My previous post about the Russian Revolution is very information-dense and hard to follow. I’ll translate it to Americanese:

Ah, Great America at the turn of the 20th century. A poor, backwards, and too-large multi-ethnic Empire of about 130 million, in which the dominant social struggle is about the oppression of the farmers, who make up about 80% of the population. The farmers hunger for land: despite America’s vast size, most of it is not arable. Bands of destitute farmers roam the countryside looking for work, sometimes turning violent.

Besides the farmers, about 10% of the population is engaged in industrial labour (eg. mining and construction) and only 1–2% works in factories. And they’re also not happy, because conditions for Industrial Revolution factory workers were horrific.

Regardless, the American Empire at the start of the 20th century is buckling under the onslaught of modernity. The old planter elite are fading away (they used to keep the farmers in check), the Emperor Nicholas of House Washington is incompetent but undercuts anyone capable of reforming the regime, and the country is in the middle of a prolonged economic contraction. Pretty much everyone is unhappy.

Then in 1905 the “barely-civilised” Japanese launch a surprise attack on Alaska, nearly wipe out the Pacific Fleet, and take Anchorage and the western half of the state. The already-rickety state, the massive military mobilisation, and the lost war trigger a failed revolution in the form of strikes, farmer violence, and mutinies in the US Army. The bureaucracy and the soldiers returning from the front crush the uprisings, while Emperor Nicholas is pushed to allow a Congress. Nicholas refuses and asks his uncle (a four-star general) to create a military dictatorship, but his uncle says he’ll shoot himself if the Emperor does not agree to create a Congress. Nicholas agrees, and then spends the rest of his life trying to fight and weaken Congress and reclaim his power, preventing the formation of a non-revolutionary political alternative.

Bloody Sunday on the National Mall, 1905

Once the 1905 revolution is put down, a heroic bureaucrat called Stollman tries to save the system with a two-pronged attack: reforming agriculture to address the farmers’ hunger for land, and repression (killing). America experiences an economic boom, but Stollman dies before he can finish the reforms.

Stollman's Necktie

A terrorist with American links shoots a British nobleman in Halifax, and all the countries of the world are ready to go to war over it. Nicholas tries to avoid war with the British, but eventually decides that war will regenerate and unite the country, solve all the annoying social problems, and deliver a great patriotic and religious victory.

The war starts disastrously for the Americans. The British, based in Canada, advance deep into America, taking Chicago, Detroit, and most of the Pacific Northwest, and threatening Albany. Nicholas, like all boys, loves soldiers and war, so he leaves the capital to live with his generals and lets his wife influence the running of the country.

She erratically appoints and dismisses government ministers at a high rate, culminating in the appointment of Porter as Minister of the Interior (i.e. the guy responsible for keeping the country together). He suffers from frequent mental breakdowns, hallucinations, and erratic mood swings, and consults mystics, spiritual mediums, and fortune-tellers for major decisions. Another blow to the populace’s faith in the government ¹.

So the war goes on, and despite more losses and a constant high rate of desertions, the state holds together and after a few years manages to deliver a lot of combat power to the front. It develops into a stalemate on the Canadian Front. The Americans haven’t lost, but they’re barely still in the fight. Over two million American boys have died.

Territorial losses by the American Empire

But the straw that finally breaks the camel’s back is the food supply to Washington DC collapsing. In response there are mass demonstrations and strikes, and the capital garrison mutinies instead of firing on them.

Finally, enter the Soldiers’ and Workers’ Councils. The Communists (the main characters) are only a small faction within them; the rest are various flavours of leftists. The Washington Council is a rival government operating from the capital, and its influence rests on soldiers and factory workers (not farmers!). Its power does not extend much beyond the city; remember that 80% of the population is in the countryside, farming. The Communists flood the front line with agitators to ideologise more soldiers.

The generals at this point have lost faith in Emperor Nicholas and see him as incompetent, so they pressure him to abdicate so someone else can take the crown. He does so, handing the crown to the next in line, but the next in line unexpectedly declines. So the monarchy is accidentally abolished. Oops.

What if Emperor Nicholas hadn’t spent so much time undercutting Congress? Perhaps at this point a powerful and legitimate Congress could’ve taken over and righted the situation.

A Provisional Government forms, but it wants to continue the war and won’t redistribute land to the farmers without an assembly. Desertions and land seizures by farmers skyrocket.

The Communists in the Washington Council declare that they want to end the war, and the British pick up on this and transport the top Communist John Lennon into America. Lennon calls for an immediate end to the war.

The Provisional Government tries to hold the centre together, balancing the right wing and the left wing (the Councils). The Secretary of Defense Kearney gambles that a successful offensive against the British will legitimise the Provisional Government, but the offensive fails disastrously.

July Days. The army had just opened fire on street protesters.

Kearney takes over the government, pivots right (alienating the left), and appoints the right wing’s Hard Man General to run the army and Restore Order. The Hard Man General moves on Washington DC to Restore Order. Kearney interprets this as a coup, denounces him as a traitor, and arms the Washington Council to resist the Hard Man General (alienating the right). To this day it’s still unclear if the Hard Man General was actually plotting a coup or not.

The Hard Man General’s commands no longer have the government’s authority. Railway and telegraph workers obstruct his units’ movements, and the soldiers refuse to obey orders anyway. Oh, and now the Washington Council has guns!

What if Kearney tried to end the war instead of starting an offensive? What if he communicated better with the Hard Man General and they managed to put down the Communists in Washington DC?

So the monarchy is gone, the Provisional Government has lost all authority, and control over the army is slipping.

At this point the Communists don’t so much seize power in Washington DC as walk into lightly defended government offices with some desultory shooting. Hooray, utopian revolution is here!

Meanwhile in the countryside, the farmers are organically seizing and redistributing land. In terms of sheer scale and social restructuring, this is the real revolution.

American peasants in the early 1900s

Okay, back to the Councils: they’ve now “seized power” in the capital, which means they control a very small fraction of the country. How do the Councils get national support, since “Workers’ Revolution!” is irrelevant to more than 80% of the population, who are concerned with land? Lennon picks the slogan “Peace, Land, and Bread”, which is exactly what everyone wants. This stroke of genius wins him a majority within the Washington Council.

It’s a popular message, but all throughout this period the farmers are already seizing the land in the countryside without waiting for a political party in the city to give them permission. Still, the Communists are legitimising the land revolution, and that’s very popular.

What if the Provisional Government or Kearney came up with the programme of Peace, Land, and Bread before the Communists did? Would they have become wildly popular and legitimate in the eyes of the farmers and the soldiers?

A note: the Councils aren’t just “communist”. There are a variety of socialist parties, including the Left DSA, the Right DSA, the Kamalists, and the Communists, and they disagree with each other. Some support a congressional democracy, some only care about the farmers, and others want to keep the war going defensively to avoid territorial losses. But the Communists are dominant within the Councils (for the moment).

The Provisional Government sets a national election in motion before it collapses. The DSA wins 40% of the vote, while the Communists get only 24%. So the Communists are dominant inside the Councils, but are a strong minority on the national level.

Here’s an important pattern: if the Communists allow a unity government of all socialist parties, they could be diluted, so they refuse a left-unity coalition and ally with one half of the DSA, leaving the other wing out in the cold with no power, only to later destroy that ally once it is convenient.

Okay, so the Communists have taken power in the capital by squeezing out the other leftist factions and taken control of parallel revolutions led by Councils in other big cities, but they don’t control the countryside at all. They don’t have a state yet.

Everywhere outside Communist control fractures into a complex mess of factions: Right Militias seeking to restore the old order, anti-Communist socialists, breakaway states, anarchists, local warlords, and others. It turns into a huge civil war.

The Communists build their army by seizing enormous stockpiles of arms (which are in the cities, where the Communists are in control!), reinstating conscription, and incorporating former Imperial officers (they can always be killed later!).

Beat the Right Militias with the Red Wedge, 1919

What’s the difference between conscripted soldiers and slavery?

Despite having most of the old Imperial generals and large armies, the Right Militias are hamstrung.

See, the Communists have tight control from above with a strong political programme: land for the farmers, peace for the soldiers, and worker power for the factory workers. Fighting for the Communists makes sense; you’re fighting for a good future, even if in practice they are incredibly brutal to their soldiers (just as the Right Militias are).

The Right Militias, on the other hand, refuse to promise land for the farmers, have no unifying figure or political programme, and embrace excessive American nationalism, which alienates potential allies in the nationalist movements (eg. Texans and California’s predominantly Mexican population).

What if the Right Militias, at last, promised land to the farmers and liberation to the national minorities to bolster their support?

American theatre in the civil war. Blue arrows are Right Militia advances.

You’d think that being surrounded by the Right Militias would be a disadvantage for the Communists, but it’s the other way around: this gives them the advantage of interior lines. The Communists control the cities and the rail lines between them, while the Right Militias are stuck in the vast countryside with no good way to shift large amounts of forces and supplies around.

So both main forces are in utter disarray, experience high rates of desertion, and can barely muster supplies, but the Communists win in the end. All that’s left for the Communists is to reconquer as much of the periphery as they can.

So how does a revolution happen?

There have to be a number of national calamities, and alternative centres of power need to keep self-destructing until only the revolutionaries are left.

How does a revolution happen? Does everyone decide “okay, it’s time for communism now”? How does a fringe group seize power over the state? I’ll illustrate with the Russian Revolution.

Ah, Mother Russia at the turn of the 20th century. A poor, backwards, and too-large multi-ethnic Empire of about 130-ish million, in which the dominant social struggle is about the oppression of the agricultural peasantry, who make up about 80% of the population. The peasants hunger for land, they want it bad, and they are not getting it. Besides the peasants, about 10% of the population is engaged in industrial labour (eg. mining and construction) and only 1–2% works in factories. And they’re also not happy, because conditions for Industrial Revolution factory workers were horrific, and it was even worse in Russia.

Regardless, the Russian Empire at the start of the 20th century is buckling under the onslaught of modernity. The old noble elite are fading away (they used to keep the peasants in check), the Emperor is incompetent but undercuts anyone capable of reforming the regime, and the Empire is in the middle of a prolonged economic contraction. Pretty much everyone is unhappy.

Then in 1905 the “barely-civilized” Japanese surprise-trounce Great Russia in a war in the Far East. The already-rickety state, the massive military mobilisation, and the lost war trigger a failed revolution in the form of strikes, peasant violence (they hunger for land!), and mutinies. The bureaucracy and the soldiers returning from the front crush the uprisings, while Emperor Nicholas II is pushed to allow a parliament. Nicholas refuses and asks the Grand Duke to create a military dictatorship, but the Grand Duke says he’ll shoot himself if the Emperor does not agree to create a parliament. Nicholas accedes, and then spends the rest of his life trying to fight and weaken the parliament and reclaim his power, preventing the formation of a non-revolutionary political alternative.

Bloody Sunday in 1905 by Wojciech Kossak

Once the 1905 revolution is put down, another heroic bureaucrat called Stolypin tries to save the system with a two-pronged attack: reforming agriculture to solve the peasant land-hunger problem, and repression (killing). Russia experiences an economic boom, but Stolypin dies before he can finish the reforms, and a few years later Franz Ferdinand gets shot.

Stolypin's Necktie

Nicholas tries to avoid war with the Germans, but eventually decides a war will be regenerative, unite the country, solve all the annoying social problems, and be a great patriotic, religious, victorious war.

The war starts disastrously for the Russians. Emperor Nicholas, like all boys, loves soldiers and war, so he leaves the capital to live with his general staff and lets his wife influence the running of the country.

She erratically appoints and dismisses government ministers at a high rate, culminating in the appointment of Protopopov as Minister of the Interior (i.e. the guy responsible for keeping the country together). He suffers from frequent mental breakdowns, hallucinations, and erratic mood swings, and consults mystics, spiritual mediums, and fortune-tellers for major decisions. Another blow to the populace’s faith in the government ¹.

So the war goes on, and despite more losses and a constant high rate of desertions, the state holds together and after a few years manages to deliver a lot of combat power to the front. It develops into a stalemate on the Eastern Front. The Russians haven’t lost, but they’re barely still in the fight. Over two million soldiers have died.

Territorial losses by the Russian Empire

But the straw that finally breaks the camel’s back is the food supply to the capital collapsing. In response there are mass demonstrations and strikes, and the capital garrison mutinies instead of firing on them.

Finally, enter the Soldiers’ and Workers’ Soviets, of which the Bolsheviks (the main characters) are only a small faction. Soviet is Russian for “council”, so it’s like a rival government running out of the same capital whose influence rests on the soldiers and workers. They don’t have power beyond this; remember that 80% of the population is in the countryside, farming. The Bolsheviks flood the front line with agitators to ideologise more soldiers.

The generals at this point have lost faith in Nicholas and see him as incompetent, so they pressure him to abdicate so someone else can take the crown. Nicholas does so, handing the crown to the next in line, but the next in line unexpectedly declines. So the monarchy is accidentally abolished. Oops.

What if Nicholas hadn’t spent so much time undercutting parliament? Perhaps at this point a powerful and legitimate parliament could’ve taken over and righted the situation.

A provisional government forms, but it wants to continue the war and won’t redistribute the land without an assembly. Desertions and seizures of land by peasants skyrocket.

The Bolsheviks in the Soviet declare that they want to end the war, and the Germans pick up on this and transport the top Bolshevik Lenin into the Soviet. Lenin calls for an immediate end to the war.

The provisional government tries to hold the centre together, balancing the right wing and the left wing (the Soviets). The War Minister Kerensky gambles that a successful offensive against the Germans will legitimise the provisional government, but the offensive fails disastrously.

July Days. The army had just opened fire on street protesters.

Kerensky takes over the government, pivots right (alienating the left), and appoints the right wing’s Hard Man General to run the army and Restore Order. The Hard Man General moves on the capital to Restore Order. Kerensky interprets this as a coup, denounces him as a traitor, and arms the Soviets to resist the Hard Man General (alienating the right). To this day it’s still unclear if the Hard Man General was actually plotting a coup or not.

The Hard Man General’s commands no longer have the authority of the government, the railway and telegraph workers obstruct his units’ movements, and anyway the soldiers refuse to obey orders. Oh, and now the Soviets have guns!

What if Kerensky tried to end the war instead of starting an offensive? What if he communicated better with the Hard Man General and they managed to put down the Soviets in the capital?

So the monarchy is gone, the provisional government has lost all authority, and control over the army is slipping.

At this point the Bolsheviks don’t so much seize power in the capital as walk into lightly defended government offices with some desultory shooting. Hooray, utopian revolution is here!

Meanwhile in the countryside, the peasants are organically seizing and redistributing land. In terms of sheer scale and social restructuring, this is the real revolution.

Russian peasants in the early 1900s

Okay, back to the Soviets, they’ve now “seized power” in the capital. Which means they control a very small fraction of the country. How do the Soviets get national support, since “Workers’ Revolution!” is irrelevant to more than 80% of the population, who are concerned with land? Lenin picks the slogan “Peace, Land and Bread”, which is exactly what everyone wants. This is considered a stroke of genius, which gets him majority support within the council.

It’s a popular message, but all throughout this period the peasants are already seizing the land in the countryside without waiting for a political party in the city to give them permission. Still, the Bolsheviks are legitimising the land revolution, and that’s very popular.

What if the provisional government or Kerensky came up with the programme of Peace, Land, and Bread before the Bolsheviks did? Would they have become wildly popular and legitimate in the eyes of the peasants and the soldiers?

A note: when I say Soviet, it’s not just “communists”. There are a variety of socialist parties like the Mensheviks, the Left SRs, and the Right SRs, and they disagree with each other. Some support parliamentary democracy, some only care about the peasants, and others want to keep the war going defensively to avoid territorial losses. But the Bolsheviks are dominant (at the moment).

There’s a national election that the Provisional Government sets in motion before it collapsed, and the SRs win 40% of the vote, while the Bolsheviks get only 24%. So the Bolsheviks are dominant inside the Soviets, but are a strong minority on the national level.

Here’s an important pattern: if the Bolsheviks allow a unity government of all socialist parties, they could be diluted, so they refuse a left-unity coalition and ally with one half of the SRs, leaving the other wing out in the cold with no power, only to later destroy that ally once it is convenient.

Okay, so the Bolsheviks have taken power in the capital by squeezing out the other leftist factions, and taken over the other parallel Soviet revolutions in the big cities, but they don’t control the countryside at all. They don’t have a state yet.

Everywhere the Bolsheviks don’t control falls into a complex mess of factions, split up between the Whites who want to restore the old order, anti-Bolshevik socialists, nationalists who want to break away from the Russian empire, anarchists, local warlords, etc. It turns into a huge civil war.

The Bolsheviks organise the Red Army through mass conscription, seize enormous stockpiles of arms (which are in the cities, where the Bolsheviks are in control!), reinstate conscription, and incorporate former Imperial officers (they can always be killed later!).

Beat the Whites with the Red Wedge, 1919

What’s the difference between conscripted soldiers and slavery?

Despite having most of the old Imperial generals and large armies, the Whites are hamstrung.

See, the Bolsheviks have tight control from above with a strong political programme: land for the peasants, peace for the soldiers, and worker power for the factory workers. Fighting for the Reds makes sense; you’re fighting for a good future, even if in practice they are incredibly brutal to their soldiers (just as the Whites are).

The Whites, on the other hand, refuse to promise land for the peasants, have no unifying figure or political programme, and embrace excessive Russian nationalism, which alienates potential allies in the nationalist movements.

What if the Whites, at last, promised land to the peasants and liberation to the national minorities to bolster their support?

European Theatre in the civil war. Blue lines are White advances.

You’d think that being surrounded by the Whites would be a disadvantage for the Bolsheviks, but it’s the other way around: this gives them the advantage of interior lines. The Soviets control the cities and the rail lines between them, while the Whites are stuck in the vast countryside with no good way to shift large amounts of forces and supplies around.

So both forces are in utter disarray, experience high rates of desertion, and can barely muster supplies, but the Reds win in the end. All that’s left is for the Bolsheviks to reconquer as much of the periphery as they can.

So how does a revolution happen?

There have to be a number of national calamities, alternative centres of power need to keep self-destructing until only the revolutionaries are left.

So I told an LLM to “Make me an RTS game” it made a real piece of shit, but I got completely sucked into improving it. And I’ve been using Codex very heavily for this:

Three weeks in, I’m trying to make my art look better for a key new unit, but the LLM is very stubborn about making everything look weird and bad. Each iteration I ask for changes, it would fix one thing and break something else, and the design process ends when I give up and decide it’s good enough.

I guessed that the problem was the LLM chose to use this library called PixiJS for drawing everything, and I figure there might not be a lot of training data for how to draw stuff with this library, so the output was bad.

But I am not going to edit the visuals by hand. I refuse to do manual labour.

My theory was that LLMs have much more training data for SVGs (an extremely common graphics format that’s text-based) than for PixiJS drawing code. So if we switched to SVG, design performance should improve, and I’ll spend less time saying “nono, undo undo”.

But I didn’t want to start from scratch with SVG, I had already put real effort into some of the existing units. Objectively they’re terrible but it was a lot of effort to get an LLM to make anything half decent, and so I’ve grown attached to them:

The units before the SVG conversion

I first asked the LLM to convert the worker (the cute little pentagon at the top) into SVG. This is what I got:

Fuck! If it can’t translate a pentagon and a line into SVG, what hope do I have to translate my sick tank model where the turret is animated and recoils when you shoot?

So, what to do? I needed to make this task legible to agents, even though LLMs are really bad at visual work.

My first idea was to render the existing PixiJS unit on a transparent background, render the SVG version, and compare them pixel by pixel. The inspiration came from this old blog post. And then calculate the diff both visually and mathematically, and make the LLM work until there was no difference:

Then I realized that each unit is made out of multiple visual components: for the worker unit above it’s made out of a pentagon and a stick.

So to make this even easier for the LLM, I could break the existing PixiJS drawing into its individual components, and then have the LLM produce SVG components that were pixel matches:

We break down the worker into the pentagon body and the stick of the worker

Once the components all match, we layer everything together and make sure the final image matches.

So I described this method to the LLM, had it create a multi-phase plan¹ to implement this harness, sent it to work, and it Just Worked.

This is, I think, what is meant by harness engineering. LLMs currently don’t have the creativity to suggest techniques such as this, and careful engineering can arrange situations where the nondeterministic stream of tokens can be leveraged to do useful work quickly and at scale. LLMs are like a very powerful but dumb horse, so for complex tasks you have to build a harness that lets it pull mindlessly in the right direction.

Was it worth it? Was the LLM actually better at producing SVGs than PixiJS drawings?

I’ve designed one unit so far using the SVG method and it felt a lot better. It’s hard to say for sure if it’s because of SVG, because I’ve also gotten a lot better at prompting for this stuff (another topic, the importance of evals), but I was able to generate a somewhat complicated unit design relatively quickly, so check out my new wizard character:

And yet, I always go to the gym. I have made it very difficult for myself to ever quit going to the gym.

Sometimes I’m really very tired and I decide “That’s it, I’m too tired, I’m skipping the gym today.” To my shock, five minutes later I’m putting my shorts on. It’s like I’m addicted, I lack the willpower to stay home, I have no control over this habit.

How? Well you see, nicotine is a habit forming drug. When smokers quit smoking, they still like going outside for five minutes every hour.

Well I just take nicotine every time I go to the gym. Now I’m addicted to going to the gym.

I don’t smoke or vape, don’t do that, it’s bad. I take a gum. It slowly releases nicotine for about the length of a gym session. Fast intake methods are more addictive, the dose is less controllable (so you keep “accidentally” increasing the dose), and the effect wears off quickly. So use gum, or something else that’s slow.

Actually, the smallest gum they sell (2mg) is too much for me so I take half a gum and throw away the other half.

Compared to being sedentary, that’s a 25% reduction in all-cause mortality risk ¹.

Without this, I definitely would not go to the gym. I hate exercise!

What’s the negative health effects of 1mg of gum 3x a week? Well they haven’t studied nicotine doses that low. But it’s probably zero ².

Won’t this get you addicted to vaping and smoking?

No you dummy, because once you start using nicotine for any other purpose, it dilutes the gym=nicotine association and ruins the entire healthy addiction. So when I have an opportunity to smoke, I don’t, cause I don’t want to ruin my special gym habit.

I have been doing this for years. If you’re seriously considering this:

• Take a one week break twice a year.
  • Especially if you find yourself being irritable on off days.
• Don’t overchew the gum, keep it at a slow drip.
• First time I tried nicotine I worked out for two hours straight. Enjoy! That effect wears off over time.

Step Zero (Most Important)

First, accept that you will live alone for the rest of your life, and that’s okay, because you’ll build a good life for yourself anyways. You must not hate or judge yourself for your loneliness, but learn to enjoy your own company. This takes a few years. You have to be able to imagine your future as a single man for the rest of your life and not flinch or feel fear, but to feel excitement at more decades that look like your current life. Don’t become proud of being single and don’t complain about it, because you have more important things in life you want to do.

Only when you feel like you don’t need the validation of a woman picking you, when you’re not afraid of women, or when you’re content with your life, will you be able to go on dates and be your relaxed, authentic self. I got to this point at about twenty-eight years old.

Lift weights or do sports. You have to be healthy. You don’t need to start worrying about having a low bf%, women mostly just like healthy men. Becoming obsessed with your physique is kind of stupid; just work out enough that you feel good day to day.

No porn, no jerking off. Porn makes you stop seeing average women as incredibly beautiful and raises your standards unrealistically high. It anaesthetises your attraction to real women. And if you stop jerking off, it’ll give you the drive to keep swiping through the very psychologically hard process of online dating.

Move to New York, or a city with more women than men. If you’re in Seattle or SF, leave.

Step 1

On Hinge, send five likes a day. At the same time, you should be judging every profile hard and rejecting 99% of them, so it can take around twenty minutes of swiping each day to send out your five likes, which is a huge pain. But remember, you’re okay being single, you’re not desperate, so you’re holding out for the one that’s right for you.

• Beware: instantly judging hundreds of women every day will, if you’re not careful about curbing this, infect your mind with negative thoughts about women in real life.

Use Photofeeler. I paid about $75 to find out the photos I thought were good were actually terrible.

Every like must have a message. Ask an interesting question and try to connect on a personal level. It can take a few minutes of reading a profile until you think of a good conversation starter, but if you can’t think of one, maybe that means you like her because she’s hot, not because you’re actually interested in her.

Be upfront in your profile. Are you a weeb who just wants someone to watch anime with? Put that in your profile. Are you looking for someone frugal and practical? Put that in your profile. Are you extremely boring? Put that in your profile. Don’t pretend to be something you’re not because you think that’s attractive. If you feel the need to change your presentation to be more attractive to women, go back to step zero.

Send no more than four messages on the app before scheduling a date.

You will be sending likes for months. It will likely be brutal, you’ll quit multiple times. It gets easier over time as you improve your profile, get more experienced with sending good messages, and start getting dates.

Step 2

The date should be something free like a park or a free museum. Do not do dinner or lunch dates; those are awful and a waste of money.

If it happens that you want coffee, buy them a coffee too, but paying for a meal should only happen when you’ve spent a few hours together and you’re both laughing and smiling and starting to get hungry. 90% of your dates should cost no more than two subway tickets.

• I want an equal with a practical outlook who can pay for themselves. You want to filter out the immature people who buy into simplistic romantic narratives. Not paying is a powerful filter for those kinds of people — does she still like you even when you’re not a source of free meals and cocktails?

Your goal is to find out if you feel good around her. Do you feel funny and charming and comfortable around her? You’re not trying to get laid or get a second date. It is to figure out if she’s a waste of your time or if you’ll still enjoy your hundredth date. Do not try to impress her. In fact, it’s better to be your cringe self. Remember: you’re okay with being alone.

• You’re looking for a woman that’s mature enough that she’s no longer looking to be swept off her feet by prince charming, but coldly calculating which man she wants to spend the next forty years of her life with.

Start clear: state that you’re looking for a life partner and want to get married, and whether or not you want kids. This won’t end the date because you’ve filtered out anyone you’re incompatible with on the app, but it’s important to set the tone for what kind of date this is going to be. I waited for the second date to talk about the pre-nup.

Then just try to have fun. Joke around, talk about your hobbies, be loose.

Is she hot but you felt uncomfortable? Do you really like her, but you didn’t feel very happy around her? No second date, goodbye. If you’re not smiling, feeling great, and laughing with her, wish her a good life and go home.

Ask about her relationship with her parents. It’s a big green flag if it’s good, and if it’s bad, see how mature she is about it. Does she see her parents as flawed people, or does she make them out to be evil and expect validation?

Disclose your own history of mental illness or depression if you have any, and ask her directly about hers. If she has a a history, you need to have a serious conversation about it, even if it’s a first date. Ask directly what her plan is and what she’s already done to treat and resolve the illness. Pay attention if she takes responsibility for it, or if she blames the illness. Plenty of people worth dating are adults building their lives and managing a mental illness.

• Look up what the numbers are on divorce rates and relationship quality when one partner has a mental illness. This is serious.
• It’s especially common these days for people to have mental illness as a permanent matter of identity. People who define themselves by their diagnosis and demand it be accepted are not at a sufficient level of mental maturity for life partnership, at least for me.
• No shame to the mentally ill, I’ve been there myself, but I would have been a terrible partner back then.

If you’re incompatible on some of these axes, you’ll be tempted to say “fuck it, why not” and compromise on your principles, especially if you’re horny from step zero. Just wrap up the date, try to make sure she has a good time, and go home.

If she offers to be friends, say no and end the date immediately. You’re looking for a girlfriend, and if you’re too afraid to say that to her face, you need to go back to step zero.

I averaged one to two dates a month. I had maybe a dozen bad dates, two enjoyable dates that didn’t work out, and one good date. She’s moved in now. There’s many indicators that we’ll have a long and good relationship. Of course, we both agree that people change over time and we’re still not done maturing, and we’re content with the possibility this relationship stops working for us. Really, only time will tell if my method is any good, but I’m not going to wait until we’re happily married for twenty years to make this post.

In practice, this is terrible. Stalin truly believed it was necessary for the good of the Soviet people to forcibly collectivize the peasants, killing millions. Mao geuinely believed in revolution, class strugle, and remaking society, killing millions.

If your institutions is full of hard-working true believers, they’ll be willing to do horrible things for the greater good. If each institution is captured by a special interest or lobby, a would-be ideological ruler has to balance all their interests to get anything done, and this acts as a check against Good Ideas.

Now don’t get me wrong, Soviet Russia and Nazi Germany were corrupt and that didn’t save their victims. But it’s not the worst thing in the world for the rulers to be mildly corrupt and susceptible to the influence of powerful players.

Ideally you have well designed institutions, a free press, an independent judiciary, and a democratic system. In practice, institutions arise organically and evolve chaotically, so you have to be very very lucky to have those. So if you don’t what are the options?

Well a president who takes bribes and can’t get anything done without appeasing a dozen powerful groups might be so incompetent there’s a famine, but he’s also not likely to engineer famines because he thinks it’s a good idea.

Let’s say you’re involved in home construction. Traditionally that involves making a foundation, putting up walls, etc.

But a robot comes out that’s very good at that.

Are you out of a job? No.

Turns out building the walls was never really the hard part, it was making sure the structure of the home could meet expected lateral and gravity loads with minimal cost while providing a pleasant and airy interior.

But they build a robot that’s very good at that too.

Turns out the hard part of homebuilding was stuff like creating a ventilation strategy that balances perceived temperature (not at the thermostat!), pressure, and operating cost.

The robot has an update, it can do that too now.

Now the difficulty is talking to customers, building trust, and helping them understand the trade-offs in front of them. Customers don’t know anything about homes, yet you must help them choose between a walk-in pantry or a dedicated laundry room on the bedroom floor, because they can’t afford both.

Somehow, they build a robot that’s good at this too.

Well the hard part is sourcing quality materials at a good price and making sure your robots show up on time with the tools and materials they need to do the job you’ve clearly specified.

New update.

The hard part is now striking deals with developers and creating an efficient production pipeline of homes to make sure their customers get their homes on time.

Update.

Market research to create a collection of pre-made home designs that satisfy 90% of customers while cutting material costs 30% due to massive bulk orders and speeding up inspections.

Update.

Building a coalition of local interest groups in order to rezone new land for development, getting the city to commit to building roads and power/water hookups.

Update.

Achieving bigger margins by designing more liveable communities with parks, schools, trails, retail/office space, and public transit connections.

And so on.

In software, we’ve been laddering our way up the levels of abstraction like this for decades.

If you’ve studied a complex system for years, like the U.S. financial system or the global supply chain network, you might be able to understand just how many abstraction layers we’ve built up in software. But for a layman without this experience, it can be hard to imagine.

And every time we’ve solved a layer of the software problem and made it easier, the demand for more software increased.

Yes, we’re laddering a lot faster now with AI. For now, humans have been comfortably moving on to the next level of abstraction above AI.

Is there an end to the ladder? I don’t see one, but the future is foggy and I can’t see very far. I don’t know.

I do know that at each level of abstraction we introduce new difficulties. People to make happy, complex trade-offs to balance, and ever longer term planning.

This is tough for both LLMs and humans. Not every engineer makes it to the next level of the ladder. And at each step the LLMs will find it harder to understand what the problem is and how to measure if it’s solved.

Problems become more wicked.

• We stop having easy solutions like “made it twice as fast” and instead it’s “if we make it twice as fast it costs 1.4x more” where a dozen stakeholders have different opinions on what the best option is.
• Solutions become irreversible and require large amounts of investment. What if we fundamentally rearchitected our system? Is that necessary to survive the next three years, or would it cost $5m of burned tokens for little reward?

I’m not saying humans will always be one step ahead of the LLMs. I don’t know how this could go. But this is how I’m thinking about it.

As to how that affects software engineering careers, I don’t know.

Maybe:

• The ladder tops out and LLMs can solve software problems cheaper than humans. Software engineering goes the way of the farriers.
• Even if software demand grows, staffing could still shrink if productivity grows faster.
• AI takes over all white collar work.
• Some software engineers create new roles with less technical emphasis but with more fun, impact, and compensation, while others fall off the ladder.
• LLMs peter out in a year and technical skills remain highly valued, everything stays the same.

One consequence of this is it can be difficult to tell where your values end and where others’ begin. One might not even realize there’s a difference.

But there often is.

2
You can change your social environment to try to shake off some baggage. Pick up new hobbies, make new friends, fashion a new you. Some values will fade, and others will rise. But they can be sticky: even if you haven’t seen someone for a while, their influence may linger on in your worldview and expectations. For the deepest bonds, this can last for years. Maybe you picked up a lot of thought patterns from an old girlfriend.

Another problem is that when you’re wandering and lost, it’s easy to end up in a different crowd with a compelling value system that still isn’t your own.

But realistically, old bonds are often the strongest and you’ll end up with your feet in different worlds. Still, the journey is likely to teach you a lot about yourself and hand you the new problem of figuring out how to live in accordance with your new values.

3
Imagine a block of wood. As you think and act, you carve grooves into that block. The more you follow a certain path—certain thoughts, certain behaviors—the deeper that groove becomes. Next time, it’s easier to follow the same path, because you slip into it.

Then you’re an adult, and you realize your life isn’t going well. You want to change. You make a firm resolution to change. You read a lot of self-help books on changing. But you still keep slipping back into the old grooves.

Change is hard. It happens in brief, rare windows where the old pattern is kicking in but you catch it in time to alter your course. Most of the time you breeze through on automatic and don’t notice until a week later that you’ve done as you always do.

But the regret of missing the window sharpens your attention for the next opportunity, and you even start to manufacture some of these rare windows.

So you make a plan, you catch that moment, you act differently and leave a faint groove in the new direction. You likely feel more bewildered than satisfied, in uncharted territory. But it’s Good.

Then you understand that fighting your nature isn’t pleasant, and it can take years to alter aspects of your core personality. It’s a life’s work that few finish, like learning a language.

You have to be realistic about what’s possible. Sometimes it’s easier to accept yourself and work around your shortcomings. This is not to be pessimistic, I’m just saying you have to pick your battles and often it’s better to work around a flaw rather than to fix it.

4
You strive to change, but your bonds expect you to stay the same. As you push to break out of your old mold, some bonds will break, some will change, and new ones will appear.

You can use this. Change is much easier in new contexts where you can metaphorically flip the block to a fresh side where you haven’t carved any grooves. We all have better and worse sides waiting to be brought out. Find people that naturally pull you toward your better self.

Which means being careful with who you allow into your life. Their values and desires will rub off on you. They may be attractive, but do you want your soul to look like theirs?

Eating is a very base human activity: you do it and it’s over. You’re the same person going out as you are going in. It’s a purely sensory experience, sometimes it feels good and sometimes it’s meh.

Eating is on exactly the same level as pooping.

2.

So it’s confusing when people pay for luxury restaurants, or any restaurant that isn’t going for the “best bang for your buck” angle.

Why do so many restaurants have Instagram pages?

Well if you ask people who go to those places, they’ll say it’s a social thing. But you can have a social gathering at someone’s apartment (which is free), a park (also free), or a cheap restaurant.

What explains this?

Well you know how Marlboro got men to buy their cigarettes in droves by plastering pictures of masculine cowboys all over the world?

Well, restaurants are also selling self-image. French fine dining sells one image. High-end tacos another. Ramen a third. They’re selling you an identity with the pretext of food.

3.

But what about people with such refined palates that they can’t stand eating anywhere low quality?

Imagine you had a friend that spent $100 a week on luxury flavoured water. They just can’t stand the subtle notes in regular water anymore. I’d say they screwed themselves over by running too far on the hedonic treadmill and they could save a lot of money by untraining their palate.

4.

And just like eating, you can put effort into pooping to make it feel better.

Ever since I started loads of vegetables, I have fantastic poops in the mornings that reliably make me feel aftershocks throughout the day. A healthy meal will fuel me for a day, while a good poop will put a lightness into my attitude, and that’s just as important. And pooping is free.

If your poops aren’t this good, you’re missing out.

Moral Mazes — Robert Jackall — 1988

A sociologist embeds with the managers of 80s corporate America, and studies their lives and worldviews. What is their social reality? What is their day to day experience like? Futher, given that bureaucratic managerial logic is ubiqitous in our society and our lives, what does that do to us?

The following quote may seem like an unreadable run-on sentence, but after reading this book I see a profound statement about American society, as if one fish grabbed the other and exclaimed “we’re living in water!”

Bureaucracy in large American corporations regularizes people’s experiences of time and indeed routinizes their lives by engaging them on a daily basis in rational, socially approved, purposive action; it brings them into daily proximity with and subordination to authority, creating in the process upward-looking stances that have decisive social and psychological consequences; it places a premium on a functionally rational, pragmatic habit of mind that seeks specific goals; and it creates subtle measures of prestige and an elaborate status hierarchy that, in addition to fostering an intense competition for status, also makes the rules, procedures, social contexts, and protocol of an organization paramount psychological and behavioral guides.

It then veers into very interesting chapters on the development of the field of public relations, so interesting indeed that I switched from the audiobook to the paperback so I could pen and highlight. The 2nd edition concludes with a very nice essay on the 2008 financial crisis that I couldn’t find online anywhere.

An excellent book, not too difficult to read, short, and very insightful.

The Years of Lyndon Johnson — Robert A. Caro — 1982, 1990, 2002, 2012

This ~3500 page series of books taught me an enormous amount about midcentury America and its political system, and I feel it generalises so that key parts of the functioning of the world have been demystified to me.

How does the senate work? How do people win elections? How do politicians turn power into money and back into more power? These problems really are so complicated they require months of mental immersion into the political world to understand, which these books granted me.

As always, Caro writes his characters and narratives so well that despite the ostensibly dry subject matter, it’s an easy read after the first few hundred pages as you get emotionally invested.

The story of the 1948 Senate Election in Texas so incredible that I memorised it and tell it to girls on first dates.

This series of four books were the logical continuation of The Power Broker, which was my highest rated book of last year.

Destiny Disrupted: A History of the World Through Islamic Eyes — Tamim Ansary — 2009

The story of the founding of Islam is breathtaking, the story of the first four Imams interesting, and the slide into the corruption is tragic but obviously inevitable.

A repeating pattern: disaster comes to the Muslim world, millions flock to a conservative reactionary movement which promises a return to better days and moral purity, and the situation gets worse. I get it, the promise of Islam is intoxicating and incredible: Dar al-Islam, Sharia, the moral purity, the discipline, and the prayer. Unfortunately the grand Islamic project seems to consistently run head first into the base nature of humans.

After reading this book, talking to Muslims, visiting a mosque during prayer, it’s clear to me that Islam is overall a good thing, the many failures of Arab and that Muslim states don’t detract from Islam as a personal creed. And I respect all those who earnestly try to adhere to it, even if they fail. But those who pretend to be good Muslims on Friday evenings for the social benefits but do not practice any of the virtues, those I scorn.

The Wehrmacht’s Last Stand: The German Campaigns of 1944–1945 — Robert M. Citino — 2012

This book covers two things, an explanation for how is it that the Wehrmacht continued to put up such a fight even when the war was clearly lost for years, and an in depth look at the usually glossed over final year of the war (which is far more interesting than I thought).

Despite the German officer corps widespread understanding there was zero chance of victory and hadn’t been for years, the army kept piecing itself together into launching more hail mary counter-attacks until the bitter end.

How is this possible? How could they know it was pointless, yet still sacrifice their own lives and the lives of their soldiers to continue fighting? And after being thrown back and shattered yet again, why did the soldiers continue to serve? What lead to this absurd and horrible system? The answer lies in Prussian military tradition.

⭐⭐

The Prize: The Epic Quest for Oil, Money, and Power — Daniel Yergin — 1990

A history of oil. The biggest lesson I learned is that oil has antigravity properties: it desires to rise up out the ground. No human effort can stop its march from the ground and into our engines.

Taught me a lot about Middle Eastern politics, how a leveraged buyout works, and oil’s relation to great power security. But despite its enormous length I feel like it was missing some deeper analysis, and was mostly a collection of interesting stories to do with oil.

Savage Continent: Europe in the Aftermath of World War II — Keith Lowe — 2012

The violence in Europe didn’t end with the war.

After peace, the continent experienced catstrophic collapse of its water, electrical, food, policing, legal, and political systems. It was anarchy and the only organisations capable of filling these vacuums were armies, i.e. millions of young men trained only in killing. They were not equipped to recover Europe, and the appetite for more funding and continued mobilization to police and stabilize europe was rapidly dwindling. And yet, all knew that to allow Europe to maintain in this savage state would invite further disaster down the line.

• Europeans returned to their homes and found them occupied, and were told it was a shame the Nazis hadn’t killed them.
• During the war those who used to live side by side had been racialised by the Nazis, once the war was over pogroms, civil wars, and state-sponsored population transfers led to their seperation into ethnically homogenous nation states.
• The Polish took over a Nazi concentration camp and used it re-enact a revenge against Germans.
• The Germans quickly convinced themselves that they too were victims of the war, victims of a dictatorship and the Soviets.

Anyways I was going to write up all these interesting things I learned, but the main thing is to cover this quite interesting period of history that teaches us quite a bit about what happens to civilized people when their civilization collapses around them. It’s not good.

Penguin Great Ideas: Civilization and Its Discontents — Sigmund Freud — 1930

We all have base instincts and desires. To do whatever we want, to kill, to cheat, to litter, to steal. But society — civilization — demands the constant and severe repression of these instincts. This creates resentment and discontentment. Sounds simple, but it’s pretty profound.

I see this in the form of Germany and Japan being very good at this kind of civilizational repression, but it comes out in their high rate of strange fetishes. And woke was another attempt of societal mass repression for prosocial causes, but it caused immense discontent. I dunno, writing this out it seems pretty simple, but really you got to read the original words.

But it’s filled with Freud’s strange theories and is kind of hard to read, so two stars.

Diaspora — Greg Egan — 1997

A very interesting sci-fi book. Heavy on the ideas, weak on the plot and characters. To be honest, the first three or so chapters are the most interesting part, the main plot was boring.

Touching the Void — Joe Simpson — 1988

A great story of a mountaineering expedition gone horribly wrong. Easy to read, gripping, and short.

Death of the Wehrmacht: The German Campaigns of 1942 — Robert M. Citino — 2007

An interesting book for a WWII nerd, but not quite as interesting as The Wehrmacht’s Last Stand.

This Is China: The First 5,000 Years — Berkshire Pub Group — 2021

A very short 120 page book covering five thousand years of Chinese history. I recommend it, mostly as a way to whet your appetite.

Penguin Great Ideas: On the Suffering of the World — Arthur Schopenhauer — 1851

A short collection of excerpts by Schopenhauer. Gotta say, this guy was a genius. Very clear eyed about many things. I don’t remember what I read in this one, but I’m sure the thoughts he introduced have stuck with me.

Penguin Great Ideas: Why I Write — George Orwell — 1946

Orwell writes well and enjoyably. Another one where I’m not quite sure I remember what he wrote about, but I remember it was a great read at the time.

The Soul of a New Machine — Tracy Kidder — 1981

A decent story about an early computer startup.

Project Hail Mary - Andy Weir - 2021

A scifi book for engineers, fun problem solving. I liked it, couldn’t put it down once I got into it.

Fine Structure — qntm — 2017

I really enjoyed this despite the plot holes and some incoherence, it was still very interesting and tense.

⭐

Penguin Great Ideas: On Friendship — Michel de Montaigne — 1580

Mainly this book taught me that people in the past could be really f—king smart. Guy was two centuries ahead of his time.

Meditations — Marcus Aurelius — 180

To be honest, not sure how much I got out of this.

Mickey 7 - Edward Ashton - 2022

Beach read. Not bad, just, not great.

Penguin Great Ideas:A Room of One’s Own — Virginia Woolf — 1929

A small collection of works by Woolf. I don’t remember what I read in this one.

The Last Policeman — Ben H. Winters — 2012

A well researched sci-fi book about what happens to society when we find out an enormous meteor will strike the earth in six months.

The plot and characters aren’t interesting, but the worldbuilding is decent.

Hyperion & The Fall of Hyperion — Dan Simmons — 1989, 1990

Very frustrating books. The first one is amazingly written, until it gets to the completely moronic ballbuster cliffhanger ending. The author might as well have added a postscript saying “buy the next book to find out what happens next!”. And so I did, and I feel he came so close to pulling of something incredible, but right at the end the pieces don’t come together right and they all fall apart.

A Farewell to Arms — Ernest Hemingway — 1929

An easy to read book, but I’m not sure what I was supposed to learn from this one. Hemingway’s a good writer in a technical sense, but I’m starting to wonder if he maybe wasn’t that bright.

The Sailor Who Fell from Grace with the Sea — Yukio Mishima — 1963

Well-written, but this book is just Mishima’s fetishistic fantasies about sailors and death. I was very unhappy after reading this. Especially after reading Confessions of a Mask, this was a huge dissapointment.