs or hacks. And the evolution of the Cabinet Office in recent years has excluded ministers, spads and the PM from almost any visibility inside the NSS, the National Security Secretariat of the CO, which has acquired power from the rest of the security/intelligence system and runs a failing empire within a failing empire. When I said in 2020 that, among the general changes to the dysfunctional No10/CO system, the oversight of NSS must change so it became visible and legible again to the PM’s office so we could participate in debates like — what are the actual priorities of the intelligence services vis Putin and Xi — some senior officials tried to pretend that zero political scrutiny of NSS was somehow a constitutional principle. After I left, this system became even more closed and dysfunctional, hence the total lack of true strategic thinking connecting ends-ways-means over Ukraine and all things defence procurement becoming more and more Kafka-esque as the MoD shipped stuff to Ukraine. I repeat: the lack of legibility of the NSS is without historical precedent in the UK for centuries and is related to broader issues of Whitehall’s dysfunction, the disgraceful shambles of the MoD etc.

SW1’s OODA loop has operated for years as a massive denial-of-service-attack on its own perceptions of reality — constant cycles of ephemeral emotional hysteria and Narrative Whiplash while No10 has no capability to execute priorities. A great recent example: Professor Ansell saying that the Zelensky Oval Office interview meant that Farage’s prospects had ‘peaked’ (widely Bluesky’d approvingly!) — an emotional spasm entirely in tune with SW1’s NPC network reflecting OODA-as-DOS-attack. This has, as I’ve argued for years, made it more and more vulnerable to history’s remorseless pattern: slow rot, elite blindness, fast crisis, sudden collapse.

The old parties lost their last chance to fix things in a sort-of normal way when the Trolley and his girlfriend told everyone in 2021 they were going ‘back to normal politics’. SW1 cheered including the Tory MPs who got culled en masse in 2024. Sunak doubled down on optimising for *pats on the head from Permanent Secretaries and lawyers*. After Starmer won, SW1’s NPCs tweeted to each other how they now had ‘serious grownups’ and we’d return to ‘normal government’.

But this was just another cycle of delusional SW1 Narrative Whiplash. The Starmer project blew up on contact with the reality of Whitehall. Now both parties are led by Dead Players. Both old parties are structurally knackered. And the NPCs tweeting ‘hurrah for the grownups, Sue Gray is the Jedi we need’ a few months ago are now Bluesky-ing ‘disgusting rhetoric from Starmer’.

Starmer is speed-running Sunak’s demented combination of a) massively raising the salience of immigration/boats with b) a set of policies that everyone who understands the details knows cannot possibly do what he’s promising.

Why is he doing it? Because, like Sunak, he’s caught between a) political advice that the country is enraged over immigration/boats and wants action, b) the adamantine priority of the dominant faction in Whitehall — i.e the force that actually orients 99% of policy — is maintaining 1) the HRA/ECHR-judicial review system and 2) the cross-party HMT/OBR/university-endorsed immigration/asylum Ponzi. Being a Dead Player optimised to ‘defend the institutions’ at all costs however pathological, Starmer has, aping Sunak, synthesised the political advice of McSweeney and the priority of the officials/lawyers actually running No10/70WH and generated his own version of Sunak’s demented combination.

If you’re not in the meetings, you can’t accurately estimate the relative levels of dishonesty and self-delusion involved.

Portugal does not allow consequence-free drug use

August 26th, 2025

Ever since Portugal enacted drug decriminalization in 2001, reformers have argued that North America should follow suit, but when Oregon and British Columbia decriminalized drugs in the early 2020s, the results were so catastrophic that both jurisdictions quickly reversed course:

Contrary to popular belief, Portugal does not allow consequence-free drug use. While the country treats the possession of illicit drugs for personal use as an administrative offense, it nonetheless summons apprehended drug users to “dissuasion” commissions composed of doctors, social workers, and lawyers. These commissions assess a drug user’s health, consumption habits, and socioeconomic circumstances before using arbitrator-like powers to impose appropriate sanctions.

These sanctions depend on the nature of the offense. In less severe cases, users receive warnings, small fines, or compulsory drug education. Severe or repeat offenders, however, can be banned from visiting certain places or people, or even have their property confiscated. Offenders who fail to comply are subject to wage garnishment.

Throughout the process, users are strongly encouraged to seek voluntary drug treatment, with most penalties waived if they accept. In the first few years after decriminalization, Portugal made significant investments into its national addiction and mental-health infrastructure (e.g., methadone clinics) to ensure that it had sufficient capacity to absorb these patients.

This form of decriminalization is far less radical than its North American proponents assume. In effect, Portugal created an alternative justice system that coercively diverts addicts into rehab instead of jail. That users are not criminally charged does not mean they are not held accountable. Further, the country still criminalizes the public consumption and trafficking of illicit drugs.

[…]

In late 2020, Oregon embarked on its own drug decriminalization experiment, known as Measure 110. Though proponents cited Portugal’s success, unlike the European nation, Oregon failed to establish any substantive coercive mechanisms to divert addicts into treatment. The state merely gave drug users a choice between paying a $100 ticket or calling a health hotline. Because the state imposed no penalty for failing to follow through with either option, drug possession effectively became a consequence-free behavior. Police data from 2022, for example, found that 81 percent of ticketed individuals simply ignored their fines.

Additionally, the state failed to invest in treatment capacity and actually defunded existing drug-use-prevention programs to finance Measure 110’s unused support systems, such as the health hotline.

The results were disastrous. Overdose deaths spiked almost 50 percent between 2021 and 2023. Crime and public drug use became so rampant in Portland that state leaders declared a 90-day fentanyl emergency in early 2024. Facing withering public backlash, Oregon ended its decriminalization experiment in the spring of 2024 after almost four years of failure.

Route march speed was reduced from 7.5 to 5 km/h

August 25th, 2025

In 1991-1992, a pelvic stress fracture incidence of 11.2% was recorded in a cohort of 143 female Australian Army recruits:

An incidence of 0.1% was recorded in a cohort of male recruits trained in the 1992-1993 year using a nearly identical program. A number of preventive strategies were instituted in an attempt to reduce the high incidence of injury in female recruits. Route march speed was reduced from 7.5 to 5 km/h, running occurred on softer surfaces, individual step length was promoted instead of marching in step, march and run formations were more widely spaced, and interval-running training replaced traditional middle-distance runs. Pelvic stress fracture incidence decreased significantly to 0.6% in an immediately subsequent cohort of 161 female recruits (chi 2 = 15.12 for 1 df; p < 0.001). It is likely that the preventive strategies reduced bone strain by reducing the frequency and forces of impact during the training period.

(Hat tip to Arctotherium.)

How the Voice might work

August 24th, 2025

Dune presents an order of highly trained experts in using the Voice, but I’ve seen little discussion of how this might work:

Your tone, pace, and even the pitch of your voice generate psychological responses. We’re wired to respond emotionally even before we process the words.

How to be taken seriously

People perceive lower voices as more confident and trustworthy, because we associate a lower pitch with authority. So, if you want to be taken seriously, like in a negotiation or conflict, drop your pitch just slightly and speak from your diaphragm.

[…]

How to sound in control

Now, if you want to sound in control, speak at a slower pace. Fast-talkers sound nervous. Slow your speech by 10 to 15 percent, and you’ll come across as more thoughtful, more powerful, and way more in control.

[…]

How to ask for a favor

But, if you want to ask for a favor, or if you’re trying to de-escalate tension, warmth and vocal smile actually matter more than confidence. Add just a touch of softness and upward inflection, and you’re good to go.

[…]

And you don’t have to be a pro voice-actor to do this. You just have to be intentional. And if you’re already doing this naturally, you’re actually using performance psychology.

The future is going to be full of good fortune

August 23rd, 2025

Richard Wiseman enumerates four principles used by “lucky” people to create good fortune:

Principle One: Maximise Chance Opportunities
Lucky people are skilled at creating, noticing and acting upon chance opportunities. They do this in various ways, including networking, adopting a relaxed attitude to life and by being open to new experiences.

Principle Two: Listening to Lucky Hunches
Lucky people make effective decisions by listening to their intuition and gut feelings. In addition, they take steps to actively boost their intuitive abilities by, for example, meditating and clearing their mind of other thoughts.

Principle Three: Expect Good Fortune
Lucky people are certain that the future is going to be full of good fortune. These expectations become self-fulfilling prophecies by helping lucky people persist in the face of failure, and shape their interactions with others in a positive way.

Principle Four: Turn Bad Luck to Good
Lucky people employ various psychological techniques to cope with, and often even thrive upon, the ill fortune that comes their way. For example, they spontaneously imagine how things could have been worse, do not dwell on the ill fortune, and take control of the situation.