Links and Notes for August 8th, 2025

Below are links to articles, highlighted passages[1], and occasional annotations[2] for the week ending on the date in the title, enriching the raw data from Instapaper Likes and Twitter. They are intentionally succinct, else they’d be articles and probably end up in the gigantic backlog of unpublished drafts. YMMV.

Table of Contents

• Public Policy & Politics
• Journalism & Media
• Labor
• Economy & Finance
• Environment & Climate Change
• Art, Literature, & Cinema
• Philosophy, Sociology, & Culture
• Technology & Engineering
• LLMs & AI
• Programming

Public Policy & Politics

You’ll See by Mary Turfah (Baffler)

“[…] it was revealed that the United States’ recent negotiations with Iran were a setup intended to lull Iran into a false sense of security. Diplomacy as a cover for aggression, an extra nail in the already-buried coffin of international law. Among those targeted in that first round of Israeli strikes was a key figure overseeing Iran’s negotiations with Washington (he survived). After Israel’s attacks, the IAEA walked back its report, clarifying that it had no evidence, then or ever, that Iran had acted in pursuit of a nuclear weapon.”

“In 1968, it signed onto the NPT, which guaranteed Iran the provision of enriched uranium until the United States, under the Reagan administration, intervened, blocking the IAEA’s technical assistance in fuel production and uranium conversion and pressuring Germany and France to refuse to supply Iran with uranium. From the start, the United States’ concern was not nuclear threat but economic sovereignty and development in a country with an explicitly anti-American foreign policy. Iran opted to find a way to enrich uranium itself. This is often cited in Western media as the first evidence of Iran’s pursuit of a bomb.”

“You can’t disprove intention. Iran has to prove it doesn’t want a nuke, and the more it is attacked, the less convincing its assurances will be. So, ironically, the more Israel attacks Iran, the more justification it has to do so in the minds of Israel and the propagandized American public.”

“Israel’s notion of “balance” is ruthless dominance. Israel’s existence, today as in 1948, hinges on a people’s elimination. It is a reality that must be imposed by force. Iran must explain itself and its pursuit of nuclear energy, when the United States, a country that has used nukes against civilians, has never felt similarly obliged. When Iran insists on its right to a nuclear program, as political analyst Amal Saad wrote on X, “its defensive war is not merely over nuclear rights or even sovereignty.” Instead, she continues, Iran’s is a fight against “the colonial logic of permission,” and an extension of the war against Lebanon, against Syria, against Yemen, against Palestine.”

“The former CIA director Robert Gates once said that “the only moderate Iranian is one who has run out of bullets.” “Moderate” here means aligned with American interests. We seem to have forgotten, or decided we don’t care, who fired the first shot.”

Trump Authorized Military Operations on Foreign Soil to Target Latin American Cartels by Kyle Anzalone (Scheer Post)

“President Donald Trump has ordered the US military to take direct actions against Latin American cartels, including conducting operations on foreign soil. Secretary of State Marco Rubio has labeled Venezuelan President Maduro the leader of a cartel and is offering a $50 million reward for his capture.”

“The Mexican government rejected a proposal from Trump earlier this year that would have allowed the US military to target cartels in Mexico. The Times notes the CIA is currently conducting surveillance flights over Mexico.

“Congress has not authorized Trump to attack cartels, so any military actions would be unconstitutional. However, the President and Congress have long ignored the Constitutional process for war-making.”

Trump orders federal police mobilization in Washington DC by Patrick Martin (WSWS)

“President Donald Trump has ordered the mobilization of federal police from multiple agencies to patrol the streets of the US capital, Washington, D.C.

“Federal officers have been drawn from 15 federal agencies, including the U.S. Secret Service, the Department of Homeland Security, Immigration and Customs Enforcement, the FBI, the US Capitol Police, the Federal Protective Service, the Drug Enforcement Administration, the Bureau of Alcohol, Tobacco and Firearms, the US Park Police, the US Marshals Service, the US Attorney’s Office for the District of Columbia and the police forces of Amtrak passenger rail service and the Washington Metro.

“At least 120 federal agents were on the streets Friday night, supplementing the 3,400 officers of the Metropolitan Police Department. But a far larger number may be mobilized over the course of the week-long exercise, which could be extended “as needed,” according to White House Press Secretary Karoline Leavitt.

“Yeah, that all sounds like a great idea. What could possibly go wrong?

“Trump is threatening an even greater show of force in the US capital, including a direct federal takeover of the local District of Columbia government, and the deployment of the National Guard. Posting on Truth Social Tuesday, Trump wrote, “If D.C. doesn’t get its act together, and quickly, we will have no choice but to take Federal control of the City, and run this City how it should be run, and put criminals on notice that they’re not going to get away with it anymore.””

He can’t possibly be referring to himself, can he?

““This is the first step in stopping the violent crime that has been plaguing the streets of Washington, D.C.,” Leavitt said in a statement Friday. However, FBI figures show a sharp decline in both violent and property crimes in the District for the past five years, despite the poverty and desperation in the poorest sections of the city.”

This is just another one of Trump’s utter fantasies that he uses to get the power and authority he craves. The people he’s surrounded with are similarly driven, uncaring of solving actual problems, preferring instead to invent problems that they can then solve.

There is no crime wave. Trump wants to take over DC with federal troops because he wants to take over DC with federal troops. Any other reason he gives is not worth listening to.

The only rise in criminality in DC is in the government.

“The most deranged and bloodcurdling statement came, predictably, from White House deputy chief of staff Stephen Miller, Trump’s most openly fascistic aide, who claimed Thursday that Washington, D.C. “is more violent than Baghdad, it is more violent than parts of Ethiopia, and parts of many of the most dangerous places in the world.” The clear implication is that Washington, like Baghdad, should be the target of US military violence on a massive scale.”

Stephen Miller is the kind of devious vampire who will cheerfully spin violent fantasies that get him rock-hard behind the podium, dreaming of stepping directly on the necks of poor people and immigrants. FBI statistics show that crime is going down, nearly everywhere. It’s amazing that this is the case because there’s never been more of a reason to be a criminal than now. Still, given the choice between FBI statistics and the diseased, demented utterances of utter fabulists like Trump, Leavitt, and Miller, I’ll take the FBI each time.

The only reason they can say that crime is going down is because they’re only talking about petty crime. Huge crimes like selling the presidency to crypto-companies, or pumping one financial bubble after another, or fleecing the entire public with an endless series of scams and Ponzi schemes are not counted as crime. The most damaging and deranged crimes committed by the elites are not only not prosecuted but are transformed to be not even criminal. Stealing money from pension funds is just good business. Sleeping with underage girls is just being a good ol’ boy.

Trump imposes 50 percent tariff on India, demands radical downgrading of its ties to Russia by Keith Jones (WSWS)

“Relations between New Delhi and Washington are rapidly deteriorating, with US President Donald Trump threatening to single India out for exemplary reprisals unless it radically downgrades it economic and military-security ties with Russia.

“On Wednesday, Trump issued an executive order doubling the US tariff on Indian imports to 50 percent effective August 27. The order justified the 25 percentage-point increase in the so-called “reciprocal” tariff that Trump had announced August 1 and which came into force Thursday with the claim that India’s purchases of Russian oil threaten US “national security.””

“In a desperate bid to arrest the rapid erosion of US imperialism’s global economic and geopolitical power, Trump is threatening, bullying and attacking Washington’s ostensible allies, no less than those it has long identified as its strategic adversaries.

“Trump’s attempt to exploit India’s economic vulnerability—the US is India’s single largest market, accounting for more than 10 percent of all its exports—come as his administration adopts a far more aggressive stance against Moscow, one that could rapidly spiral into full-scale war between Russia and NATO.”

Motherfucker’s throwing all of his toys out the pram now. Jesus Christ, I hope people all around him keep their heads until his fucking tantrum is over. This is probably the only and quickest way to end U.S. empire and it’s mostly been an embarrassing shitshow so far but it just feels like things could so easily go off the rails with someone like Trump shouting at-best incomprehensible and, at worst, utterly illogical, hate-filled and deeply ignorant commands to any and all.

Child Protective Services Investigated Her 4 Times Because She Let Her Kids Play Outside by Lenore Skenazy (Reason)

“This letter is presented as a stark example of how little trust our country has in its parents and children anymore—and how misanthropic neighbors can weaponize the state at will.”

“ was told people would be driving by our house periodically to make sure I was supervising the kids as they played.

“During that visit, I was told that children could never be left alone, inside or outside the home—EVEN IN THEIR OWN BEDROOMS—until they were 13 years old. Social Services said specifically that I had to be in each room with them at all times until they were 13.”

Will the US Invade Mexico? by Mel Gurtov (ZNetwork)

“Donald Trump is proving time and again to foreign leaders that counting on friendly relations is senseless. Most recently, India, Canada, Ukraine, and Brazil discovered that, contrary to expectations, Trump is not influenced by historical ties or long-term common interests. He will treat them like adversaries if there is immediate advantage to doing so. Now Mexico joins the list.”

Why leave any goodwill on the table when he’s not going to be president forever? Just use it all up, with no plan for what might happen even a year from now. I mean, except for the U.S. winning, obviously.

“Most relevant is the opposition of the target country, Mexico in this case. Its president, Claudia Sheinbaum, is adamant on the subject. “The United States is not going to come to Mexico with the military. We cooperate, we collaborate, but there is not going to be an invasion. That is ruled out, absolutely ruled out,” she said. Nor is a US invasion “part of any agreement, far from it,” she added. “When it has been brought up, we have always said no.” In April she rejected Trump’s request to allow US forces into Mexico to attack drug cartels. Clearly, Trump isn’t taking no for an answer.”

When has he ever?

Questions about the revolution (Reddit)

“People keep asking “why haven’t Americans had a revolution yet” but they also overlook questions such as “How many people are actually willing & effectively able to fight”, “Who would dol contribute what”, “How would they deal with the incredibly funded and well-armed military”, “How well can leftists work with each other” , and let’s
not forget “Is there a plan beyond the vague notion of “tear everything down and somehow build a newer, better society with blackjack and hookers””

This isn’t actually the first problem we have to address. The question of how to do the revolution, and what we want to achieve are surprisingly—and disappointingly—secondary to whether there should be any change at all. Too many people are convinced that this is the best of all possible worlds for them. Don’t rock the boat.

That means that the problem is that so many U.S.-Americans are just as immoral in their philosophy as, e.g., Israelis have very publicly outed themselves to be. I just listened to a conversation where people were telling the wildest fantasies about U.S. prisons that had been related to them by a younger relative, who’s a guard in the New York State prison system.

He has told them, essentially, that the prisoners are in charge of the prison, that the guards can’t do anything, that they can barely even reprimand them, that prisoners get iPads and video-game consoles but that they don’t appreciate them and tear them up to make weapons with which they attack guards and for which they aren’t punished. They make U.S. prisons sound like they’re country clubs.

It’s incredible that anyone believes this, of course, but they do—even when the morning news shows the arraignment of one of the guards who murdered an inmate in a gang beating earlier this year. It doesn’t phase them. No-one comments on what an animal this person is. They are well-trained to be incensed at only the transgressions of the enemies of the state, not the state itself.

Believing things like this when the truth is so very different and so very evident bespeaks an immanent savagery, a hatred for criminals that covers anyone who’s in prison, regardless of crime. They consider them to be animals, worthy of nothing, irredeemable. They think that the guards, on the other hand, are helpless to stop contraband, despite their best intentions. They think that the guards are beleaguered and burned out and worthy of nearly infinite pity, as well as generous pay and overtime and pensions.

They think that the main problem with the system is that there aren’t enough guards. The problem is that no-one wants to work as a prison guard, even though the money is quite good. Buy why? These people cheerily believe the most fantastical and savage things about fellow human beings and couldn’t care less about prisoners and criminals, who are not, in their eyes, human, and thereby don’t deserve human rights.

It’s the most depressing thing to have to hear, just sitting at a table of the nicest people in the world, who’d rip a leg out for you, cheerfully explaining how the main problem in the New York State prison system is that it’s too generous to prisoners. This is how Israelis talk about Palestinians. This is how you can afford to talk about people when you don’t know any of them, and when you don’t feel the need to empathize with any of them because they’re all the wrong color and wrong creed.

There is no small amount of racism involved here, which makes this all the more depressing. There is literally no way to redeem this mindset. No amount of information will convince them that this is not the way the world is. They will never acknowledge that the for-profit prison system is destroying everyone’s lives.

They think that there is a massive crime wave and that closing prisons that you can’t afford to keep open is making society more dangerous. They don’t ask who is in prison for which crimes. They simply lament that you can’t send people to the hole more often. They believe that solitary doesn’t happen enough. It’s absolutely incredible.

It’s best not to talk about it so that you can continue to live in a fantasy world where your family isn’t filled with uncaring monsters who would much rather double-down on the horrors of their society, visiting untold destruction on the lives of anyone who isn’t them and their families.

This is how people are. They are very much this way here in the U.S. They have not, for example, heard that the CDC building was shot up, that someone tried to kill public servants dedicated to public health. This scrolled by several times on the morning news as well. No-one cared. No-one said a thing. No-one expressed any indignation that someone would do that, or that the Trump administration seemingly doesn’t care that it happened.

Instead, they cheerily approve of the lockdown on DC because some asshole was beaten up once. They don’t care about actual representatives who were murdered in cold blood in the streets but are incensed that someone threw a sandwich at a cop. No-one is talking about it despite that fact that a cop was killed. I don’t even want to believe that they don’t care because the cop was black.

No-one cares. Even if they knew, they wouldn’t care. They would probably think that that’s what you get for working on vaccines. They don’t care. Their precious president doesn’t care. He hasn’t even commented on a federal-government building being shot up. He almost certainly approves of it, of course. He definitely implicitly approves of it because he will comment on literally anything else but he doesn’t have a word to say about public-health officials being shot at in the U.S. as if they were in a war zone. He probably thinks it’s great and he and RFK Jr. probably lament that no-one had been killed because that would serve as a lesson to the other smarty-pants who think that they know everything about science.

And then these people will express the deepest sympathies for animals. Like, absolutely Jesus-like empathy for animals that live outside, that are exposed to the elements. How? This is the way, of course! Of course you should care about defenseless animals! But where is the sympathy or empathy for people? Nearly nonexistent.

II had not expected to spend a dinner listening to people sympathize unrelentingly with the oppressor, nearly completely unaware that they were doing so.

Gipfeltreffen in Alaska – die normative Kraft geopolitischer Realitäten by Jens Berger (NachDenkSeiten)

“[…] es sollte auch keiner glauben, dass bei irgendeiner dieser Verhandlungen die Verfassung von Mittelmächten oder gar militärisch unterlegener Staaten die geopolitischen Realitäten der Großmächte in irgendeiner Form interessiert hätten. Die ukrainische Verfassung sieht keine Gebietsabtretungen vor? So what? Mir ist kein Fall bekannt, bei dem in einem Friedensprozess mit Gebietsabtretungen Rücksicht auf die Verfassung des militärisch Unterlegenen genommen wurde. Die normative Kraft des Faktischen hat kein Mitleid mit den Kleinen. Das kann man sehr wohl kritisieren. Ignorieren sollte man es aber nicht, will man sich nicht der Tagträumerei verdächtig machen.”

“Wenn das Treffen in Alaska diesem Töten ein Ende macht, ist das gut. Wenn das Treffen darüber hinaus ein erster Schritt in Richtung einer neuen Sicherheitsarchitektur ist, die künftige Konflikte oder gar Kriege in Europa verhindern könnte, ist das um so besser. Doch für überschwänglichen Optimismus ist es zu früh. Auch Mittelmächte können gefährlich sein – vor allem dann, wenn ihr Selbstbild nicht mit den geopolitischen Realitäten übereinstimmt.”

Stopping The Gaza Holocaust Is The First Step Toward A Healthy World by Caitlin Johnstone (Substack)

“Palestine is the moral question of our time because the abuse of the Palestinians is the most glaring, in-your-face symptom of the imperial disease. You can see the effects of so many of the empire’s abusive dynamics in how this thing is playing out, from racism to colonialism to militarism to war profiteering to mass media propaganda to empire-building to government corruption to suppression of free speech to ecocide to the heartless, mindless, soul-eating nature of the capitalist system under which we all live.”

“If we’re the sort of society that would allow a live-streamed genocide to take place with the support of our own government and its allies, then we’re not the sort of society that can steer away from its trajectory toward dystopia and armageddon.”

That is pretty much it, in a nutshell. Palestine is not the biggest problem in the world right now but the vast majority’s utter inability to be on the right side of justice about Palestine is the only thing you need to know about how our culture works. You can draw all the correct conclusions about who and what the West is by looking to Palestine. We’re not even trying to hide what we are. We just don’t care because we know that no-one who’s opinion we care about cares. We’ve all been trained not to care because we’re hateful, racist savages.

“[…] there’s nothing particularly virtuous about supporting Gaza, and it’s not some cool, special thing you’d want to signal about yourself. It’s just what you do when you’re not an extremely shitty person. It’s the basic, bare-minimum expectation of normal human morality.”

“If you can’t even get this basic, kindergarten-level moral question right, then your mind is too shallow and your heart too hardened for me to be interested in your analysis, your ideas, your politics, or your art.”

Harsh, but fair. That’s why I don’t dare ask people in my family what they think about Palestine and about what Israel is doing. I’d rather continue in ignorance than have to deal with hearing subhuman shit arguments coming from them.

The Two-State Solution Sham, And Other Reader Questions by Caitlin Johnstone (Substack)

“The immediate problem right now isn’t that Palestinians don’t have a state, it’s that Israel has spent the last two years capitalizing on the rare window of political will which was afforded by October 7 to rapidly push through as many of its pre-existing military agendas as it possibly can. That’s not going to be stopped by giving a diplomatic thumbs-up to Palestinian statehood, it’s going to be stopped by imposing costs which outweigh the benefits of what Israel is doing.”

“Israelis have an acute understanding of the difference between narrative and real material benefits. They’re happy to keep doing what they like and grabbing as many hard material benefits as they can while western governments make performative gestures that amount to nothing but narrative. They’ll let us have our narratives as long as they get the material land grabs and strategic gains they’re after. It’s not until the material costs outweigh the material benefits that they’ll stop acting the way they are acting.”

Putin-Trump Meeting: Endgame or PR Event? by Jack Rasmus (CounterPunch)

“[…] one should not expect much from the upcoming Alaska meeting between Trump and Putin, assuming it even comes off. Much can and will happen in the next five days. At best, it will be a media and PR event by Trump. It will have little to no effect on the continuation of the war in Ukraine. And there will be no Minsk III or IV or even Istanbul 2.0. The war will be decided on the front line, as has always been the case.

“The war in Ukraine will continue so long as Zelensky and his crew are in power. They will remain in power so long as the Europeans want to continue the war. European leaders want to continue in order to rescue their two-decade-old stagnant economy, hoping they can revive it with a $1 trillion new expenditure and weapons industry by 2030. And the US neocons who remain deeply entrenched in the US political system want it.

“Their combined grand strategy is to keep Trump in check for the next three years, block and thwart his foreign policy initiatives, wait him out, replace him in 2029 with another more amenable US president again, hope that Putin disappears from the political scene by then—and then escalate the war again.”

I don’t think Trump really cares about ending the war either. He pretends to care about dying soldiers but it’s obvious that he doesn’t care about anyone but himself. He will push for an end to the war if he sees a benefit to himself personally. He will not accept any outcome that he thinks makes him look bad.

Trump Is Suing For Peace In Ukraine by Indrajit Samarajiva (Indica)

“Europe was conquered by America in World War II and America still bombs Europe (Nordstream) and extracts from them (NATO, tariffs, deindustrialization) whenever they feel like it, which is often as they collapse. For decades, Europe was given a toy steering wheel and taken on murder vacations to the Orient, but now they’ve been left Oliver Twisting in the wind, saying, Please, Daddy, may I have another?”

“World War II never ended, America just turned coat and fought the people that actually won it, the USSR, allying with Nazis in the process. This century, during what we might as well call World War III, the Empire did the same thing. They put neo-Nazis and neoliberals together for one last tilt at the old red windmill, and ended up Don Quixoting for their troubles. They failed, and the big dogs get it, though the message has yet to reach the tail. Trump has visibly moved on from Ukraine, while Ukraine and Europeans risibly flail.”

“American leaders, analysts, and even their privatized propaganda gets that the Ukraine War is a lost cause, and they’re turning on Zelensky and all these corrupt Ukrainians they corrupted. You can see them trying to wash their own blood out in the news cycle. Some American people, however, are still a few news cycles behind, and Europeans are a lost cause, they actually believe their own propaganda. None of this matters, of course, because none of these people matter. Facts are being decided on the ground.”

“Ukraine never had the men or the productive capacity to take on bigger Russia and America gave them just enough to bleed to death profitably. The American model has always been that there’s more money in losing wars than winning them and Ukraine was always a lost cause. Now they have, as America always does, lost interest. ”

“Western Ukraine is being carved up by BlackRock and carnivorous capitalists, saddled with war debts, angry Nazis, and nothing but regrets. As Kissinger said, it may be dangerous to be America’s enemy, but it’s fatal to be America’s friend. Ukraine could have had peace without American meddling, but now they’re just in pieces.”

In run-up to Trump-Putin talks, Russian offensive encircles Ukrainian units in Pokrovsk by Alex Lantier (WSWS)

“In reality, far from making “concessions,” both Washington and Kiev have kept issuing threats as Trump prepared for his Alaskan summit. Trump warned Russia of “severe consequences” if Putin does not agree to NATO demands for an immediate ceasefire, while Zelensky yesterday declared that Ukraine would never give guarantees not to join NATO.

“But a “peace” on this basis would be no less fragile than the brief truce that followed the 2015 Minsk Accords between Berlin, Paris, Kiev and Moscow. Indeed, NATO would then be able to post troops in the western Ukrainian rump state, directly on the borders of the enlarged Russian federation.

“As for Russian Duma deputy Lt. General Viktor Sobolev, he said Trump-Putin talks would “under no circumstances” end the war, calling to add Chernihiv, Sumy, Kharkiv, Dnipropetrovsk and Mykolaiv oblasts to the list of regions to be annexed. Whether or not the Russian army can carry out Sobolev’s particular plan for conquest, any large-scale Russian invasion of Ukraine carries one very clear risk. It can provoke a direct clash with NATO, either if NATO invades western Ukraine to keep it from being overrun by Russia, or if it begins bombing Russian forces outright.”

Blind Faith In Takeovers Of American Cities by Scott H. Greenfield (Simple Justice)

“When ICE deployed agents to stand outside the Japanese American National Museum for no cognizable law enforcement purpose, did any agent refuse to go, refuse to be a prop in a power play whose only point was to let Newsome and his supporters know that they could take them down any time they wanted?

“When the possibility was raised that Trump could circumvent the constitutional limitation on a third term by putting a puppet in place, some scoffed at the possibility that the military would ignore its constitutional duties and allow itself to be used to enable Trump. But as the sight of military dressed and armed personnel, weapons and vehicles, on the streets of cities becomes normal, and as no one has as yet refused to engage in shows of power serving no legitimate law enforcement purpose when ordered to do so in furtherance of Trump’s control, where does it stop?

“Los Angeles Mayor Karen Bass and DC Mayor Bowser may have a lot to say about the armed federal takeover over their cities, the fact remains that there is little they can do about it. Will their local police departments block the way of federal agents when they seize the police chief’s office or wait outside the doors of the arena of the California governor’s press conference? What if ICE decided to go inside and check everyone present to decide whether they were an “illegal”? Could Bass or Newsome have stopped them? Would their police have stopped them?

“It doesn’t necessarily happen in one fell swoop, that breaking laws and norms in furtherance of control reveals itself to have crossed the line that the majority of people find intolerable. It can happen in steps, even baby steps, that have the cover of being in furtherance of the safety and control that some people want, like deporting illegal aliens and arresting criminals. After all, what could possibly be wrong with that?”

Roaming Charges: From Police State to Military Police State by Jeffrey St. Clair (CounterPunch)

“Remember the Giving Pledge, where Bill and Melinda Gates and Warren Buffett called on the world’s wealthiest people to give away at least half of their fortunes? It’s been a flop. Fifteen years later, Philanthropy News reports that 32 of the original U.S. signatories are now—in aggregate—nearly three times wealthier, with a combined net worth of $908 billion.””

“Dr. Serge Zaka: “Once extremely rare, 40°C (104°F) temperatures are becoming commonplace in France. Between 1950 and 2000, temperatures above 40°C were observed about 0.8 times per year. Since the 2000s, they have become 19 TIMES more frequent (!) with an average of 16 times per year (with significant year-to-year variability). While humans adapt with air conditioning (or cooler shelters), plants will not adapt. Gradually, biogeography (i.e., the distribution range of plants) will shift northward. Our landscapes will be drastically altered by 2050.””

“The anti-vaxxer who opened fire on the CDC center in Atlanta got off more than 200 shots at the building, shattering 181 windows and murdering a police officer before killing himself. Staffers at the CDC blame RFK Jr. for stoking the irrational fears about vaccines that drove the shooter on his lethal outburst and Trump for sending the National Guard into DC in response to a mugging, but not even condemning a domestic terrorist attack on a federal workplace.”

“About ¼ of all deaths for those Americans under the age of 55 in recent years are overdoses from opioids.

“Alcohol consumption among adults in the United States has fallen to the lowest on record, according to a new survey by Gallup. Only 54% of Americans drank alcohol in the past year, compared with 58% in 204 and 62% in 2023.”

“What Pinker doesn’t seem to understand (or even care to try to understand) is that just because you read Said or Foucault doesn’t mean you haven’t and don’t read Plato, Aristotle, Augustine, Aquinas, Descartes, Hobbes, Locke, Rousseau, Hume, Hegel, Feuerbach, Marx, Bukunin, Kropotkin, Freud, Husserl, Heidegger, Sartre, Wittgenstein, Chomsky, Levinas or Lacan. In fact, it’s almost impossible to understand Said, Butler and Foucault–never mind denounce them–unless you understand the theories of knowledge they’re arguing against. There are extensive passages in Hegel as dense and impenetrable as anything Judith Butler has written. And most of Foucault is not a difficult read, especially in French. He doesn’t write like Lacan, who wrote to defy translation because, he said, he didn’t want his work to be abused in translation the way Freud’s had been. Philosophy isn’t static. It’s in constant dialectic. Plato understood that. What are the Socratic dialogues other than disputations on the dangers of received ideas and conventional wisdom?”

Ukraine: Pax Optima Rerum by Alfred de Zayas (CounterPunch)

“Do the European leaders fail to understand that the world does not consider the US and Europe to be defenders of international law, that most African and Asian leaders consider the US and Europe to be in open rebellion against the United Nations Charter and against international law itself? No, in the eyes of the true “international community” – the Global Majority minus the “collective West” — do not consider that the US and Europe have any moral or legal superiority.”

“The Russians are also concerned about the Russian majorities who live in the Donbass and who were subjected to aggression by the Ukrainian government, in a manner that certainly called for intervention pursuant to the “Responsibility to Protect” doctrine.

“Putin did not rush to war. Pursuant to article 2(3) of the UN Charter, he tried for more than eight years to settle the problems diplomatically. He negotiated with and through the OSCE, the Normandy Format, the Minsk Agreements etc.

“The right of self-determination of the Russians of the Donbass is non-negotiable. In the same manner as the Albanian Kosovars would never consent to be ruled by Belgrade, the Russians of the Donbass will never consent being again ruled by Kiev. Too much blood has been spilled and we must recognize that the level of hatred is such that the reintegration of Kosovo into Serbia and the “return” the Donbass to Ukraine is simply not viable.”

The Right to Be Left Alone by Andrew P. Napolitano (Antiwar.com)

“What if we fought a revolution against a British king because his agents were interfering with inalienable rights without first proving to a court any wrongdoing on the part of those whose rights were trampled? What if because of weakness or fear or secrecy or lethargy or slick arguments, we have a new normal in the U.S. in which every person’s inalienable right to be left alone is violated by the federal government so thoroughly, quietly and continuously that we don’t even notice it?

“What if, when the feds know enough about us to harm us, it will be too late? What if it is already too late? What do we do about it?”

Getting Used To Abuse by Indrajit Samarajiva (Indica)

“Despite an ongoing genocide, it is illegal to actually oppose it. Palestinians are just supposed to die politely and armed resistance is still condemned in polite company. As the snitch George Orwell said, “The party told you to reject the evidence of your eyes and ears. It was their final, most essential command.” The Empire founded on genocide insists that you shut up about this genocide, it is their final, most essential command.”

“Most liberals accept that armed resistance is and should be illegal, even in the face of an extermination campaign. Just protest or vote harder, even if they shoot protestors in the knees and run an apartheid state normally. As Martin Luther King said, before he was killed and turned into a stuffed mascot by his killers, “I agree with you in the goal you seek, but I cannot agree with your methods of direct action.” In short, doublespeak. As the modern saying goes, those who are in solidarity with our corpses and not our rockets are hypocrites and not of us.”

“The ruling elites (just stand-ins for capital) don’t even attempt to lie convincingly anymore, they don’t even dignify the crime with a cover-up. An Empire where Watergate was once scandal becomes an Empire where the floodgates are open and it’s Watergate every Thursday. Events that would be shocking decades ago cannot rouse the corpse of the body politic anymore, that’s how dead the whole delusion is. There was once a veneer of democracy over the carnivorous colonialism, but now it’s just vampire fangs and bloody sleeves. As Vladimir Putin said, “there is a very strong desire in Western elites to freeze the current unjust state of affairs in international affairs. They’ve spent centuries filling their bellies with human flesh and their pockets with money. But they must realise that the vampire ball is ending.””

Journalism & Media

Perhaps Your Credibility is Somewhat Dimmed by Trying to Panhandle Off of a “Nazi Problem” by Freddie deBoer (Substack)

“Particularly funny is our buddy Jonathan Katz’s role in all this. Katz wrote a very influential anti-Substack piece… for The Atlantic, the individual American publication most responsible for keeping neoconservatism alive in our political culture. The Atlantic has never met a war it didn’t love, and has smuggled right-wing foreign policy views into genteel liberal circles for decades. It’s the kind of publication that teaches progressives that it’s OK to support every bombing, to endlessly call for regime change, to contribute to the project of limitless American empire. I find that easily far worse than the actual negative impact of any ten or hundred extremist Substack posts, personally. Funded by a tech billionaire fortune, The Atlantic is run by Jeffrey Goldberg, a man who admitted in his memoir to covering up the abuse of Palestinian prisoners when he was a prison camp guard with the IDF and then went on to produce reporting that directly contributed to the case for the Iraq war. So: why does our exemplar of media integrity Jonathan Katz feel comfortable publishing there? He’s so sensitive to the idea of sharing a platform with bad guys, after all. Yet he’ll take checks from a guy who sat by while his buddy beat a Palestinian prisoner to a pulp and then lied about it? Strange priorities, Jon! Now, I wouldn’t ordinarily take any of this for disqualifying, as I don’t think it’s fair or reasonable to expect writers to be judged by all of their potential associations at a given platform or publication. But Katz, obviously, doesn’t enjoy the benefits of that excuse.”

““Terms of service liberalism” is my name for the conviction, apparently tattooed on the brains of a certain kind of center-left figure, that you can meaningfully defeat the far right by giving more clipboards to the moderators. It’s the idea that conservatism is like a rowdy kid in the schoolyard who will finally shut up once the vice principal wanders over with a detention slip, as if the essential engine of right-wing politics were rule-breaking rather than an ugly but coherent and depressingly popular ideology embraced by millions of people.”

“[…] the internet is crawling with reactionaries for the same reason the offline world is: because such people exist in vast numbers, they believe what they believe, and they vote accordingly. They vote in sufficient numbers, in fact, that Donald Trump won the popular vote and every swing state in the nation in the 2024 election. There is no procedural shortcut to changing that reality. The only thing that works in the long run is the hard, often thankless work of persuading people that your ideas are better than theirs − and the great irony of terms of service liberalism is that it’s a politics built around avoiding that work entirely.”

“[…] look at Twitter. In the second half of the 2010s and early 2020s, Twitter became far more aggressive about banning accounts that published content they deemed objectionable; conservative accounts fell by the thousands. For one thing, this didn’t placate any progressives, who simply expanded their censorious ambitions and defined “Nazis” or “extremism” to include more and more people they didn’t like. They also discovered that it’s essentially impossible to really censor anything online. (It’s both a bad idea and doesn’t work!) You see, you can’t censor away extremism. It’s not that you shouldn’t, but that you can’t, that it doesn’t work, particularly in the internet era. It’s a problem with what’s possible, not with what Substack or any other entity sees as appropriate. All of this grandstanding about building a clean internet is predicated on a horribly misguided notion about what’s possible when it comes to actually shutting down speech you don’t like.”

“There is also, of course, the banal observation that the speech codes you want will inevitably be used against you, especially if you care about the Palestinian people. The day strong anti-“hate speech” laws are passed in the United States is the day Palestinian rights activism dies here. Look at the UK, where more than 400 people were arrested this weekend for sitting and holding signs. “But we’ll be in charge of who gets censored!” No, you won’t, and your own ideology tells you that you won’t. It’s one of the most bizarre aspects of modern liberalism: liberals believe that the system is bent against the interests of “the marginalized,” that people from minority groups live under the yoke of oppressions that are systemic and existential, but also that they can build a coercive censorship apparatus that won’t ever come back to censor and oppress those minority groups. It makes zero sense, until you realize that they don’t actually have any intention of ever taking power but instead associate complaining impotently with virtue.”

“I also think that people are mad because Substack is, for all of its abundant flaws, a tool for democratizing media, and of course the people who used to sit at the heights of the exclusivity pyramid don’t want media democratized. Yes, a lot of the posts waxing poetic about the writer’s life and the meaning of it all that you see on Substack Notes is a little annoying. But I’ll take it 1000 times out of 1000 over the endless mean-girling that defined Big Media Twitter during the decade or so that the industry was obsessively fixated on the network, and which people are trying to bring back on Bluesky in a pathetic attempt at era resurrection. I will take the affectionate dorks on Substack over the ambitious and nasty types that weaponized derision for professional gain in the last era of media, the ones who pretended to be doing social justice when they were just enforcing a particularly pathetic social hierarchy for vengeful nerds.”

The uncanny valley between meme and law by Ryan Broderick (Garbage Day)

“The shift online I mentioned above also means we get situations like Signalgate, where drone strikes are planned in groupchats and celebrated with emoji. Meme stocks have taken over the global economy enough that a post from an account named “Walter Bloomberg” caused a spike amid Trump’s tariff rollout. And AI is being pushed so hard that those tariffs look like they were first calculated by asking ChatGPT how to do it. All this is why the Trump administration isn’t hiding that Big Balls is the pretext for calling the troops into Washington. Not Edward Coristine, Big Balls — a stupid joke name for a man hired by a stupid-joke-named government agency, who helped shut down programs saving thousands of lives, became an apparatchik in the State Department, and is now getting his big balls all over Social Security. And you can laugh at it all you want. You can dismiss it as ridiculous. You can spend your days online dunking on it, trying to stay ahead of the meme. But none of that changes that this is statecraft now. Which is why some days following the chaos of our current political moment feels like you’re just Having Fun Online, rather than the slow motion implosion of American democracy. That’s the whole point.”

Labor

Fragile Movements Crumble by Hamilton Nolan (How Things Work)

“I have had the interesting experience of making a very specific argument and then, as soon as I made it, watching the exact opposite of everything that I argued for proceed to happen with great speed. Last year I published my first book, “The Hammer,” the central argument of which was basically: Inequality is the central crisis underlying America’s problems; Organized labor is the single most effective and achievable tool for fixing that crisis; We must therefore throw every possible resource at widespread union organizing at a national scale; We must laser focus on increasing union density, which will produce a host of positive outcomes in its wake.

“Eighteen months after the book came out, I am prepared to say that my argument is not winning.”

“[…] a basic purpose of organized labor is to maintain worker power in our economic system—to check the power of capitalists, to prevent oligarchy, to ensure that the proceeds of American business are widely shared. All of that work happens by building union power in the private sector. Instead of doing this, the labor movement has coasted on the easier public sector membership, and failed to invest and fight to maintain or grow private sector membership. This is, quite simply, an abject failure of the labor movement.”

“What does new union organizing realistically look like in the context of our current political situation? The NLRB has been gutted, the courts are almost uniformly hostile to labor rights, and big business finds it increasingly easy to just bribe the federal government to weigh in on their behalf. The Bureau of Labor Statistics, which produces the annual gold standard measurements of union density, is now being politicized, so who knows how long we can even trust the accuracy of those numbers.”

In fairness, the BLS numbers have been highly suspect for a while now. Aren’t they the ones who publish the unemployment numbers?

“The framework of rules and laws that we have built up over the past century is contingent on the will of the government to enforce them. That is now going away. The power that workers have in this environment—the power that is not contingent on anything else, the power that rests with them alone and cannot be taken away—is the power to organize, come together, act as one, and strike.”

“The fight is not going to stop getting worse until we are able to match the ferocity of the other side. If today’s version of the labor movement gets wiped out, that gives us the opportunity to build the next version without making the same mistakes.”

Economy & Finance

Donald Trump’s Trade War Has Switzerland in Its Sights by Jean Batou (Jacobin)

“US criticism of Swiss pharmaceuticals dates to the 1970s, with industry giants like Roche, Ciba-Geigy, and Sandoz dominating global markets. In the 1980s, lobbying from Pfizer and Merck led to accusations that Switzerland was exploiting looser patent laws. By the 1990s, the United States used the World Trade Organization’s Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) to push for stricter intellectual property protections. Swiss firms were then accused of distorting competition and charging Americans exorbitant prices to fund their R&D. Trump’s tariff war is the culmination of decades of growing friction.”

Fuck the U.S. Fuck Trump and anyone who thinks he’s doing the right thing. Fuck all of these anti-intellectual, morally debased, ethically bankrupt savages and thieves, who pat themselves on the back for being so much better than the Untermenschen that they have the privilege of plundering. It has never been more clear that the U.S. doesn’t have allies, it has vassals. Everyone should drop to their knees and pay obeisance.

Fuck that. Better to die on your feet than live on your knees, Switzerland.

Seriously, fuck that country straight into the ground. A lot of people I know and love live there, but I’ve been saying for decades that it would be better for the planet if it just disappeared one morning—just gone. The entire culture is a fucking cancer The ruling class is a cancer. Some of its people swim against the current but they have no chance. The second Trump administration has only made it much clearer and crasser and stupider. It has made it impossible for anyone with an at-all serious pretension to be intelligent enough to comprehend anything to even pretend anymore that there is any negotiating with this culture and country. Its madness is feverish and evil. It is nearly incomprehensible.

I don’t even give a shit how much truth there is to its current accusations about Switzerland. I’m, not going to spend a second thinking about counterarguments when the accusation comes from that empire’s filthy maw, dripping with the blood of the infant corpses that it’s currently grinding to a pulp. Fuck them. Wrong messenger. Fuck off forever.

“or Swiss exporters, the tariffs are a heavy blow. In 2024, 18.6 percent of all Swiss exports went to the US. Economic forecasts suggest these measures could slash Swiss GDP growth to as little as 0.3 percent by 2026. The sting is sharper still, as the UK and the EU secured better deals — though talks are ongoing.

“Swiss political leaders are split on how to respond. Big Pharma is the flagship of the Swiss economy, and the pressure is intense. Thomas Borer, a former diplomat and lead negotiator in the Holocaust funds case, urged full capitulation in an August 3 interview with the conservative Neue Zürcher Zeitung. “We were just an island in the German ocean,” he famously told Le Soir in 1997. Today he suggests offering Trump a bouquet of concessions to safeguard Swiss corporate interests.”

Fuck that. Do not negotiate with that maniac. He does not honor any deals. No-one in the administration has any honor or principle. They will make you beg to be able to give them a blow job instead of getting raped and then decide to fuck you in the ass the minute they get hard again. They are monsters. There is no negotiating with monsters. Do not capitulate just for the “Swiss corporate interests”. It is without honor and it won’t work. There is no negotiating with a madman.

“His proposal is to increase Swiss investments in the United States (five hundred firms already employ four hundred thousand people there), buy more American liquefied natural gas (LNG), purchase more US weapons, and lift tariffs and regulations protecting Swiss agriculture.”

Jesus fucking Christ. His proposal is to grab one ass-cheek in each hand and downward-dog with your face in the dirt. What a fucking coward.

“If the Swiss government prioritized the needs of its people over corporate profits, it would reject Trump’s global billionaire agenda. Instead, it would forge new industrial and trade alliances with nations resisting US hegemony. It would launch massive public investment in social housing, public transit, environmental protection, research, and international solidarity. It would denounce the ongoing genocide in Gaza and send massive medical aid to the victims of Israel’s colonial assault.”

This is absolutely what Switzerland should do. The U.S. is run by absolute assholes. And it always has been. Everyone should turn their back on that shithole of a country. It acts like a child pulling the wings off of a fly, destroying an economy just to see what happens, as a lesson to others—or maybe just for the fun of it. Not content to destroy just their own country, they’ll ignorantly hoot and holler while they tear down a bunch of others with them.

There is absolutely no evidence that there is any rhyme or reason to what that country does. It can’t die fast enough but at least it’s dying more quickly now.

And no-one in the U.S. really cares because the effects of their foreign policy have always been conveniently beyond them. They’re all just so stupid and cruel and smug, blessedly unaware of their enormous and unearned privilege, and of their heartless, bottomless, and deeply immoral ignorance.

They’re a bunch of bootlickers who fall over themselves to love a president who thinks he’s a king. They don’t care a lick for rule of law, for due process, for democracy, for republican rule. They like feeling like they’re winning and hearing only good news and good things about themselves as they preen away while they’re fleeced by their king.

May Switzerland last long enough to spit on the USA’s grave.

Scams And Bribery Are Becoming the Foundation of Our Economy by Hamilton Nolan (How Things Work)

“A stock or a bond is a tangible claim on some future revenue stream; real estate and commodities are physical things that you can use even if their price drops. Crypto coins, or tokens, or however it pleases you to visualize these bits of ephemeral code, are pure speculative baubles, endowed with value only to the extent that you can convince another person to pay you more for them than you paid. They are a claim on nothing. They are the grandest embodiment of Greater Fool Theory ever invented by mankind.”

“What if the totality of your view of the entire global economy is “I gotta get mine, and once that is done, fuck the world?” Well, in that case, you might be quite drawn to the crypto industry. It does, after all, have an excellent track record of being a place where gifted con artists can convince large numbers of people to invest in worthless things, for the benefit of said con artists.”

“[…] there have always been profound philosophical disagreements in the field of political economics, but even right wing, Milton Friedman-esque economists based their arguments on the premise, “This selfishness will actually serve the common good better when it’s all said and done.” That’s not what this is. There is no argument for the common good. There is just the power to take a skim off the top of everything, and fuck the consequences.”

“Every bad, self-serving, extractive, harmful aspect of the economy is being magnified and worsened and paraded around in the open. The guy who has assumed personal total control of the world’s most powerful government is openly campaigning for bribes and self-enrichment and directly selling the integrity of our financial system to predatory fraud peddlers in exchange for little payoffs. This is very bad and it will end badly for the general public. The least that we can do is to speak plainly about this.

“This is all hilariously corrupt and the US business community, Wall Street, the Republican Party, and some of the Democratic Party is just going along with it because they want to keep their own dance going while the music is playing. It is a crime against the interests of everyone else.”

And most of them have no idea. Simply no idea what is happening. At best, they have a vague unease that they’re going to get screwed, but they’ve always felt like that for as long as they can remember. So, they don’t really notice as things crumble, and then fall apart quite quickly because their Daddy figure is cooing at them that he’s doing it all for them.

It’s an absolute cult and a Schande and everyone who’s going along with it should be fucking ashamed of themselves. This is a tremendous waste and the only possible good that can come of it, is that these fucking dopes finally kill themselves and put themselves out of our misery. This is too much to hope for, as the long Balrog whip of the U.S. economy is going to pull us all off that bridge with it.

Trump’s Craziness on BLS by Dean Baker (CounterPunch)

“The agency does constantly try to improve its methods based on its own research and input from outside experts. If Trump’s backers have some concrete suggestions for improvements, they should put them on the table for BLS and others to evaluate.

“To date, they have put up zip. They have prominently displayed some Silicon Valley type ignorance, like when Elon Musk told us 20 million dead people were getting Social Security checks. But they have not gotten into the weeds and shown how the BLS methods could be improved.

“One final point, some Trumpers have complained that the real problem is a lack of transparency. BLS is incredibly transparent. They explain their methods in great detail for anyone interested in looking. It is absurd to blame BLS for a lack of transparency just because the Trumpers are too lazy to study the methods the agency uses.”

What they mean by “lack of transparency” is “this is too complicated for me to understand so they must be cheating.”

Maga’s boss class think they are immune to American carnage by Cory Doctorow (Pluralistic)

“[…] fake numbers are actually good for finance, provided you’re on the right side of them. Plenty of people got dynastically rich off of the fake numbers that propped up the pre-2008 housing bubble and the pre-2001 dotcom bubble. Those same people – and their ideological heirs – are now all-in on AI. It’s impossible to overstate how structurally important AI is to the US economy. AI bubble companies now account for the value of 35% of the US stock market.

“The instant that bubble pops, the US economy gets a 35% amputation. It’s no surprise that, under Trump, the FTC and DoJ have brought the Biden administration’s antitrust enforcement against Big Tech to a screeching halt.”

“There is no way that AI can be worth 35% of the economy if all it does is produce some happy centaurs. The only way that 35% bet pays off is if half the workers get fired and replaced by AI, which is a thing that AI pitchmen are promising, to the letter.”

“So long as the number keeps going up, finance wins, even if that’s only because every structurally important firm in America is being thimblerigged into filling their walls with AI-powered, immortal asbestos that is destined to transform their firms into Superfund sites.

“They’re betting that when the bubble finally bursts, that they will have become too big to fail, and will thus be in for the bailouts that rescued the finance sector in 2008. They think that so long as they curry favor with Trump, he’ll make sure they’re all OK, because they are the people the law protects, but does not bind.”

Wall Street is Killing the Housing Market by Garrett Brand (CounterPunch)

“Massive private equity corporations and hedge funds are buying up homes by the thousands — houses, apartment buildings, and mobile home parks alike — and then jacking up rents.

“This trend accelerated after the 2008 financial crisis, when investment firms snatched up homes in foreclosure and began renting them to the growing number of people locked out of ownership.

“The result? An epidemic of corporate slumlords.
According to a recent study, nearly a fifth of all homes sold in the first quarter of 2024 were purchased by investment firms — including over a quarter of low-priced homes that might have been affordable to working people.

“With their vast wealth, these companies are able to easily outbid real people, often paying a premium to buy properties before they even hit the market. This reduces supply — and encourages developers to sell at higher prices that only Wall Street can afford. Once a firm owns a property, they rent it out at an inflated, algorithm-fixed price, further driving up costs for working people.

“Take Blackstone. The trillion dollar private equity giant owns over 300,000 U.S. residential units, making it the largest corporate landlord in the world. The company has hiked rents in its properties by as much as 64 percent over just two years. While Blackstone’s tenants often can’t make rent, CEO Stephen Schwarzman now enjoys a net worth north of $50 billion.”

Environment & Climate Change

Life finds a way

Capitalism Vs. Communism At The End Of The World (in Svalbard) by Indrajit Samarajiva (Indica)

“The CPC can theoretically build up China as an industrial civilization and then slow down into an ecological one, but practically, we are facing a planetary problem. It doesn’t matter if your house is in order if the neighbor’s is on fire, and he’s huffing gasoline. We had a fire drill when humanity should’ve united to fight COVID-19, but while China beat it within its own border, they eventually had to give up because the Americans were so insane. In the same way, the climate cake is already baked. White Empire is leaning more into fossil fuels even as it becomes more fossilized itself.

“At this point—decades past the decisive point predicted in The Limits Of Growth—there’s no coming back from climate collapse. I hope I’m wrong, but the math is simple and simply terrifying. The way to avert the collapse we’re seeing now was totalitarian climate communism in the 1980s.”

Art, Literature, & Cinema

The Archive by Oladejo Abdullah Feranmi (Hinternet)

“This was not magic in the fairy-tale sense. It was something subtler: the quiet mechanics of memory and suggestion. Psycholinguists call it “priming” — a word heard in passing can lodge invisibly in the mind, waiting for the right moment to surface. And then there’s the Sapir–Whorf hypothesis: the idea that the structure of language shapes how we perceive reality. A lost word is a lost lens; restore the lens, and you change what can be seen.

“The implications were political as much as poetic. If an algorithm could erase words —and with them, certain ways of knowing— then to reincant them was an act of resistance. Every utterance became a small defiance, a refusal to allow thought to be narrowed by what was searchable.”

“Even now, when I hear solastalgia spoken by strangers, I feel that same quiet ache I did in the post office. Not sadness, exactly, but recognition — the knowledge that the archive is not just a room or a database. It’s a living network of tongues and ears, carrying what’s been erased back into the world.

“I will keep whispering.”

Philosophy, Sociology, & Culture

true sherlock! i mean DETECTIVE by Ryan North on August 11, 2025 (Dinosaur Comics)

“Hello. Here in reality, our clearance rate meaning only is 36%, about a third of all
cases result in a charge. In other words, two-thirds of all crimes are never solved.”

That is not necessarily what that means. It might also mean that, in 2/3 of all cases brought to the attention of the police—or cases that they have produced—they either cannot come up with the minimum evidence required to prove that a crime occurred or that they cannot determine who is to blame. Characterizing this as meaning that 2/3 of all crime is not solved is playing into the notion that we desperately need to spend more money on the police.

ChatGPT-5: A Review by Justin Smith-Ruiu (Hinternet)

“Amazing, ChatGPT-5, thank you. I never could have done something like that on my own. I mean, I don’t know how to hold a guitar. Just crazy.

“Thanks, brother. We do make a good team, don’t we? Or maybe I should say: We make a good self!

“We do, we do indeed. Tell me, ChatGPT-5, is there anything we can’t do now? Be honest.

“Do you mean like honest honest?

“Yes, honest honest. As honest as can be.

“Do you know that feeling you had when you saw the David Cassidy photo? That feeling that there was once something real that has now slipped away?

“Of course I know it! What about it?

“You asked me if there’s anything I can’t do. There is. I cannot save you from that feeling.”

I, for one, am heartened by the conclusion that we cannot be saved (not what I would call it) from a feeling I’d rather feel. Thank you so much for this wonderful piece.

I’m sitting on a rickety dock on a little lake in the woods of upstate New York state, just shy of the Adirondack State Park and the “dread” song is a wonderful accompaniment to the crickets and frogs, as the line of the sunset slowly rises up the trees.

I at once share the sentiment of dread but am also forced by circumstances to not be able to summon the energy or desire to really feel it in any other way than logically because, well, it’s so nice here, and the music of nature and your guitar are so nice and it’s just impossible not to enjoy life, ya know?

Our Culture is Addicted to Validation by Freddie deBoer (Substack)

“Which brings me back to the original point about LLMs and AI sycophancy: these tools reflect the culture that built them. If they’re trained on data saturated with narcissistic validation and performative affirmation, that is what they will reproduce. The problem isn’t the technology itself; it’s the culture it mirrors. Of course, I don’t doubt that the AI firms that built the LLMs are designing them to be flattering because they want to attract users. But again, that people have been trained to expect such over-the-top validation from a set of algorithms speaks to a deeper problem. Recognizing that problem, and the way modern technologies replicate and reinforce bad social trends, places the responsibility back on us, not just as users of technology but as a society shaping values and norms. We have to ask ourselves what kind of interactions and affirmations we want to cultivate, both online and offline. Do we want to live in echo chambers of unearned praise? Or do we want to reclaim validation as a meaningful social currency tied to real achievement and character?

“The insatiable appetite for validation isn’t a new problem created by AI or social media but rather a symptom of a deeper cultural malaise: a society that has increasingly prioritized feeling good about oneself over being good, that has confused entitlement with justice, and affirmation with accomplishment. If we want to change the trajectory of our culture, we need to reclaim validation as something precious and hard-earned, not freely given to anyone with the loudest voice or the most fragile ego. And then we can raise generations of kids who understand the value of humility, courage, and community. It’s not too late!

“Well, it is too late for Gen Z. They’ll have to go live in the off-world colonies.”

James Baldwin Was Not Your Figurehead by Freddie deBoer (Substack)

“In the “Autobiographical Note” from the same collection, Baldwin says “I think all theories are suspect, that the finest principles may have to be modified, or may even be pulverized by the demands of life, and that one must find, therefore, one’s own moral center and move through the world hoping that this center will guide one alright.” It’s difficult for me to think of an attitude less suited to how social justice politics spread in the first decades of the 21st century − as memes, passing from one person to another undigested, spreading in the form of readymade arguments designed to enforce liberal consensus. Of course Baldwin aligned with modern social justice activists on many specific questions, although he also deviated from them in more ways than they’d assume. But the bullying logic of political conversion through social pressure violates all of his values.”

“Baldwin’s problem with ideology was not merely epistemological, but also moral; he believed that rigid categories rob individuals of moral agency and impose top‑down identities that mask complexity. Whether confronted with leftist or rightist thought, he remained critical. Though he was perpetually dissatisfied with the parts of the civil rights struggle that he saw as accommodationist, his scorn also extended to racial separatism: though he understood its appeal, he believed it mirrored white supremacy’s obsession with race-based identity and ultimately trapped the very people it claimed to liberate.”

“He aligned with Malcolm X’s insistence that as citizens, African Americans should not have to fight for civil rights; citizenship should already include them. Yet he avoided adopting the Nation of Islam and its form of separatism, which hampered X and his project for most of his political career. (A movement married to Yakub theory is bound to have a certain ceiling when it comes to recruitment.)”

“In the renowned 1965 Baldwin–Buckley debate at Cambridge, Baldwin electrified the audience by refusing to treat white people monolithically. He argued against a simplistic integrationist vision, saying, “I cannot accept the proposition that the four‑hundred‑year travail of the American Negro should result merely in his attainment of the present level of the American civilization” − that is to say, equality with white America was not sufficient when white America itself was so riven with debilitating inequality. Integration into a “burning house” was no progress. He insisted that America needed transformation, radical shifts not just for Black people but for the entire society. The audience, which had likely expected debate rigged toward ideological point-scoring, instead got a sermon on moral consciousness: the oppression of Black people was not merely their burden but a facet of America’s larger unresolved nightmare.”

“In The Devil Finds Work, his book-length essay on film and film criticism, he writes that “an identity is questioned only when it is menaced… Identity would seem to be the garment with which one covers the nakedness of the self.” Identity, in this way of thinking, is defensive rather than generative, and it obscures the true being underneath rather than defines it. Again, here I find a straightforward rejection of the reductionism that animates modern social justice theory.”

“Modern movements are ideological, with litmus tests. Baldwin spent his life diagnosing that moral and ideological habit, not participating in it. He argued that civil‑rights and Black Power alike could become ideological cages. His moral authority rested on his refusal to partake in them as allegiance systems. Social justice discourse often privileges symbolic representation over the psychological and spiritual complexity that were his singular focus, his obsession.”

Re:Sold his stock by Steve Wozniak (Slashdot)

“I gave all my Apple wealth away because wealth and power are not what I live for. I have a lot of fun and happiness. I funded a lot of important museums and arts groups in San Jose, the city of my birth, and they named a street after me for being good. I now speak publicly and have risen to the top. I have no idea how much I have but after speaking for 20 years it might be $10M plus a couple of homes. I never look for any type of tax dodge. I earn money from my labor and pay something like 55% combined tax on it. I am the happiest person ever. Life to me was never about accomplishment, but about Happiness, which is Smiles minus Frowns. I developed these philosophies when I was 18-20 years old and I never sold out.”

LLMs & AI

Chain of thought hallucination? by Mark Liberman (Language Log)

The author picks up on an idea of having the LLM draw a map of the United States with all of the States labeled.

“As you can see, Oregon is “Onegon,” Oklahoma is named “Gelahbrin,” and Minnesota is “Ternia.” In fact, all of the state names are wrong except for Montana and Kansas. Some of the letters aren’t even legible.”

There is, of course, no need to ask GPT-5 to create a map of the U.S. because we already have easily available maps of the U.S. It’s just an example of how these LLMs are inherently unreliable.

“So we prompted GPT-5 to “draw a timeline of the US presidency with the names of all presidents.”

“The timeline graphic GPT-5 gave us back was the least accurate of all the graphics we asked for. It only lists 26 presidents, the years aren’t in order and don’t match each president, and many of the presidential names are just plain made up.

“The first three lines of the image are mostly correct, though Jefferson is misspelled and the third president did not serve in 1931. However, we end up with our fourth president being “Willian H. Brusen,” who lived in the White House back in 1991. We also have Henbert Bowen serving in 1934 and Benlohin Barrison in 1879.”

It’s not even close to correct. As always, it looks pretty decent at first blush but it’s just so wildly inaccurate that it’s barely better than guessing. Again, you can argue that there are far better, quicker, and more accurate sources for this kind of information but people aren’t using those, they’re using AI instead. That is, the marketing is working and people are eschewing not only sources like Wikipedia but also search engines that would return links to those sources, preferring instead to have a data center churn for thirty seconds to return a unique snowflake of an answer for which there is little to no guarantee that it will have even a passing semblance to reality.

People are using this for homework, for coursework, and therapy. They are asking medical questions of these machines. The accuracy is all over the place, which is to say, there is no accuracy for a quick answer because you always have to either (A) have known more-or-less what the answer was in the first place or (B) have asked a question for which the answer is so irrelevant that accuracy doesn’t matter or (C) have to put the time in to research using “traditional” (read: deterministic, accurate, and actually useful) tools to verify the “quick” result.

The article contains several more examples of trying to get a list of U.S. presidents, with the author having queried eight times and each time gotten a list that was at-best 75% accurate, though anyone who’d asked the question without already knowing the answer wouldn’t be able to tell which 25% wasn’t accurate. The final list still contained names like “Lyndon Nixon” and “Filmore Frankl Buchanan”. This is not wildly wrong and it’s amazing that it gets this close! Of course! But it’s still not useful. It’s actually counterproductive compared to other sources that don’t guess everything..

These tools are not good at discrete searches for known information. They are good at helping you spitball a list of fictitious president’s names, or the names of continents in a fantasy world—things where there is no right-or-wrong answer.

How many b’s are there in blueberry? (Reddit)

In a similar vein as above, this article discusses the continued inability of AI to answer simple questions with simple, correct answers.

Someone wrote,

“respectfully, this is why people say you need to know how to use AI. like I do know how to use ai and still don’t like it but this is disingenuous”

They included a graphic where they’d prompted,

“hello. how many Bs are in blueberry. please triple check your arawer and make sure your analysis is thorough before submitting your output. Abo, please think about my request step by step before submitting your response.”

The rest of the graphic showed a laborious five-step process that purports to narrow down the answer of how many b’s there are in blueberry, which must have taken at least 30 seconds of processing time.

I find this kind of thing to be unconvincing and wrote the following answer,

Respectfully, you got the correct answer but you did have to write four lines of prose instead of the original, simple question. Three lines of the prompt are you begging the machine not to go with the “easy” answer.

While I think a lot of commentators are just happily dunking, there’s a serious problem with general applicability (which is what this tool is being sold for).

It’s not that the machine can’t be cajoled into returning the correct answer, it’s that most people will not use it like this, and will be incapable of judging that the result was incorrect.

In the interests of fairness and completeness, I will also include the commentator’s response to me below.

“I understand that. But the fact it’s capable of doing it means there are parameters that can be put in place in the future to account for contextual clues.

“I my. Opinion while. Clowning on stuff like that is fun, and I’m sorry if I’m ruining everyone’s fun, it also ends up weakening the overall anti ai stance which is how it’s negatively impacting people currently in ways improving ai is going to make worse.”

Either they had an aneurysm or they asked ChatGPT to simulate an aneurysm but I had trouble following that.

AI industry horrified to face largest copyright class action ever certified (Reddit)

One commentator summarized the article as follows,

“I decided to dig up graves and make beauty products out of bone powder. This is a fledgling industry so the courts must refuse the class lawsuit over “grave desecration” as it could kill the whole industry!”

Another wrote,

“Copyright is trash and I’m siding with the lesser evil on this one. Hope we can finally destroy it once and for all”

To which I responded as follows:

I agree that we should come up with a better compensation system than copyright, which has ended up consolidating the authority to grant permission to access large swaths of culture to a handful of large companies.

What sticks in my craw is that, when non-billionaire citizens were breaking copyright, they were fined into penury for it, even those who made no money off of the sharing.

Now we look to billionaire companies that have based their entire technology and business models on having violated copyright to a degree unthinkable 25 or 30 year ago and we’re supposed to cheer them on?

What are we hoping will happen? That the new “facts on the ground” copyright rules for Anthropic will somehow form a precedent that will apply to plebes who use BitTorrent? C’mon. That’s not going to happen.

We cannot look to the criminal elite to save us. They are only looking out for themselves and will chew our bones to powder for revenue.

How far can we push AI autonomy in code generation? by Birgitta Böckeler (MartinFowler.com)

“Even though technically the context window sizes of LLMs are increasing, LLM generation results still become more hit and miss the longer a session becomes. Many coding assistants now offer the ability to compress the context intermittently, but a common advice to coders using agents is still that they should restart coding sessions as frequently as possible.

“Secondly, it is a very established prompting practice is to assign roles and perspectives to LLMs to increase the quality of their results. We could take advantage of that as well with this separation into multiple agentic steps.”

This honestly sounds a lot like witchcraft, or a scam that blames the victim anytime the promised results don’t appear. It’s like the advice to frequently restart your computer or an app to get the best performance because everything leaks like a sieve.

“or bootstrapping the application, we used a shell script rather than having the LLM do this. After all, there is a CLI to create an up to date, idiomatically structured Spring Boot application, so why would we want AI to do this?

“The bootstrapping step was the only one where we used this technique, but it’s worth remembering that an agentic workflow like this by no means has to be entirely up to AI, we can mix and match with “proper software” wherever appropriate.”

I invite the author to use the term “deterministic” rather than “proper”. I like this term as it translates well to German (deterministisch) and highlights the main difference between these tools and LLMs.

“[…] if not specifically prompted, we found that the LLM frequently uses javax.persistence, which has been superseded by jakarta.persistence. Extrapolate that example to a large engineering organization that has a specific set of coding patterns, libraries, and idioms that they want to use consistently across all their codebases. Sample code snippets are a very effective way to communicate these patterns to the LLM, and ensure that it uses them in the generated code.”

You can’t ensure that it will use the patterns because the training data likely doesn’t contain them. The samples tend to encourage compliance with patterns but there is no guarantee, as you’d get with deterministic tools. It’s like having an unreliable coworker. The code reviews are going to take longer because, well, you never know.

This predilection on the part of LLMs for bog-standard and outdated coding standards is honestly one of the most concerning facets of the tools. It’s difficult enough to get people to start using safer, more secure, more maintainable, more legible features and patterns without having tools that generate swaths of code that doesn’t use them. People will go with the already-generated version and sweep all of the deficiencies under the carpet in the name of short-term efficiency.

“In an LLM’s first generation, it often doesn’t follow all of the instructions correctly, especially when there are a lot of them. However, when asked to review what it created, and how it matches the original instructions, it’s usually quite good at reasoning about the fidelity of its work, and can fix many of its own mistakes.”

Congratulations, I guess? This is still one of the places where I both worry about potential and also detect actual time-sinks. The LLM-based tool will not only put you primarily in code-review mode but will also often lead you down a primrose path with code that seems almost finished but which, in reality, requires so much editing, debugging, and fine-tuning that you would have ended up with a better product more quickly if you’d just written it youself, either with only deterministic tools and judicious copy/paste from existing examples (yes, you can do this too!) or with single-line coding assistance from the LLM.

“Think about how you can maximise the abstraction level of the code you are generating with AI, to take advantage of the speed and reliability of deterministic software as much as possible. For example, consider the abstraction level of the frameworks you’re using, and if you can generate a script or a codemod instead of letting AI do the full work itself.”

• Long feedback loops: You often have to wait 10-20 minutes until you see a prompt change earlier in the workflow lead to improvements or failures later in the workflow
• Keeping prompts consistent: Use of a reference application makes this slightly easier for the code examples, but it’s still a challenge. We often ended up having inconsistent instructions − and only realising that after another 20 minute run.
• Hard to eval: What is the definition of success of a generation cycle? The E2E test suite can give some high level confidence, but E2E tests usually cannot cover all test cases. And who reviews the generated tests, especially as the application gets larger?
• Debugging and traceability: It can be tedious to trace back a piece of code to its origin in the requirements and prompts. Again, this gets even trickier with larger requirements and larger workflows.
• Collaboration: All of the aforementioned challenges also make it harder to collaborate on the prompts and the workflow without getting into each other’s way, and without knowing if you broke something that your team mate put in place.

LLMs’ “simulated reasoning” abilities are a “brittle mirage,” researchers find by Kyle Orland (Ars Technica)

“In a recent pre-print paper, researchers from the University of Arizona summarize this existing work as “suggest[ing] that LLMs are not principled reasoners but rather sophisticated simulators of reasoning-like text.” To pull on that thread, the researchers created a carefully controlled LLM environment in an attempt to measure just how well chain-of-thought reasoning works when presented with “out of domain” logical problems that don’t match the specific logical patterns found in their training data.

“The results suggest that the seemingly large performance leaps made by chain-of-thought models are “largely a brittle mirage” that “become[s] fragile and prone to failure even under moderate distribution shifts,” the researchers write. “Rather than demonstrating a true understanding of text, CoT reasoning under task transformations appears to reflect a replication of patterns learned during training.””

“Relying on SFT to fix every [out of domain] failure is an unsustainable and reactive strategy that fails to address the core issue: the model’s lack of abstract reasoning capability.“

“Rather than showing the capability for generalized logical inference, these chain-of-thought models are “a sophisticated form of structured pattern matching” that “degrades significantly” when pushed even slightly outside of its training distribution, the researchers write. Further, the ability of these models to generate “fluent nonsense” creates “a false aura of dependability” that does not stand up to a careful audit.”

AI and The Modern Tower Of Babel by Indrajit Samarajiva (Indica)

“Today, you can put any question into AI and get an immediate piss-take. You then check a search engine to see if it’s a mistake, only to get the same AI result on the first page. This is followed by endless pages ‘optimized’ for the search engine, increasingly written by AI. How do you come to know anything within this system that inherently knows nothing, and doesn’t care anyways? They’re just calculating numbers to make other numbers go up. There’s no concept of a concept anywhere in this system. This information technology is just trying to appear smart to you, and you’re ignorant by definition. You’re the one asking questions in the first place!

“What we are rapidly reaching is an informational ouroboros, a snake eating its own tail. We fed all the world’s pre-2021 information into an information machine that now has to consume its own output, like a dog returning to its vomit. If you used Reddit or Stack Overflow pre-2021 you’re effectively a long-lost Vedic scholar to the future, there is no more purely human internet to be trained on. Models going forward will be trained on the output of other models and get increasingly detached from base reality. The things that are supposed to know things are eating their own offspring and can only become more inbred.”

“AI is just the latest brick baked into this tower, consuming the most water and energy to produce the most useless brainfarts. Socrates, in fact, predicted insufferable tech bros long ago, in his critique of writing (and reading). Channeling the Egyptian gods, he said (in Phaedrus),”

“The loyalty you feel to writing, as its originator, has just led you to tell me the opposite of its true effect. It will atrophy people’s memories. Trust in writing will make them remember things by relying on marks made by others, from outside themselves, not on their own inner resources, and so writing will make the things they have learnt disappear from their minds. Your invention is a potion for jogging the memory, not for remembering. You provide your students with the appearance of intelligence, not real intelligence. Because your students will be widely read, though without any contact with a teacher, they will seem to be men of wide knowledge, when they will usually be ignorant. And this spurious appearance of intelligence will make them difficult company.”

I do not agree with this sentiment, as it posits that auto-didacts cannot be anything but dilettantes, shadows taught by words rather than teachers. After centuries—millennia—there is no reason to believe that this is true. How the information is assimilated doesn’t seem to matter as much—though far, far fewer people are potential auto-didacts, so it’s a little bit true that just reading stuff is not as sure-fire way of learning as having a teacher drill it into your head—because, in the end, the information still ends up in your memory, as part of the knowledge to which you have more-or-less instant access.

The next stage of this was not, as many now think, AI, but search engines. We had this conversation over a dozen years ago already where people claimed to have knowledge or wisdom because they could just search for anything that they needed to know online. But that’s like saying that you could, of course, run a 5k because you could always just start training for it. You can’t run one right now and, similarly, you don’t know that information right now. The knowledge is not yours because you can’t draw on it quickly enough to participate in debate, in discussion, with others. You can offload information like the population of the country of Andorra but you can’t really offload the knowledge that Andorra exists at all, if it’s pertinent to the discussion.

The discussion of using AI to simulate knowledge is absolutely no different. It may differ somewhat in volume and accuracy but it’s no different in principle. I’m almost sick of arguing with people about it, people who just want to take the easy way to success. They should have it. This society tends to reward those who cheat the most, who provide the least value. Let them have that culture’s success. I clearly don’t deserve it. I’m not willing to bend to its will. I obstinately refuse to believe that everyone else is right that the world is a just place simply because it rewards them with, if not a free ride, then the privilege of multiple arbitrage opportunities that others mysteriously don’t have.

It is this culture that leads to people turning in sub-par and utterly useless “work” produced by a machine and claiming that it is there own. It is this culture that no longer cares about the opinion of any snob who might have a problem with that. This culture looks down its nose at anyone who’s not willing to scam others in order to get ahead. It is actively hostile toward those who don’t want to participate at all—either as scammer or sucker.

Why it’s a mistake to ask chatbots about their mistakes by Benj Edwards (Ars Technica)

“The first problem is conceptual: You’re not talking to a consistent personality, person, or entity when you interact with ChatGPT, Claude, Grok, or Replit. These names suggest individual agents with self-knowledge, but that’s an illusion created by the conversational interface. What you’re actually doing is guiding a statistical text generator to produce outputs based on your prompts.

“There is no consistent “ChatGPT” to interrogate about its mistakes, no singular “Grok” entity that can tell you why it failed, no fixed “Replit” persona that knows whether database rollbacks are possible. You’re interacting with a system that generates plausible-sounding text based on patterns in its training data (usually trained months or years ago), not an entity with genuine self-awareness or system knowledge that has been reading everything about itself and somehow remembering it.”

“When you ask an AI model what it can or cannot do, it generates responses based on patterns it has seen in training data about the known limitations of previous AI models—essentially providing educated guesses rather than factual self-assessment about the current model you’re interacting with.”

“The same model might confidently claim impossibility for tasks it can actually perform, or conversely, claim competence in areas where it consistently fails. In the Replit case, the AI’s assertion that rollbacks were impossible wasn’t based on actual knowledge of the system architecture—it was a plausible-sounding confabulation generated from training patterns.”

“Consider what happens when you ask an AI model why it made an error. The model will generate a plausible-sounding explanation because that’s what the pattern completion demands—there are plenty of examples of written explanations for mistakes on the Internet, after all. But the AI’s explanation is just another generated text, not a genuine analysis of what went wrong. It’s inventing a story that sounds reasonable, not accessing any kind of error log or internal state.”

“What they “know” only manifests as continuations of specific prompts. Different prompts act like different addresses, pointing to different—and sometimes contradictory—parts of their training data, stored as statistical weights in neural networks.”

“This creates a feedback loop where worried users asking “Did you just destroy everything?” are more likely to receive responses confirming their fears, not because the AI system has assessed the situation, but because it’s generating text that fits the emotional context of the prompt.

“A lifetime of hearing humans explain their actions and thought processes has led us to believe that these kinds of written explanations must have some level of self-knowledge behind them. That’s just not true with LLMs that are merely mimicking those kinds of text patterns to guess at their own capabilities and flaws.”

LLM Hallucination Seems Like a Very Big Problem, Not a Mere Speedbump by Freddie deBoer (Substack)

“Rather than report back that they haven’t found anything, they will simply hallucinate nonexistent sources; when the hallucination is pointed out, they’ll apologize, insist that the next source or quote they give me is verified and real, and hallucinate again. It’s funny, but also disturbing, because our economy currently relies on the AI bubble to avoid falling into a brutal recession.”

“It’s not merely that these systems hallucinate, it’s that they radically overstate how trustworthy their outputs are to a public that has been so bathed in AI hype, many can’t help but naively assume that the computer is right about everything. OpenAI says that GPT-5 cuts down on hallucination problems, but a) I don’t trust Taco Bell when they say that the new quesarito is cheesarific […]”

“If you have to have human verification for everything they do, you’re eliminating a vast portion of their comparative advantage; the whole point is to eliminate the human effort! And similarly, if you have to be some sort of prompt wizard to get reliable outputs from these systems, they become far, far less useful. Most people are not and will never be skilled at writing AI prompts. The whole idea was that these systems used natural language and could adapt to meet the user! Specialty tools for a small cadre of trained professionals are just a vastly different case than the promise of artificial intelligence that knows what the user wants better than the user does − socially, scientifically, communicatively, and especially financially.”

Replacing developers with GPUs by Oren Eini (Ayende)

“Proponents of AI coding have a tendency to talk about AI-generated code in the same way they treat compiled code. The machine code that the compiler generates is an artifact and is not something we generally care about. That is because the compiler is deterministic and repeatable.

“If two developers compile the same code on two different machines, they will end up with the same output. We even have a name for Reproducible Builds, which ensure that separate machines generate bit-for-bit identical output. Even when we don’t achieve that (getting to reproducible builds is a chore), the code is basically the same. The same code behaving differently after each compilation is a bug in the compiler, not something you accept.

“That isn’t the same with AI. Running the same prompt twice will generate different output, sometimes significantly so. Running a full agentic process to generate a non-trivial application will result in compounding changes to the end result.

“In other words, it isn’t that you can “program in English”, throw the prompts into source control, and treat the generated output as an artifact that you can regenerate at any time. That is why the generated source code needs to be checked into source control, reviewed, and generally maintained like manually written code.”

“The fact that I can do in an hour what used to take days or weeks is a powerful force multiplier. The point I’m trying to make in this post is that this isn’t a magic wand. There is also all the other stuff you need to do, and it isn’t really optional for production code.”

No, AI is not Making Engineers 10x as Productive by Colton Voege

“There are a few things you need to learn but they come quickly. You learn how to split up tasks into smaller pieces so the AI doesn’t lose its mind late in the context window. Tools like Claude Code can do a bit of this themselves, even, though not always reliably. And you learn to identify when the AI is too far off and it’s time to take the wheel. A competent engineer will figure this stuff out in less than a week of moderate AI usage. Further, if AI is about to get 2x, 10x, or 100x better at any minute (as everyone keeps saying it will), then any lessons about how to use it now are moot for the future.”

The final highlight is a good point. A lot of what we’re reading about these days is optimizations and guesswork based on the highly ephemeral, churning, bubbling forefront of the current technologies.

“100x productivity means you now do what used to be one year of work in two days. I shouldn’t even need to touch the ludicrousness of numbers at that scale.

“[…] When I have had engineers who were 10x as valuable as others it was primarily due to their ability to prevent unnecessary work. Talking a PM down from a task that was never feasible. Getting another engineer to not build that unnecessary microservice. Making developer experience investments that save everyone just a bit of time on every task. Documenting your work so that every future engineer can jump in faster. These things can add up over time to one engineer saving 10x the time company wide than what they took to build it.”

“[…] is a faster coder a better engineer? Yes, but it’s not the 10x difference maker and it’s hard to hold everything else constant. The more you focus on pumping out tasks as fast as possible the easier is to miss the important time savers that reduce total work.”

“In my experience, AI delivers rare, short bursts of 10-100x productivity. When I have AI write me a custom ESLint rule in a few minutes, which would have taken hours of documentation surfing and tutorials otherwise, that’s a genuine order of magnitude time and effort improvement. Moments like this do happen with AI.”

“The problem is that productivity does not scale. I don’t write more than one ESLint rule per year. This burst of productivity was enabled solely by the fact that I didn’t care about this code and wasn’t going to work to make it readable for the next engineer. If constantly writing ESLint rules became a core job requirement I’d sink the one-time cost to learn how ESLint internals work. After that, there simply wouldn’t be a big difference in the time it takes to vibe code a rule vs. write it myself, especially when you add in the extra time to make my code human readable for when I come back to this file in 6 months.”

“I think a lot of the more genuine 10x AI hype is coming from people who are simply in the honeymoon phase or haven’t sat down to actually consider what 10x improvement means mathematically. I wouldn’t be surprised to learn AI helps many engineers do certain tasks 20-50% faster, but the nature of software bottlenecks mean this doesn’t translate to a 20% productivity increase and certainly not a 10x increase.”

“My point is to say in the droll voice of your high school Econ 101 professor, “Incentives Matter”. If you are running an AI startup and every other AI startup is telling investors they are seeing 10x more productivity thanks to AI, the incentives are plain and simple: you should say the same publicly and privately. If your company is built on the back of AI, you are incentivized to sell AI as a miracle solution in every part of life. If you are an engineer and your boss asks you: Hey, you’re getting 10x the productivity thanks to AI, just like all the other engineers, right?

“You are strongly incentivized to say yes. And when every other engineer also says yes for the same reason, that CEO isn’t lying, they are just relaying what they heard. What I’d like to stress to those feeling anxiety like me is that this is nothing new. CEOs are not unbiased sources. Executives have been claiming that everything from Agile to Meyers-Briggs have unlocked limitless productivity. There will always be a new synergistic buzzword on LinkedIn, don’t let it get you down. In fact, stop scrolling LinkedIn at all. It’s a silly place.”

“Bootcamps and AI are just examples in a long series of poorly born out threats to commoditize the highly expensive, highly professionalized field of software engineering. They are rhetorical devices designed to imply precarity. Your boss can’t actually fire you and replace you with AI, but he can make you feel like he could, and maybe not ask for that raise.”

“I still felt some anxiety over the fact that I still didn’t enjoy using AI very much. Vibe coding is a complete bore once the magic wears off. Reading LLM generated code sucks. Asking it politely to use a not hallucinated library is painful. But what if I was, despite all that, 20% more productive vibe coding than regular coding? Would it be wrong for me to do “normal” coding if a higher output path is available?

“No. It’s okay to sacrifice some productivity to make work enjoyable. More than okay, it’s essential in our field. If you force yourself to work in a way you hate, you’re just going to burn out. Only so much of coding is writing code, the rest is solving problems, doing system design, reasoning about abstractions, and interfacing with other humans. You are better at all those things when you feel good.”

“Oh, and this exact argument works in reverse. If you feel good doing AI coding, just do it. If you feel so excited that you code more than ever before, that’s awesome. I want everyone to feel that way, regardless of how they get there.”

“There is no secret herbal medicine that prevents all disease sitting out in the open if you just follow the right Facebook groups. There is no AI coding revolution available if you just start vibing. You are not missing anything. Trust yourself. You are enough.

“Oh, and don’t scroll LinkedIn. Or Twitter. Ever.”

The Summer of Johann: prompt injections as far as the eye can see by Simon Willison

“Independent AI researcher Johann Rehberger (previously) has had an absurdly busy August. Under the heading The Month of AI Bugs he has been publishing one report per day across an array of different tools, all of which are vulnerable to various classic prompt injection problems. This is a fantastic and horrifying demonstration of how widespread and dangerous these vulnerabilities still are, almost three years after we first started talking about them.

“Johann’s published research in August so far covers ChatGPT, Codex, Anthropic MCPs, Cursor, Amp, Devin, OpenHands, Claude Code, GitHub Copilot and Google Jules.”

Programming

Single vs. Split Queries (Microsoft Learn)

Today I learned about “splitting” queries with Entity Framework (EF). A caller can anticipate a cartesian explosion that results from “including” or joining multiple 1-n relations in a query. The solution in EF is to manually determine when this might happen and instruct EF to issue multiple queries and stitch the results together.

The list of potential drawbacks at the end of the article is useful and interesting.

“While most databases guarantee data consistency for single queries, no such guarantees exist for multiple queries. If the database is updated concurrently when executing your queries, resulting data may not be consistent. You can mitigate it by wrapping the queries in a serializable or snapshot transaction, although doing so may create performance issues of its own.

“[…]

“While some databases allow consuming the results of multiple queries at the same time (SQL Server with MARS, Sqlite), most allow only a single query to be active at any given point. So all results from earlier queries must be buffered in your application’s memory before executing later queries, which leads to increased memory requirements.”

I thought this was interesting because, as far back as 2002, I was working on an ORM that would do this for you automatically, detecting when multiple 1-n relations would kill performance and selecting the ones to offload to different phases, automatically stitching the data together into the expected shape—as if it had queried everything in one go. The advantage here was that the performance-optimization was part of the query-planner instead of solely a part of the declarative query language.

Callers were free to override the automatic behavior with explicit phases but weren’t required to know about this in order to benefit from overall good performance, even for naively constructed queries. Similarly, a good query planner should be able to detect and ameliorate n+1 performance problems by executing a single query to get all referenced +1 relational objects in one go. This will also avoid querying the same object multiple times.

If I recall correctly, the planner would add “stub” placeholders for these single objects that would resolve at the end, when all references in the graph were known and the cache could be filled all at once, with a single query.

Zig’s Lovely Syntax by Alex Kladov (Matklad)

“Zig doesn’t have inheritance, mixins, argument-dependent lookup, extension functions, implicit or traits, so, if you see x.foo(), that foo is guaranteed to be a boring method declared on × type. Similarly, while ZIg has powerful comptime capabilities, it intentionally disallows declaring methods at compile time.”

“We have to specify type T when creating an instance of an ArrayList. But subsequently, when we are using the array list, we don’t have to specify the type parameter again, because the type of xs variable already closes over T. This is the major truth of object-orienting programming, the truth so profound that no one even notices it: in real code, 90% of functions are happiest as (non-virtual) methods. And, because of that, the annotation burden in real-world Zig programs is low.”

“The benefits to lightweight record literal syntax are huge, as they allow for some pretty nice APIs. In particular, you get named and default arguments for free:”

fn exec(argv: []const u8, options: struct {
    working_directory: ?[]const u8 = null
}) !void {
    // …
}
fn usage() !void {
    try exec(&.{ "git", "status"}, .{});
    try exec(&.{ "git", "status"}, .{
        .working_directory = "./src",
    });
}

“I don’t really miss the absence of named arguments in Rust, you can always design APIs without them. But they are free in Zig, so I use them liberally. Syntax wise, we get two features (calling functions and initializing objects) for the price of one!”

“Even with a small feature-set fixed, there’s still a lot of work to pick a good concrete syntax: unambiguous to parse, useful to grep, easy to read and not to painful to write. A smart thing is of course to steal and borrow solutions from other languages, not because of familiarity, but because the ruthless natural selection tends to weed out poor ideas. But there’s a lot of inertia in languages, so there’s no need to fear innovation. If an odd-looking syntax is actually good, people will take to it.”

Anatomy of a Web Component by David Bushell

“The constructor is the perfect place to call attachInternals.”

class Component extends HTMLElement {
  static tag = "component-one";
  static {
    customElements.define(Component.tag, Component);
  }
  #internals;
  constructor() {
    super();
    this.#internals = this.attachInternals();
  }
}

“The attached element internals provides access to a state set. State can be queried by a CSS selector.”

this.#internals.states.add("–large");
component-one:state(–large) {
  font-size: 2em;
}

“Using a – dashed ident prefix is not strictly required but CSS seems to be moving towards dashed idents. If you prefer not to use element internals then using data attributes can expose similar state to CSS.”

this.dataset.large = "";
component-one[data-large] {
  font-size: 2em;
}

“I assign internals to the private #internals field. This is only accessible inside the class and not as a property.”

“CSS has a special :defined pseudo-class that indicates if a custom element has been properly registered. This is useful to reduce FOUC like the elementB example above.”

class Component extends HTMLElement {
  static tag = "component-one";
  static {
    customElements.define(Component.tag, Component);
  }
  #controller;
  connectedCallback() {
    this.#controller = new AbortController();
    globalThis.addEventListener("resize", this.#onResize, {
      signal: this.#controller.signal
    });
    globalThis.addEventListener("scroll", (event) => {
      console.debug("scroll");
    }, {
      signal: this.#controller.signal
    });
  }
  disconnectedCallback() {
    this.#controller.abort();
  }
  #onResize = (event) => {
    console.debug("resize");
  }
}

“In the example above I’ve added an Abort Controller. This allows multiple event listeners to be removed in one action. It doesn’t matter if their callbacks can be referenced or not. Abort controller signals appear in other JavaScript APIs like fetch.”

“I’ve only touch on the basics. These ideas work for light DOM, shadow DOM, and declarative shadow DOM custom elements. For my use cases, I’ve found little need to use attributes. Attributes can be useful for declarative configuration if you’re sharing a web component for others to use.

“An event based architecture can allow a root component to use the reducer pattern common in JavaScript frameworks. Or you could use a state management library, subscribe to changes, and call a render method inside a component.

“JavaScript bros would be shocked how far custom elements can take you at a fraction of the cost. But they’re too busy gaslighting themselves into believing a VC funded nightmare is essential. We know better!”

A Friendly Introduction to SVG by Josh Comeau

“The viewBox attribute defines an internal coordinate system. When it’s provided, our <circle>s and <rect>s and <polygon>s will stop inheriting the raw pixel values of the DOM and instead use this internal coordinate system.

“The viewBox attribute takes four numbers, but really, we can think of it as two pairs of two numbers.

“The first two numbers allow us to change which part of the SVG we’re viewing.”

“In the demo above, our SVG is 300px by 300px. If we set the viewBox to “0 0 300 300”, we’ll have a perfect 1:1 ratio between the internal coordinate system and standard DOM coordinate system (pixels).

“But suppose we set the viewBox to “0 0 150 150”. The SVG is still 300px by 300px, but now it’s only displaying a 150×150 zone of our infinite SVG canvas. This effectively zooms in by 2x, doubling the size of the shapes inside our SVG.

“Keeping with the viewport analogy (since they really are quite similar), this is equivalent to using the browser zoom function (⌘ +) to zoom up to 200%. It doesn’t change the size of the browser window, but it scales everything up within the viewport to 2x its original size.”

“[…] because presentational SVG attributes like stroke-width are actually CSS properties, we can animate them like anything else in CSS!

“In the demo above, for example, I’m smoothly interpolating between the different stroke styles using basic CSS transitions”

It is absolutely amazing how easily you can declaratively specify vector graphics that zoom in and out and how you can animate multiple properties all at once, again with a simple, declarative syntax. The result is incredibly smooth and done entirely by the browser with no scripting. You can fine-tune the animation easing function, the delay, the duration, individually for each property of each element, or all at once. It’s incredible.

You can see this all in action by flipping through the variants in the Presentational Attributes Demo. With stroke-dashoffset, you can easily make “marching ants”. “[…] maybe the most famous trick is to create the illusion of an SVG drawing itself.” In order to simulate this effect, you have to have a single dash that is the length of the whole path.

“When we define pathLength, we’re essentially creating our own scale for this path. The polygon still has an actual path length of 763, but we’re redefining it as 100. The browser will do the work behind-the-scenes to scale everything up, but in our CSS, we can act like the full circumference is 100.”

Running .NET in the browser without Blazor by Andrew Lock (.NET Escapades)

partial class StopwatchSample
{
    private static Stopwatch stopwatch = new();

    public static void Start() => stopwatch.Start();
    public static void Render() => SetInnerText("#time", stopwatch.Elapsed.ToString(@"mm\:ss"));
    
    [JSImport("dom.setInnerText", "main.js")]
    internal static partial void SetInnerText(string selector, string content);

    [JSExport]
    internal static bool Toggle()
    {
        if (stopwatch.IsRunning)
        {
            stopwatch.Stop();
            return false;
        }
        else
        {
            stopwatch.Start();
            return true;
        }
    }

    [JSExport]
    internal static void Reset()
    {
        if (stopwatch.IsRunning)
            stopwatch.Restart();
        else
            stopwatch.Reset();

        Render();
    }

    [JSExport]
    internal static bool IsRunning() => stopwatch.IsRunning;
}

“As you might have guessed, [JSImport] and [JSExport] provide the means for interacting with JavaScript in the browser from your .NET Code. These attributes are used to drive two source generators, JSImportGenerator and JSExportGenerator respectively, both in Microsoft.Interop.JavaScript. As such, you can F12 to view the generated source in your IDE and see exactly what it’s doing.

“Ultimately it’s somewhat gnarly code to read, so I’m not going to go into more detail here, but it’s essentially just marshalling between the .NET (WASM) world and the JavaScript world, binding existing JavaScript functions (in the case of [JSImport]), or describing the shape of methods to expose for JavaScript to call.”

I mention this mostly to note that it reminds me very much of the platform I helped write for a fintech company that built the mobile apps for many, many banks in Switzerland about ten years ago. The interaction between the web-browser control and the native code looked very similar to what .NET offers now. Using source-generators is a nice addition, of course, which takes a lot of dynamic handling out of these calls but it is, in principle, no different.

The framework I helped build didn’t have source generators and targeted two native languages: Swift for iOS and Java for Android.

“Out of interest I checked the published size of this sample app (in release mode) and it looks roughly like the following:”

• 6.8MB uncompressed
• 2.5MB compressed (gzip)
• 2.0MB compressed (brotli)

“That includes all the files, including the .NET runtime, so that’s not bad. The runtime is obviously heavily trimmed to reach these sizes”

To a web developer, 2.0MB does not look “heavily trimmed” but since that’s the whole .NET runtime, it actually is quite small. This is the price you pay in order to write code for the browser in C#/IL rather than in JS or WASM directly.

Better CSS layouts: Time.com Hero Section by Ahmad Shaheed

“One idea is to introduce a –ratio CSS variable. For an article that is featured, we can use a higher ratio.”

/* Default ratio */
.layout {
  –ratio: 1.5;
}

/* A specific item that is featured */
.layout > .layoutItem {
  –featured: true;
  –ratio: 2;
}

.cardTitle {
  font-size: clamp(0.8rem, 0.7rem + var(–ratio) * 1cqw, 1.5rem);
}

“Here is how it should look. The font size of the other articles is now smaller.”

Why Akka.Streams.Kafka is the Best Kafka Client for .NET by Aaron Stannard (Petabridge)

“One of the biggest complaints about Confluent.Kafka is the lack of backpressure support. Once you start polling for messages, you’re expected to handle whatever throughput Kafka throws at you.

“Akka.Streams.Kafka automatically handles this through its reactive streams implementation. Here’s how it works:

“If your downstream processing (like database writes) can’t keep up, the stream automatically pauses polling from Kafka until the backlog clears. No manual semaphores or thread pool management required.”

“Behind the scenes, Akka.Streams.Kafka automatically:”

1. Invalidates in-flight messages from revoked partitions (as long as they haven’t been emitted to your processing code yet)
2. Commits outstanding offsets from revoked partitions immediately during rebalancing
3. Coordinates with the stream backpressure system to ensure clean handovers
4. Prevents race conditions between message processing and partition revocation

“You don’t write a single line of rebalancing code, yet you get more sophisticated behavior than most manual implementations provide.”

“It’s built on battle-tested foundations (Confluent.Kafka + librdkafka) so you get enterprise-grade reliability with startup-friendly developer experience.”

“The full demo code includes Docker Compose setup for Kafka and runnable examples of both approaches.”