↵ select ↓ ↑ navigate esc close

Daily Briefing: November 11, 2026, morning

Daily Briefing ·

Listen: https://serve.buildtall.systems/07a9132a0b77a2192aaf6967e357a4de82e61438f6b3a013649ef047ef87edbf.mp3


The briefing for November 11, 2026, morning covers fourteen articles, spanning science, spaceflight, technology, climate, economics, and cultural commentary.

Several articles concern science and space. The Nautilus piece titled Famous Fossil Could be Rosetta Stone for Hidden Hominin Diversity asks what the Taung child would grow up to be, but the corpus provides no further substance. The Nautilus article titled Lean Muscle Mass Isn’t a Cure-All for Metabolic Health reports that a new study is complicating the picture, but offers no additional detail.

The NASASpaceFlight.com article titled Isaacman lays out SLS path to the 2030s reports NASA Administrator Jared Isaacman's explanation of decisions to standardize the Space Launch System, move toward a stretched Centaur V upper stage, and stop work on Mobile Launcher 2. Isaacman links these choices to higher flight rates and lower costs, and says industry responses to NASA's search for future commercial lunar transportation pointed to the 2030s rather than an immediate replacement. The article notes that Artemis IV was previously intended to introduce Block 1B and the Exploration Upper Stage, but NASA moved toward a near-Block 1 configuration using Centaur V, which is already in production for United Launch Alliance's Vulcan rocket. The stretched version planned for NASA remains in development, and Isaacman says NASA originally intended to use the existing Centaur configuration but pursued the higher-performance version to move Orion closer to the Moon and simplify the landers' task. He expects the stretched stage to fly on Vulcan for Department of Defense missions as well. Material testing is progressing, and NASA and ULA are carrying out development activities in parallel, with NASA helping address production constraints. The first mission to use the new stage remains undecided; Artemis III's revised low-Earth-orbit mission will not need an upper stage, leaving the final Interim Cryogenic Propulsion Stage available for Artemis IV. Isaacman says modifications to Mobile Launcher 1 are being designed and hardware ordered, with deferrals as long as practical. On Mobile Launcher 2, Isaacman acknowledges disappointment but says the visible progress did not mean it was close to operational readiness, and that launching on an eight-month cadence does not require a second mobile launcher. He says NASA should concentrate on getting very good at operating a single mobile launcher, and some components, along with hardware from the canceled Exploration Upper Stage, have been provided as government-furnished equipment to help Blue Origin with its pad rebuild and landers, while also giving younger engineers hands-on experience. He describes reusing ML-2 as highly unlikely. NASA's goal remains to fly crews at least once a year, ideally with less than twelve months between missions, and improving production and ground operations could preserve the possibility of two launches in 2028. The article notes that NASA's request for information on commercial lunar transportation, issued in March with responses due in April, sought industry input but no end-to-end solution was offered for 2028 or 2029. Isaacman does not describe the transition as abrupt; Orion is likely to remain part of the architecture into the 2030s, potentially using other launch vehicles and transfer stages.

A series of articles concern technology, corporate issues, and programming. The Slashdot article titled Fireworks AI Confirms Wider Breach Affecting Notion AI Chats reports that the October breach at Fireworks AI was broader than initially disclosed, with attackers gaining access to customer account and billing information as well as some datasets, incomplete model weights and secrets. Notion disclosed that fewer than one percent of Notion workspaces were impacted, and upon analysis of new data, Notion found that the breach included Notion AI chat transcripts used for the Insights feature, which for the vast majority of impacted workspaces were unattributable, meaning they were not labeled with who wrote them or which workspace they came from, and the transcripts did not contain tool call results. The article cites The Hacker News, which reported that Fireworks initially said on October 10 it detected unauthorized use of internal credentials allowing access to environment variables in evaluation jobs in a limited number of customer accounts, with activity detected on October 6. The latest bulletin reveals attackers reached the development environment via an unnamed third-party cloud provider hosting research and development machines on October 23, and on October 24 the unauthorized third party used it to reach Fireworks development machines and run an unauthorized workload that led to additional customer data impact. As of October 25, no further evidence of successful unauthorized access has been detected. The breach provided access to resource metadata, billing information, and model weights and datasets. Fireworks AI says payment card numbers are not stored and were not impacted, and it is alerting customers whose data was impacted.

The Slashdot article titled Google Fi Brings Back Network Switching With 'Secondary' Partner reports that Google Fi is bringing back network switching through an unnamed secondary 5G carrier. Most users will be on the primary network, but Google Fi will automatically switch users to the secondary network if it meaningfully improves performance based on smart thresholds. At launch there is no manual switching. The report says that when Project Fi first launched, it included Sprint and T-Mobile, and that went away when T-Mobile acquired Sprint in 2020. Google is not naming the second network but did share that Ultra Wideband is supported, and the service switches between two of America's fastest 5G networks. Fi Wireless notes that users have top priority so data is never deprioritized on the primary network, and there is highest QCI priority on the backup network, meaning subscribers are among the last to be slowed during congestion. This is available internationally when local partner networks are available with Flexible and Unlimited Premium plans. Network switching will first be supported on Pixel 9 and newer devices that have a Google Fi eSIM and no other SIM active, using Android's Multiple Enabled Profiles eSIM technology that allows two carrier profiles on a single chip. The updates start rolling out today with gradual availability.

The TechCrunch article titled X launches its own clipping feature to spur influencer engagement says X is jumping into clipping, the practice of cutting long videos into short, shareable clips that can go viral, aimed at influencer engagement.

The TechCrunch article titled Nvidia’s crisis has corporate governance experts reaching for the popcorn reports that governance experts say nobody has seen anything like what is unfolding right now.

The post by Eli Bendersky titled Monads in Lean demonstrates how to use monads in the Lean programming language, specifically Lean 4. It begins with a motivating example of evaluating numerical expressions supporting numbers, addition, and division, where division by zero must be reported as a failure using the Option type. The article defines a safeDiv function returning an Option, and a first eval function that manually checks for none values at each step, which is tedious. It then notes a common pattern: examine an Option, return none if none, otherwise continue with the result, and extracts a helper called withSome with signature that takes an Option and a continuation and returns an Option. Using this helper and an infix operator, the article shows a cleaner evaluator, and says this has essentially invented monads. Next, the article shows that the Monad type class in Lean, in simplified form, has pure and bind methods, and the Option instance defines pure as some and bind as withSome. The standard bind operator is available, and the evaluator can be written with pure and bind, or using do notation as syntactic sugar. The article then discusses monads for modeling side effects: pure functions are easier to reason about, but side effects are inescapable, and monads encapsulate them. The bind method takes a wrapped value and pushes it into the next computation, chaining binds sequences computations while tracking side effects, decoupling what computations do from how they are sequenced. The article introduces the Except type as an alternative to Option for error values, and writes a generic evalM function that works for any Monad, taking a division function as argument, so the same evaluator can be used with Option or Except. Next, the article presents the writer monad with a LoggedOption structure that carries an Option value and a log of strings, and implements Monad by wrapping pure with an empty log and bind by appending logs. This allows an evaluator that logs operations without explicit log management. Then the article covers the state monad, defining State as a function from initial state to a pair of result and updated state, and demonstrates an evaluator supporting variable assignment and lookup using an association list as state. The evaluator uses store and load expressions, and operations like num and add just do their work while div propagates none; the monad hides the bookkeeping. The article notes that in a language with mutable state, side effects are easy, but in functional languages monads carry state explicitly with elegance. Finally, the article discusses the IO monad, showing a greet program using do notation, and notes that IO is a state monad where the state represents the outside world. It shows an evaluator that logs to standard out using OptionT, a monad transformer composing Option with IO. The article concludes with monad laws in Lean: left identity (pure followed by bind equals f), right identity (bind with pure equals original), and associativity, and demonstrates that Option obeys these laws. Lean does not enforce laws for Monad but has LawfulMonad requiring proofs, useful for proving theorems about programs.

The post by Richard Seroter titled Daily Reading List – November 11, 2026 (#907) begins with a note about a pair of customer chats in Japan about AI-first teams, questions about gathering requirements for agents, and possible future blog post. The reading list includes a Gemini 4 Carbon Developer Guide with callouts for when to use the frontier model and when Argon is right; a Gemini 4 Carbon and /swarm blog on Antigravity for teams of agents with orchestration and visualization out of the box; a Harvard Business Review article on why leaders need to rethink AI oversight, arguing leaders respond with more oversight and governance but must provide trust and agency too; a Claude announcement for Whiteboards, Sheets, and Data, noting that model API consumption alone cannot sustain revenue and growth, so Anthropic is shifting to higher-margin office productivity; a CodeMender guide for pull requests, emphasizing that code generation should not skimp on security practices; a Lyft engineering blog on hiring junior engineers, noting many companies hire for mindset not just hard skills, but candidates must prove they have created production-quality code; a Google Cloud piece on bootstrapping an agent database instance in two seconds, suggesting a database that scales to zero and handles bursts; an InfoWorld article on panic over cloud capacity, saying clouds are not full and certain AI compute is oversold or pre-committed but business is open; a JetBrains report on the state of Go in 2026, describing it as healthy with experienced developers, modern AI tools, and reliance on the standard library; and a Google research blog introducing the Multimodal Knowledge Agent benchmark, which appears useful because quality and cost matter.

Two articles concern global economics and energy. The Slashdot article titled US and Rest of World Must Follow China's Lead On Green Energy, Says John Kerry reports on John Kerry's comments to The Guardian. Kerry says China is leading the world in moving away from fossil fuels and the rest of the globe, including the US, must follow. At Cop31 in Turkey, there is speculation about whether China will step into a leadership role it has long avoided. China's latest five-year plan adopted in March 2026 pledges to actively participate in and lead global climate governance. In 2025, clean-tech exports surpassed two hundred twenty billion dollars, and China accounts for about eighty-five percent of solar and eighty percent of lithium-ion battery supply chain production capacity, with nearly seventy-five percent of electric cars manufactured there. Kerry says the only country that has hugely accelerated is China, while the rest of the world is stumbling. He notes China realized it had too much dependency on oil and gas from other places and decided to be independent; other countries since the war in Iran and the interruption of the Strait of Hormuz have made similar decisions, leading to a massive uptick in renewables. For autocratic states, seeing China's economic benefits will spur adoption of low-carbon growth, because they can read a spreadsheet and see where costs are exorbitant in fossil fuels. Kerry argues renewables together with storage and transitional gas are cheaper and more effective than fossil fuels, and the marketplace has made its decision, moving to electrons and electrostate status, which is unstoppable.

The Noahpinion post titled India is quietly doing great argues that while the author has become bearish on China's continued economic and technological outperformance, he is more bullish on India than in a while. The reason is that the two countries are being asked to do different things: China's future success is measured by eclipsing the combined economies of the US and its allies, seizing the lead in AI, reaching Taiwan or Korea's income level, or dominating Asia, which the author is increasingly skeptical about; India's job is simpler, just to escape poverty and become an upper middle income country, which it can do. In GDP and Human Development Index terms, India is around where China was in the late 2000s. The challenge is to make the leap China made in the 2010s. The article notes that extreme poverty in India has mostly been conquered over the last two decades, lifting a full tenth of humanity from deprivation, and unlike China, India reduced poverty while also reducing inequality. The World Bank attributes this to strong consumption growth, rising employment especially among women, and expanded social transfers including food subsidies. The article says a strong welfare state is important, but economic growth takes the lion's share of credit. India has been growing faster than China since the pandemic, at about six and a half percent, which represents almost a doubling of living standards every decade. Bloomberg expects growth to accelerate to eight percent by the end of the decade. The article highlights government investment increasing by a factor of four since before the pandemic, with highway spending up fivefold and railway spending up eightfold, and notes this is correcting chronic deficiencies and changing the physical look of the country, including bullet trains and impressively clean and quiet subways. Manufacturing is growing alongside services: output was nine percent higher year on year, electronics exports up ninety percent with American-bound iPhones a large share, and engineering exports are diverse, led by automobiles, industrial machinery, iron and steel, and electrical machinery. This happens despite China's attempts to throttle Indian industrialization by denying technological exports and factory expertise. The article notes challenges: private investment and foreign direct investment will have to rise, and structural issues like bad rural education, female underemployment, and incomplete urbanization need correction. But for now, India looks like a big economic success story, perhaps the brightest in the world.

Several articles concern culture, secrecy, and intellectual discourse. The Slashdot article titled M John Harrison Becomes First Sci-Fi Author To Win Booker Prize reports that eighty-one-year-old M John Harrison has won the Booker prize for The End of Everything, a postapocalyptic novel set on a near-future Kent coast, becoming the oldest winner in the award's history. Gaby Wood, chief executive of the Booker Prize Foundation, says while plenty of previous winners had dabbled in science fiction, Harrison is the first Booker winner to have emerged from the bona fide SF world. Harrison, who has been publishing for six decades, overtakes Margaret Atwood, who was seventy-nine when she shared the prize in 2019. He received the fifty thousand pound award and trophy at a ceremony in London. The novel is set a few decades from now, years after a disaster nobody can properly account for, and follows Phillip Tennent and his ageing aunt Marnie trying to survive alongside the mysterious iGhetti, vast shape-changing beings whose origins and intentions are never explained. At one hundred ninety-eight pages, it was the shortest book on the shortlist. Mary Beard, chair of the judges, says the decision was unanimous and the book turns you upside down, raises a wry smile, and leaves a twinge of optimism. She says one main point is that if or when such an apocalypse comes it will be incomprehensible, and the old rules won't matter; this is the politics of now seen through the lens of the future.

The Marginal REVOLUTION post titled Spying makes a comeback points to a Free Press article and excerpts an argument that a recent story of a wealthy, powerful man making an irresponsible decision is also a harbinger of the erosion of privacy, the acceleration of espionage, and the increasing value of secrets. More time and resources will be devoted to uncovering secrets, with honeypotting only part of a larger picture that includes more spying, more private surveillance, more attempts to record and search conversations using AI, more hacks of phones and computers, and more attempts to use AI to figure out details of people's lives. Life will feel much more cloak-and-dagger, especially for the wealthy and influential, who are likely to take ever greater care to minimize public scrutiny, protect digital privacy, prevent journalists from going through trash, travel incognito, and guard their language in private conversations. The post notes that NDAs for hookups and even dinner dates are now a thing. The author says this is worth pondering because in the pending world secrets will be more valuable than before; the super-wealthy and their companies can be decisive actors, there are more prediction markets than before, and AI is equalizing the value of high-quality but publicly available information.

The Astral Codex Ten article titled Escaping The “Outside Criticism” Trap begins with a description of a rhetorical trap for discrediting anyone: call them an arrogant fanatic with a totalizing ideology, and when they present evidence of engaging with critics, say they cherry-picked critics who operate within their worldview, so they exclude genuine critics. This always works because everyone has people close to their ideology whom they respect and people far away whom they don't. The article illustrates with five circles of critics for a psychiatrist prescribing escitalopram: the Colleague who suggests bupropion, the Biopsychosocialist who suggests diet and lifestyle, the Freudian who suggests repression analysis, the Zealot who suggests religion, and the Scientologist. In practice, the author devotes a lot of time to the first two, a little to the Freudian, and very little to the last two. If portrayed as blinkered, pointing out the engagement with the first two leads to the accusation that those are loyal opposition, and the author admits to not engaging with Freudians, zealots, and Scientologists because they are wrong, which is then used as evidence of refusing outside critics. The article calls this a Kafkatrap, where any denial is further evidence of guilt. It notes this argument was common in the first Social Justice Wars and is used against rationalists and effective altruists, demanding engagement with increasingly hostile critics. The author declines to abandon unpopular beliefs in favor of popular ones or to make the attackers high-status leaders. Part two asks how to address the epistemic problem. One tempting approach is to use objective evidence of which critics are worth engaging, but this has two flaws: perception of powerful and popular is influenced by current beliefs, and it ensures movement toward Power, which the article likens to demons whispering. Instead, the author proposes a power law: halve the amount of energy devoted to each step away from current beliefs. Spend most time on core, half on near periphery like biopsychosocial solutions, half that on far periphery like Freudianism, and half that on obviously crazy ideas. The key idea is to read critics a little but not too much, to check whether they are as silly as thought. The trap will insist that a few books are not enough and years of immersion are required, but there is no way to spend years on every idea, so concentrating on correct-seeming ideas and devoting progressively less energy to more outlandish ones is the way to go. The article lists many book reviews as examples of investigating views very different from the author's, including Just Giving, The Body Keeps The Score, A Future For Socialism, Inventing The Future, Madness And Civilization, Capital In The Twenty-First Century, Seeing Like A State, Unlearn Your Pain, My Brother Ron, Chronicles Of Wasted Time, What's Wrong With The World, writings on neoreaction, Singer on Marx, San Fransicko, The Others Within Us, I Saw Satan Fall Like Lightning, America Against America, The Cult Of Smart, Sadly Porn, Mussolini On Fascism, and the Frankfurt School.

The window is dominated by detailed treatments of monads in Lean, India's economic trajectory, and the nature of intellectual criticism, alongside technical and scientific reporting.

  1. After 7 Months, Glasswing Overwelms Maintainers with Underwhelming Findings
  2. Fireworks AI Confirms Wider Breach Affecting Notion AI Chats
  3. Famous Fossil Could be Rosetta Stone for Hidden Hominin Diversity
  4. WE'VE BEEN COOKING!! Come see what all the fuss is about inside Bitcoin Quest :)
  5. Inflation is Still Always a Monetary Phenomenon (X, Lars Christensen)
  6. Google Fi Brings Back Network Switching With 'Secondary' Partner
  7. Lean Muscle Mass Isn’t a Cure-All for Metabolic Health
  8. X launches its own clipping feature to spur influencer engagement
  9. A great old-timey game-programming hack (2013)
  10. Monads in Lean
  11. M John Harrison Becomes First Sci-Fi Author To Win Booker Prize
  12. Isaacman lays out SLS path to the 2030s
  13. Protecting SSH Keys with macOS Secure Enclave
  14. Spying makes a comeback
  15. US and Rest of World Must Follow China's Lead On Green Energy, Says John Kerry
  16. Escaping The “Outside Criticism” Trap
  17. Nvidia’s crisis has corporate governance experts reaching for the popcorn
  18. India is quietly doing great
  19. Daily Reading List – November 11, 2026 (#907)