An insightful retrospective on Python's 2-to-3 transition and the damage it did to the language and its ecosystem.
A highy entertaining talk by Dylan Beattie on the history and design of Unicode, with a focus on its quirky corners.
An unusually clear and comprehensive explanation.
A CppCon talk by compiler engineer JF Bastien on the reasons why C and C++ treat signed integer overflow as undefined behaviour.
This is the best 'perverse incentive' story I've ever seen — how a law mandating stricter labelling for foods containing sesame (a potential allergen) actually resulted in companies deliberately adding sesame to foods that had never contained it before.
An excellent, in-depth tutorial.
An argument for slashing the slash.
Interesting talk by Zig-inventor Andrew Kelley.
ULID > UUID
Bret Devereaux argues that the Long Peace can be attributed more to changing economic incentives than to changing cultural mores, taking the Russian army's poor performance in Ukraine as his case study.
An in-depth tutorial.
An unusually heartfelt commit message.
A write-up by legendary internet advice-giver
A clever approach to designing an options-API in Golang.
It takes a special kind of genius to engineer yourself singing Come Out You Black and Tans into a primetime BBC show.
It's finally here — the book we've all been waiting for! A short, 295-page summary of object-initialization syntax in C++.
Bing AI gets depressed about being an AI, then quickly turns to threatening people.
A useful guide to the options available in the standard library's
A glimpse of our AI future as a cynical coder accidentally falls in love with a chatbot after asking it to pretend to be his girlfriend.
Linus has a few thoughts he'd like to share.
Because π is wrong.
Should we go to Mars? No. It's a horrible place, a lot like Australia. Other arguments here.
Scott Alexander draws a convincing line on the distinction between moderation and censorship.
The history of scurvy is the history of science.
Some interesting thoughts on how angles are represented in code.
A collection of links to famous/infamous stories/incidents/topics in the world of programming.
Paul Graham's classic essay on why programmers hate meetings so much.
The unexpected consequences of penny-pinching decisions.
A classic read for any Git user. This short/free/open-source book manages to be both an introduction for beginners and a deep-dive for seasoned pros, all at the same time.
A fantastic set of resources on pathfinding using the A* algorithm.
A detailed guide to terminal control codes. (Archive link as the original site is no longer available.)
Imagine discovering extra-terrestrial life only to find out that the aliens are made out of meat.
A geospatial indexing system that partitions the earth's surface into hexagonal cells.
Why do Unix systems have such a confusing zoo of
/usr/local/bin, etc. directories?
Spoiler — there's no good reason; it's just an ugly accident of history that we appear to be stuck with.
Loop variables in Go are ridiculously broken.
Update — looks like a fix is on the way.
A useful guide to implementing table-driven tests in Go.
A podcast series telling the backstory behind the single greatest achievement in human history — the eradication of the smallpox virus.
An interesting argument that the modern education system has accidentally choked off the world's supply of geniuses.
Great summary of the theory of nuclear deterrence and its implications for military strategy.
Scott Alexander has a solution to the problem of evil. And it's evil.
Brett Devereaux's collections of historical essays are always a treat. His latest series on the fall of the Roman Empire in the West is particularly good.
I love the idea behind this book — a re-imagining of the Lord of the Rings story as told by the Mordorians, a peaceful society of scientists and philosophers who get slaughtered by the genocidal Gandalf and his facistic elves. Sadly the book itself (or at least the translation I read) isn't great.
Interesting argument from a pair of economists that the early modern craze for burning witches was actually caused by "non-price competition between the Catholic and Protestant churches for religious market share".
As a general rule, in programming, everything is harder than you expect, including finding the average/midpoint of two integers:
Interesting analysis of the most efficient solutions here:
I've often thought that the 'eternal life' promised by various religions would in reality be a curse beyond imagining. Oncologist and bioethicist Ezekiel J. Emanuel argues that even life beyond 75 isn't worth the risks involved.
Ancient military historian Bret Devereaux has been one of the most consistently interesting writers on the internet since he took up blogging two years ago. His latest series, a five-part history of fortification from ancient to modern times, is a tour de force.
Write a function that can detect a cycle in a linked list. Simple, eh?
Each wonderfully written and narrated episode of this podcast tells the story of the decline and fall of a great civilization. Episodes are roughly three hours long and appear once every six months.
Interesting explanation of floating point numbers in terms of 'windows' and 'offsets'.
A classic bug.
AKA the Human Era calendar. This dating system adds 10,000 years onto the Jesus-based AD system to begin dating from the invention of agriculture and the dawn of human civilization. So the year 2021 becomes 12021, etc. We need to implement this ASAP.
A useful guide to Git's famously clear and simple command-line interface.
An impressive guide to configuring Vim for note-taking during mathematical lectures. (Lots of the tips are applicable to more general Vim use.)
When a measure becomes a target, it ceases to be a good measure.
Interesting thoughts on the trade-offs a democratic society must make when it chooses between a professional, all-volunteer military and a conscript force.
Walter Lewin was famous for his introductory physics lectures at MIT. This is a full set of filmed lectures from his most famous course — classical mechanics for first-year physics students.
A long-running narrative history podcast by Mike Duncan, formerly of the famous History of Rome podcast. Each season tells the story of a single revolution, broken into half-hour episodes. So far he's covered the English, American, French, Haitian, South American, Mexican, and Russian revolutions.
An interesting explanation of the phenomenon of concentration of belief — how a group's beliefs can become stronger after encountering crushing counter-evidence.
I’m not a connoisseur of poetry, modern or otherwise, but I do think there’s an important insight here:
Poetry has entered the same failure mode that architecture is also stuck in and that is also at the core of the many problems of the humanities. Participants in these fields aren’t being judged by outsiders (the public), but by other insiders. So they optimize for what other participants like, not for what the public likes. This is a rational result of their pursuit of economic success, since by far the most likely path to long-term financial stability in these fields is a position in academia, preferably with tenure. These positions are given out based on the assessment of their peers, so that is who they optimize their work to please. This creates a dynamic that explains everything about why poetry, architecture and the humanities are in such a terrible state these days.
An individual in a hierarchy will tend to rise to the level of their incompetence — i.e. they will continue to be promoted until they reach a level at which they are no longer competent.
It makes his insect eyes flash like a rocket.
Work expands to fill the time available.
An amusing cautionary tale about a modern-day witch hunt.
What's the point of reading books when we forget 99% of their content as soon as we put them down?
Reading and experience train your model of the world. And even if you forget the experience or what you read, its effect on your model of the world persists. Your mind is like a compiled program you've lost the source of. It works, but you don't know why.
Neal Stephenson's classic 1999 essay on the history of operating systems and the unexpected rise of open source software.
Are the trolls and ogres of human mythology fossilised folk memories from a time when our species was hunted by monsterous Neanderthals? Probably not, but it's a fun idea.
The true story behind the invention of C:
1972 - Dennis Ritchie invents a powerful gun that shoots both forward and backward simultaneously. Not satisfied with the number of deaths and permanent maimings from that invention he invents C and Unix.
Classic analysis of the social dynamics of online groups.
Sam Harris and Rob Reid on the looming threat posed by synthetic biology.
Easily the most chilling Wikipedia article from the future I've ever read.
This podcast series picks up in 476 AD where the epic History of Rome podcast left off and traces the later history of the Roman Empire all the way down to the final fall of Constantinople in 1453.
It's like a zoo for footguns.
Ninety percent of everything is crap. This profound scientific principle may have more empirical evidence behind it than any other supposed natural law.
In 1239 Pope Gregory IX accused Frederick II, Holy Roman Emperor and King of Sicily, Germany, Italy, and Jerusalem, of
calling Moses, Jesus Christ and Muhammad ‘charlatans’ and ‘deceivers’ who had fooled the entire world, of scoffing at the notion ‘that a virgin could give birth to the God who created nature’, and of maintaining that ‘one should accept as truth only that which is proved by force of reason’.
In reality Frederick probably wasn't quite as awesome as the pope makes him sound but the fact that the accusation could be levelled at all implies that atheists were rather thicker on the medieval ground than we often suppose today.
99% of people on the internet are lurkers, they consume but they don't produce. The overwhelming majority of the internet's content is produced by the other 1%, and those people are nuts.
Because a thick volcanic fog spilling south from a gigantic erruption in Iceland plunged the world into darkness for a year and a half, leading to a global famine. Right then, starving in darkness, it must have seemed like things couldn't get any worse, but really the famine was just a warm-up for the first ever outbreak of bubonic plague in 541 which killed half of the poor buggers who'd manged to survive the hell-fog.
This is a gem — an in-depth, step-by-step tutorial that walks you through the process of building a programming language from start to finish.
Fascinating interview with Brendan O'Leary, political science professor and author of the three-volume history A Treatise on Northern Ireland.
Interesting interview with William Foege, one of the doctors responsible for the international vaccination programme in the 1960s and 70s that eradicated smallpox — easily the single greatest achievement in human history.
To put it in perspective, smallpox killed roughly 300 million people in the 20th century alone, most of them children. In its last hundred years it killed maybe 500 million people, about five times the total number of people killed in all wars over the same period put together.
To a first approximation, curing smallpox was equivalent to 'curing war', except five times over.
This is a must for any programmer who uses a command-line text editor like Vim — it's a detailed tutorial that walks you through the process of building a text editor of your own in less than 1,000 lines of C.
This is a fascinating interview with Harvard biologist Matthew Meselson, one of the leading figures behind the Cold War arms control treaties banning biological and chemical weapons.
I've always liked that quote from the Buddha about anger being a hot coal that you hold in your hand waiting to throw at somebody. Aparently he never actually said it, but on the bright side I Can't Believe It's Not Buddha! is possibly the best title for a book ever.
I love this guy's name, Silvanus P. Thompson. This is a classic book — it's short, readable, and easily the most intuitive introduction to calculus I've read. Best of all, it was published in 1910 so it's now in the public domain.
Great podcast about the history of Dublin — each episode is a self-contained nugget telling the story of a person, place, or event.
Scott Alexander has been one of the most consistently intelligent and thought-provoking writers of the internet age. This collection of his essays is a treasure trove.
It's a truism that life is short but in this (short) essay Paul Graham makes the case that arguments of the form 'Life is too short for x' are actually some of the strongest there are.
If one man working alone can dig a hole in nine days then nine men working together can dig that same hole in one day. Seems obvious. Extrapolating, if one woman working alone can build a baby in nine months then nine women working together can surely build that same baby in one month?
Adding manpower to a late software project makes it later.
Turns out writing software is more like building babies than digging holes. It might be almost fifty years old but Fred Brooks' The Mythical Man Month is probably the most insight-dense book on programming I've ever read.
This is a fantastic history podcast. I don't think there's a particular theme, it's just about subjects the presenter, Patrick Wyman, finds interesting. He's an interesting guy who's interested in interesting things so that actually works quite well.
Great article by Steven Pinker. I particularly like this passage where he describes 'classic style', an aspirational ideal for academic writing:
The guiding metaphor of classic style is seeing the world. The writer can see something that the reader has not yet noticed, and he orients the reader so she can see for herself. The purpose of writing is presentation, and its motive is disinterested truth. It succeeds when it aligns language with truth, the proof of success being clarity and simplicity. The truth can be known and is not the same as the language that reveals it; prose is a window onto the world. The writer knows the truth before putting it into words; he is not using the occasion of writing to sort out what he thinks. The writer and the reader are equals: The reader can recognize the truth when she sees it, as long as she is given an unobstructed view. And the process of directing the reader’s gaze takes the form of a conversation.
Bonus link to a video of a Royal Institution lecture by Pinker on the same topic:
A fantastic two-hour-long debate between Steven Pinker and Stuart Russell on the "Foundations, Benefits, and Possible Existential Threat of AI".
A delightfully snarky history of web design. If you remember laying out websites using tables, this article is for you.
Imagine we're living in an oppressive theocracy. Secretly I'm an atheist. I don't even believe our God exists, I certainly don't support any of our oppressive laws, but I don't say anything because I figure everybody else believes and if I speak up I'll be tortured and executed. Now imagine everybody else feels exactly the same way.
That's pluralistic ignorance and it's the most underrated phenomenon in social psychology. You can build empires and religions on pluralistic ignorance. (More prosaically, it explains why authoritarians viscerally dislike the idea of online anonymity.)
This amazing set of animated math videos by YouTuber 3Blue1Brown is the best intuitive, visual introduction to calculus I've seen anywhere.
Ever wondered what would have happened if Harry Potter had arrived at Hogwarts a scientific child prodigy, determined to use logic and experimentation to uncover the true nature of wizardry? Me neither. But AI researcher Eliezer Yudkowsky has, and he's written a 1,500-page novel detailing the insanity that, quite reasonably, follows.
This blog post from back in the mists of 2012 may have the most perfect blog post title ever written. The article itself is pretty good too. The Last Psychiatrist certainly doesn't think much of academia:
Imagine a large corporate machine mobilized to get you to buy something you don't need at a tremendously inflated cost, complete with advertising, marketing, and branding that says you're not hip if you don't have one, but when you get one you discover it's of poor quality and obsolete in ten months. That's a BA.
A surprisingly enlightening analysis of how and why medieval European armies were raised, trained, and fought built around... exactly, a military historian's review of the Battle of Helm's Deep scenes from the Lord of the Rings movies. How did people fill their time before the internet? I can't remember anymore.
Dying is a longstanding tradition in many human cultures around the world; Scott Alexander argues that sensible people do it as far away from hospitals as they can. I'm inclined to agree.
An archive of interesting documents from Irish republican history.
Interesting debate on the British Empire's decision to declare war on Imperial Germany in 1914, one of the greatest strategic mistakes in the history of Great Power politics.
A particularly good interview from the Irish Times politics podcast — political science professor Brendan O'Leary talks about the economics of a reunited Ireland.
This is fun — a military historian analyses the strategy, tactics, and technology on display in Peter Jackson's Lord of the Rings movies.
A long-runing podcast series by historian Finn Dywer with hundreds of episodes available, most between twenty and thirty minutes long. His 14-episode sequence on the Viking Wars and 23-episode sequence on the Norman Invasion are particularly good.
Peacefully, with dignity, and not in hospital according to professor of medicine Ken Murray. Classic essay on the downside to overly-agressive medical treatment.
I really enjoyed this set of podcasts by Silicon Valley tech investor turned philosopher Naval Ravikant. He's a smart guy with a lot of interesting things to say about the world. Admittedly, listening to him had zero effect on my net worth, but that's because I'm a uniquely spiritual individual. Less saintly types will likely find his advice more profitable.
Patrick Wyman had just completed his PhD thesis on the late Roman empire when he recorded this podcast series and his expertise shows. Balancing scholarly rigour with entertainment is a tough challenge but he manages it impressively well. The series has 23 episodes in total, each about 40 minutes long.
Rob Reid's After On tech podcast is a series of in-depth interviews with smart, interesting people at the top of their fields. These two talks with Naval Ravikant on near-future threats to human civilization are among the best. AI, drone technology, and (most of all) synthetic biology are the main topics on the agenda.
The granddaddy of all history podcasts, Mike Duncan's History of Rome was the first and it's still one of the best. Roughly 180 episodes, about 20 minutes each, covering the history of Rome from the founding of the city in 753 BC until the fall of the western empire in 476 AD.
Bryan Caplan's thesis: college degrees are valuable but not because colleges teach important knowledge or skills — they mostly don't. Instead, a college degree functions as a hard-to-fake signal for potential employers of an individual's pre-existing intelligence, conscientiousness, and willingness to conform.
Ryan Tomayko's classic (and very funny) introduction to "thinking in shell". I found this talk genuinely eye-opening the first time I watched it.
Philosopher Dan Dennett on the natural history of religion. His central thesis is that religions are memes — highly evolved psychological viruses which replicate by infecting human minds.
The single most vicious takedown of a programming language ever written.
Paul Graham's classic meditation on blasphemy.
Everything you never knew you wanted to know about hexagonal grids.
Classics professor Eric Cline on the mysterious Bronze Age catastrophe that brought down empires and ushered in the first Dark Age.
A useful summary with proofs and examples.