# LLM Context File for James Edward Dillard This file contains selected public writings and pages from https://jdilla.xyz, formatted in markdown. It is intended to help LLMs and other tools better understand my work, interests, and perspective. --- # Blog ## Good tokens 2025-10-31 ### Trick or TreatA very happy š Halloween š¦ to you and yours.
Worth your time
The most American Building. My grandfather slept in it while it was unfinished in between training stops in World War 2. My father took classes in it. I went to it on field trips. Itās a wonderful building.
Preach, Nabeel, Preach. I wonder why āeducationā rather than āageā has been what has sorted our politics š¤.
On relationship between growth and trust, arguing that living through periods of higher GDP growth leads to higher societal trust. On one level, this squares well with the idea that trust is a mixture of competence, commitment, and character, with societies delivering growth being seen as competent. On the other hand, I would expect higher levels of trust to also unlock opportunities for faster growth.
Things I learned
Home field advantage in the NFL is actually real and it basically disappeared in 2020 when no fans were in the stadiums. Via CrƩmieux.
More than 98% of new vehicle sales in Norway were EVs in September. From Elective via Anton.
Unconfirmed but from a reliable source: Amazon drivers are paid 12 cents per packaged delivered.
A growing share of Americans (+13%) say Religion is gaining influence in American life according to Pew. Iām not sure how to square this with the thing I learned last week, that support for declaring the United States a Christian Nation is falling amongst Christians or that the fastest growing Catholic sects are the strictest ones. Strange things are happening!
For the first time in 35 years, no rap songs are in the top 40. Rolling Stone.
Musings
Should we care about process or outcomes? Some really successful people (see Tom Brady here ) seem more to favor the process over the result while others favor the result over the process (see Phil Knight / Nike and Sam Altman). How should I make sense of this?
What would have to change for Western Society to become less individualistic? Is it possible for Western Society to become more individualistic? What would it look like to short individualism?
LLM corner
Episode 3 of Dangerously Skip Permissions is next week: LLM pricing is broken, but not in the way that you think with my friend Anjali Shrivastava.
## Supra Podcast: How AI is changing the future of software development ### AI in the Trenches: Tools, Teams, and TransformationsI got to join Marc and Ben from the Supra podcast to talk about how AI is changing how software teams operate.
Three things I took away from this conversation.
First, is that there are some things that AI doesnāt change. At the end of the day, youāve still got to define the problem, define the approach, define the details. AI changes the tools, the artifacts, and the process, but it doesnāt change the basic facts of problem solving.
Second is AI is changing how software is made at three levels simultaneously: individuals, teams, and organizations. Individuals are trying out tools (e.g., Claude Code) and putting them into their workflow. Then there are some teams that are starting to adopt some of these tools en masse and reorganize their processes around them. Finally, there are organizations that are trying to figure out what all of this means for the āstandardā way of working and shipping software.
To get this right, organizations need to be willing to change across 4 dimensions:
Tools - What are the tools that are available to us? What are their benefits and limitations?
Tactics - How do we coordinate with these tools to achieve a result? What are the artifacts that are created? What is the size and roles of people on the team?
Training - How do we build competence on these new tools and tactics? How do we give people space, opportunity, and resources to experiment?
Values - What does great work look like? What is important and celebrated?
Without all of these working together, organizations will fail to get value out of a transformative technology ā and I have to be honest, now is a moment where Iād rather be at a small company experimenting with new ways of working than at a large company where I have to be concerned about how this works at scale.
You can find the whole episode here:
Spotify: https://open.spotify.com/episode/7nozDSwSk3fuAK4TQWxm5l?si=oA0qIwIJShqTCjMOLFFC0Q
Apple: https://podcasts.apple.com/us/podcast/81-i-spent-3-months-at-an-ai-native-startup-where/id1737704130?i=1000733726676
YouTube: https://youtu.be/GbOw8_JViPA
Substack: https://suprainsider.substack.com/p/81-i-spent-3-months-at-an-ai-native
Similar to Shoe Dog ā and different in the way that Nike is different from Trader Joeās.
Three things I want to take away from this book:
I didnāt know whether or not to laugh or cry when he said that Trader Joeās target customer is overeducated and underpaid.
This book helped me better understand how retail, goods, and media are interconnected. The transition from network tv to cable tv happens at the same time as Trader Joeās is shifting away from homogenized consumer packaged goods to the more varied assortment we see today. A similar version of this happened with Facebook and the DTC brands of the 2010s.
## Good Tokens 2025-10-24 ### pigs have banks?The Alpha Terrace Historic District in Pittsburgh, PA. One of my dream places to live.
My guilty pleasure on YouTube right now are videos claiming Ancient Egyptians had access to advanced technology that allowed them to machine vases out of hard stone. Iām agnostic as to whether or not this is true, but I canāt look away! A second thing that makes these videos delightful is that they all pit themselves against mainstream archeology which just cracks me up. Who are these mainstream archeologists? What are they doing to hinder this message? I see the evidence for advanced manufacturing but these mainstream archeologists seem like a mythical species.
Why is Switzerland so rich? This is good, but I think it misses a couple of things. First, Switzerland was spared the physical and human losses of both World Wars. Second, thereās a cultural element that the post doesnāt speak to. Switzerland is both highly individualistic and highly communal, a mix of live-and-let-live and weāre-all-in-this-together that I believe allows it to make more pragmatic decisions, the benefits of which compound over time.
Some of the strongest US-China copium Iāve ever seen.
Creating a village for your child. I wish it were easier to do this.
What happens when someone dies on an airplane. Via Uri.
11 states and half of the counties in the US have more senior citizens than children. This sounds outrageous but Iām curious how much this has changed over time and the degree to which this is just more about longer life spans. Someone should analyze this the way Brian Potter analyzed US pedestrian deaths.
Costcoās Kirkland Brand drives more revenue than all of Procter and Gamble combined.
One of the great joys of having children is that they ask obvious questions you havenāt considered. This week it was: āWhy do we call it a piggy bank?ā 1 It turns out that this (possibly) comes from the name of the clay, pygg, that was used to make jars for storing coins and that shaping them like pigs was a visual pun.
Support for declaring the United States a Christian nation is falling amongst Christians.
Someone told me this week that in France they say that there are six reasons someone will pay for something: Security, Pride, Novelty, Comfort, Money, Friendliness.
The Tiny Teams Playbook. This rhymes with some of what I learned this summer while āinterningā with Roo Code. See also prototype first development.
Dead Framework Theory - the idea that LLMs are freezing frameworks like React into the internet. I thought like this at first, but I no longer think that this is true and I actually think LLMs will make it easier to bootstrap new frameworks provided those frameworks have real advantages over what theyāre replacing because LLMs make it so much easier to adopt new tools.
Peter Steinbergerās Agentic Coding Guide.
Living Dangerously with Claude.
The actual question was much funnier. My 5 year old made a piggy bank at church, causing my 3 year old to ask, āDaddy, do pigs have banks?ā As I think about this, it gets even more puzzling, because I'm not sure he's ever been to a bank. ↩
A message from my sponsor
Person Do Thing is on Amazon. Youāre here so you know Uri already, but Iāll just say that my family loves this one and that it makes a great gift for the person in your life that loves games.
The Resonant Computing Manifesto. See if you see anyone you know š.
The universe as an evolving organism. I have no idea whether or not this is true, but I really enjoy this style of conversation about black holes and space and what we know and what we donāt. There should be more of this.
Frequency reduces difficulty. Via Mark Larson.
What happened to .400 hitters?
āYou have enough ideas to destroy Amazon.ā Not just founders, product managers too.
Monarch butterflies produce a āsuper generationā that live 8x as long as the other generations to allow them to complete their migratory cycle. Again, via Secrets of the Forest, which has to be the childrenās book of the year in the Dillard household. Itās beginning to rival the Kroger App for introducing me to delightful facts. And I donāt say that lightly!
If anyone knows the author, send her my compliments.
Brand Mascots are actually persuasive to children.
This is how the Canadian Supreme Court dresses. And even worse, theyāre changing it. Why, Canada, why?
"Life is 10 per cent what you make it, and 90 per cent how you take it" āIrving Berlin. Sometimes I think the quotes at the end of The Browser our aimed directly at me. I promise you my kids will grow up with this one memorized.
I was on the Demystifying Cosmetics Podcast talking about what I learned making high performance biomaterials from kelp.
If thatās not enough, weāre doing a live show of --dangerously-skip-permissions on Friday at 2 pm ET. Come and hang out.
(I have to be the only person putting out a podcast on beauty ingredients and coding with AI the same week)
Video models are zero-shot learners and reasoners
Mike Judge on the lack of AI software productivity in the data and a response.
## Slop is a choice ### "The devil's oldest strategy is... promising godlike creation without godlike effort"2025 is the year of slop. Meta has made an infinite slop machine. So has OpenAI. If this isnāt bad enough, our jobs have been invaded by workslop.
The implication behind all of this is that slop is the fault of the LLMs or their creators.
But for a moment, I want to ask you to considerā¦ is this true?
I donāt dispute that weāre seeing a lot more AI generated slop than we were 2 years agoā¦ but I do wonder what itās displacing. When I see this AI video of a woman jumping through a glass bridge, I wonder what human content it is replacing in those users feeds. What is the quality of that content? Would it qualify as slop too?
Here is what I believe to be true.
Making something excellent takes care and focus. Sometimes that care and focus is the years you spent prior to the morning you make something excellent and the final thing hops out of you almost fully formed. Sometimes that care and focus is the years you spend refining something until it is excellent.
Most creative work, including mine, isnāt particularly high quality. If you want to be uncharitable, you could call this slop.1 Internet platforms have made it easier for people to create and display their work. LLMs have made it easier to create. This means that we see a lot more slop.
There will be some Sora posts that will be funny, wonderful, even beautiful. There will be many that are slop.
At the end of it all, slop is a choice. My choice and your choice. āThe devilās oldest strategy is, of course, promising godlike creation without godlike effort.ā
I will note for a second that it was not called slop when it was fully human generated low quality content. ↩
Mike Judge has a great piece poking at the AI hype where he asks essentially, āIf these tools are so great, where is the explosion of AI created stuff in the world?ā
The whole piece is worth a read, but one of the most interesting things to me about it is the data he brings to bear on the question.
He looks at:
And then concludes from these that AI coding tools are ābullshitā ending with the call for people who claim that they are now 10x software engineers because of AI, to show the receipts.
First, I want to concentrate on what I love about this. āIf this is so great, where is it in the data?ā is absolutely the right question to be asking.
And there is definitely a dog that isnāt barking here. The data that he cites arenāt perfect (more on this in a second) and yet really impactful things tend to move really obvious metrics. The gains in life expectancy between 1870 and 1970 are really easy to see.
So on one hand, I love this challenge. On the other, I think he goes too far in calling it all bullshit and saying that it doesnāt work.
Iāll hold myself out as the example that Mike asks for. While I wonāt claim to be a 10x engineer, I had never completed a meaningful software project in production before GPT-4 launched. Since then, Iāve coded this blog / portfolio site myself, launched an AI based local news site that has hundreds of weekly readers, and I have a third unreleased prototype that I think could be a real product.
Analyzing myself against Mikeās charts:
Now Iām open to the idea that Iām the exception rather than the ruleā¦ but I also too humble to think that Iām a unicorn on this dimension.
Thereās plenty of room for middle ground here. Itās totally possible that: 1. AI tools are net negative for most software engineers 2. AI tools are transformative for people like me 3. People like me are a minority
Intuitively, I doubt that this is true and yet I donāt have hard data beyond my personal experience to bring to bear on this question. Itās certainly something Iāll be thinking about over the coming months. A more likely explanation in my view is that we havenāt unlocked the right combination of values, tactics, organizational design, and training to unlock AI software productivity at scaleā¦ but I canāt prove that at this point.
Worth a ponder.
## Sympathy for the Devil ### Book thoughts on The Power BrokerThe Power Broker by Robert Caro deserves its reputation as a masterpiece. One of the best books Iāve ever read and up there in my pantheon of non-fiction books with Breaks of the Game and What it Takes.
Reading this book from the perspective of 2025, I think Robert Caro got Robert Moses wrong.1 Yes, he was obsessed with power. Yes, he ruined neighborhoods and destroyed communities with highways and slum clearance programs. Yes, he failed to treat the black citizens of New York City as having equal significance with its white citizens. He had all of these failings and it is important not to shy away from them.
But he was also a genius. An operator and a builder. He reorganized the New York State government into the form it still holds today. He built parks and stadiums and power dams.
Yes, Moses hoarded power. He had sharp elbows. He pushed the limits. He was also prepared, on time, organized, and ready to work tirelessly. He didnāt enrich himself or his family, but instead funneled money into getting power, so he could do more building. He built with all the limitations and blindspots of the generation he grew up in but he did build. He could be petty and attention seeking and yet he was also fantastically productive, not for just himself and his ego, but also for the people of New York.
Itās important while reading the book to pay attention to the other characters coming of age alongside Moses, people like Jimmy Walker, who are comically corrupt by todayās standards. When you do, you get the feeling that Caro is comparing Moses not to the other men and women who couldāve been chosen to lead at the time but to a hypothetical perfect public servant that never did exist and never could exist.
I first heard of Robert Moses in a City Planning Survey course that I took as a sophomore at UNC Chapel Hill. The TA who taught the class had an infectious enthusiasm for city planning and so that summer I found myself reading the Jane Jacobs classic Death and Life of Great American Cities.2
So I came into the book with a sense that the problems with American Cities are downstream of the choices that Robert Moses and his disciples made to orient them towards cars and away from walking and public transportation and neighborhoods with a sense of place.
But after reading the book, I think the problem for American cities isnāt that Robert Moses ruined them, but that there have been no Robert Moseses since: civic leaders with the intellect and power to reshape cities with a new, imperfect vision of what the good life is. We know from Amsterdam and Paris that this sort of transformation is possible and yet weāve chosen to stay stuck in time, living with the same problems year after year.
The Robert Moses of 2025 would probably not favor cars over other forms of public transportation or redline minority neighborhoods. He would be imperfect in new ways that we cannot yet see or understand. But he would build.
Itās a credit to Caroās immense ability that I can read his work and come to very different conclusions about what it means. ↩
Still one of my favorite books to this day. ↩
Best enjoyed this week in a sunny corner of a park
If youāve ever wanted to buy a life sized dinosaur, now is your chance. Someday my son is going to find out I had this opportunity and didnāt take it and will never look at me the same way again.
The Quiet Ones by Nikunj Kothari. An ode to the people that do the little things to make a company or a team effective.
Illiteracy is a policy choice. We donāt talk about Mississippiās education system often enough (although careful readers of Good Tokens will recognize this from a previous edition). Every single state should be studying their approach to literacy. See also the Sold a Story .
Altoids by the fistful. Via my friend Daniel.
I now realize that everything I lorded over other peopleāall the things I gatekept without consciously understanding that this was what I was doingāI didnāt need to do that. It really didnāt help anything. For some number of people who interacted with me, Iwas the problem. I couldāve been more tolerant or forgiving, I couldāve said āletās find out together,ā I couldāve let other people have the fun once in a while.
"The devilās oldest strategy is, of course, promising godlike creation without godlike effort." Slop is a choice.
Iāve become obsessed with the tops of trees, in particular in the morning or the evening when the sun is hitting them. For some insect or bird or leaf that spot is the center of the world.
Letās see if I can land the plane on this one. Iām surprised that there isnāt more nostalgic fiction about growing up in evangelical Christian circles. Thereās satirical stuff like Saved but nothing that Iām aware of like The Big Sick that both pokes fun at being a child of immigrants while also on some level clearly feeling affection for it. Is this out there and I just donāt know about it?
Are we at the point where āyes, andā¦ā is overrated? If not, how long until we get there?
Something I struggled with this week: for someone like you and me, in 2025, what does it mean to live a good life? At 19, it was easier for me to articulate an answer to this question I actually believed than it is now in many ways. If you feel like you have a good answer to this, consider this me humbly requesting that you write it and share it.
Apparently Marie Antoinette never said āLet them eat cakeā, according to a recent Rest Is History Bonus episode. Iām a sucker forthings we think that arenāt actually so. Also from a RIH bonus episode: apparently the US now requires people to share their social media handles to get a travel visa. What are we doing here people?
China installed more industrial robots last year than the rest of the world combined. This is one of those stats that a 17 year old is going to be citing in an AP History Exam in 2084 about why China won the war for Taiwan.
Badgers air out their beds to keep them clean, via Secrets of the Forest.
āYou are going to continue sucking for the rest of your career.ā A call from Nerajno to embrace learning.
Episode 2 of Dangerously Skip Permissions. Mark your calendars, tell your friends. Tell people you donāt even like.
A list of ways to run more than one Claude Code instance at once. I was hoping to build in this space but I may be too late.
The future is compounding teams
Simon Willson on designing agents loops.
What does a UI look like that all users are able to edit? What primitives are needed to build it?
Fuzzy compilers in less than 30 seconds.
Making a note to try out Microsoftās amplifier.
Human / AI synergy and having a theory of mind.
## Good tokens 2025-09-26 ### Mars life clues, smart veggie tactics, and AI musingsThis weekās episode is best paired with a hot cup of coffee and Wild Ways by Josh Ritter playing in the background. Last episode was too LLM heavy, for which I apologize. Iāve done my best to group all of that into LLM corner so as not to let it overshadow everything.
Uri says we should not allow 18 year olds to sign long term contracts. So, so many thoughts here. 1. I remember a conversation I had with my best friend when he was a brand new army officer out of college ROTC about all the 18 year old privates he worked with that had 19% car loans. 2. Jonathan Haidt opened my eyes to the way social media companies get teenagers to agree terms of service that they very obviously should not be able to agree to without their parents consent. I cannot believe we allow this! 3. Matt Levineās Certificate of Dumb Investment continues to seem underrated to me.
It appears we have evidence for life on Mars.
PSA: How to fold fitted sheets, via the Browser. I sent this to my wife and she very nicely said to me something to the effect of āisnāt this the same way I taught you to do it?ā š¤£
"any study of Internet culture is basically a study of crazy peopleā. Also: "Be careful who you pretend to be, because you are who you pretend to be.ā
Dwarkeshās advice for explaining your announcement / launches / blog posts for Twitter.
āWhen outsiders succeed, itās usually through reframing problems in āparadigm shiftsā. They benefit from not being too attached to existing theories.āFrom a thread on outsiders solving problems.
Our parenting hack of the year so far is having cut vegetables ready at the table when our kids get home from school. The percentage of vegetables consumed is up like 10x and compliance to the routine of coming in, washing hands, and sitting down at the table has risen as well. Recommended and thanks to Emily Oster for the suggestion.
German chocolate cake was invented in United States, via the Kroger App. Someone needs to figure out why the Kroger app has so many delightful facts in it. This is someoneās passion project! I'll buy you a nice bottle of wine if you find this person and introduce them to me.
80% of Swiss are satisfied with their lives. I am not sponsored by the Swiss government, but I am open to it if they are reading.
The Pangolin is the only mammal with scales. Peacock is the name of the males only; the female are peahens. The species is called peafowl. Via The Animal Book.
Thereās no such thing a quality time with your kids. My mom said this to me over and over again as child. Itās quantity of time, not quality of time.
The rise of parasitic AI. This is the first moment where Iāve seriously contemplated the AIs taking over.
ChatGPT Is Blowing Up Marriages as Spouses Use AI to Attack Their Partners
āShe does that to her family. She does that to her friends. She does that to me,ā he lamented. āShe doesnāt seem to be capable of creating her own social interactions anymore.ā I worry a lot that the sycophancy of the agents have made me less flexible with people who (of course) are less likely to defer to me. I am not sure how to measure this, but I wish I could.
Sort of a musing, but I think we owe Blake Lemoine and apology.
How to Claude and Claude Code Camp. I want to be on Claude Code Camp.
The Pope says we wonāt find God in the AI.
If you are good at code review, you will be good at using AI agents. I wonder what it would look like to teach editing as a skill. Is there anyone that does this?
A promising approach to prompt injection attacks.
I canāt wait to experiment with Net Dollar.
## Levels of problem solving ### there's levels to thisI'm blatantly stealing this from Matt Holden who taught it to me, but I think about it all the time and I want a reference page to be able to point myself and others too.
When working on a team, there are three levels to work on:
Level 1: Agree on the problem to be solved
Level 2: Agree on the approach to solving the problem
Level 3: Agree on the details
Many disagreements happen when you skip these levels or give level 3 feedback when people are looking for level 1 recognition.
## Book thoughts: Passport to Magonia ### Exploring the eerie parallels between UFOs and folklore.This is a book with a handful of big ideas: 1. There are considerable similarities between the UFO stories of the 1950s and 1960s (then current, the book was published in 1969) and the stories of fairies / angels and demons / other mythical creatures from before the space age 2. Those similarities are interesting even if you donāt believe that UFOs come from extraterrestrial life 3. Even if these phenomena arenāt real the way the Empire State Building is real, they still impact the world in real ways 4. The (then current) UFO stories are folklore in the making, which makes it interesting in its own way
Iāve never gone deep on aliens / UFOs so Iām not up on the lore, but I think most of the points above are now mainstream?
Beyond this, there were a ton of stories about weird things happening, including a series of stories from 1890s America that just seemed bizarre. The one that will stick with me is the Mystery Airship of 1896 and 1897 where (potentially?) an airship floats around the western and midwestern states, occasionally stopping and having conversations with local farmers. You can choose to believe this or not, but either way itās a fun wikipedia read.
On the whole, this increased my belief in the supernatural marginally.
## Good Tokens 2025-09-12 ### AI Insights, Markdown Magic, and Workplace WanderingsShameless self promotion
Matt Holden and I are doing a YouTube show about building with AI called --dangerously-skip-permissions. The first episode is āHow did we get here?ā. Matt and I have been having 1:1 conversations for more than a year now about what tools weāre using and how weāre using themā¦ and now weāre having those conversations in public. I especially enjoy the way that Matt is able to connect whatās happening with LLMs today with previous eras of computing innovation. Give it a listen if thatās your thing!
Matt Holden on Markdown coding
OpenRouter has market share by LLM model. Interesting and unexpected in some ways!
On fact checking with AI. I really enjoyed this one. I have a draft blog post in my head called āVibe Craft: How to do serious work with AIā but every time I try to write it, it falls flat. This is spiritually related to that.
Office building visits are up among people that live less than 5 miles from their office. As someone who made major life changes during the pandemic, I feel the pang of regret.
āA soldier will fight long and hard for a bit of colored ribbon.ā ā Napoleon
āEveryone has a plan until they get punched in the face.ā ā Mike Tyson
## Good tokens 2025-09-05 ### Wildflowers, Luck, and Ancient Math MysteriesReplacing lawns with wildflowers š. When Iāve made it, I wonāt tell anyone, but there will be signs.
Cate Hall on how to increase your surface area for luck, which is one of the biggest things I learned from Henry Oliverās book on Second Acts. Cate is quickly rising up the list of people whose work I rush to read. Along the same lines: How to Get Ahead in DC.
āEven the context has contextā. Wherein Soren blows my mind and sells me on decentralized edge intelligence.
Should I have kissed her? Some how I missed this one in August of 2022. Itās my favorite type of Uri post.
How can you not love this? A 3,700-year-old Babylonian clay tablet holds a trigonometric table more accurate than any today. Humans are amazing and beautiful.
Drones are downing helicopters.
Yucca man. Iām a sucker for ādoes this Bigfoot like creature actually existā stories (see season 1 of the Wild Thing podcast), but this one also has so many great Southern California places in it. Like taking a mini vacation.
Nuclear batteries. āA 157W Voyager-based RTG that launched in 1977 will produce about 88W today.ā The clean up problem seems insurmountable.
Noah Smith, Dan Wang, and James Cham talk about Danās new book Breakneck.
Why Swiss Kids Walk to School Alone. This is one of the things that made me fall in love with Switzerland. They do this as 5 year olds! Part of it is safety but part of it is teaching agency. The walk to school is a part of the education. This should be our aspiration for American neighborhoods.
Your idea sets the ceiling for your videos potential and other good advice from Paddy Galloway.
Chips on shoulders put chips in pockets.
Whatās the steel man case for formality?
What does our society overemphasize now in a way that will seem silly in 25 years?
The secret to engineering is embracing that getting new errors equals progress.
## Good tokens 2025-08-21 ### Mob Mentality, Age Gates, and Tech MusingsA group of kangaroos is called a mob. A group of jaguars is a shadow.
Adding an age check reduced online porn traffic in the UK by 47%. Whether or not you believe it is right, I believe it is unlikely that ten years from now people will be unable to get porn online without verifying their age in some way.
What kids say about getting off their phones. Freedom is the killer app.
Zhengdongwang on whether or not AI is a normal technology. Zhengdongās 2024 letter and productivity advice are some of my favorite recent pieces of internet writing.
Devon Zuegel on how to build a new town. I want to do this someday.
Iām in love with the conspiracy theory that the terra cotta warriors are fake. I donāt believe it is true, but I love going down the rabbit hole. Someone make the definitive YouTube video on this!
Disposable delivery drones are a thing.
Product / Market Experiments:āExperimentation is a skill developed via learning-by-doing, and angels have a skill advantage in that domain because of having more operational experienceā. Filed under āwe all experiment too little.ā
Rooting for Austin Vernon.
Under the hood with Claude Code. Also, AI coding agents and IDEs ranked.
Be the person who writes things down.
Life should be much rougher.
Most of your audience never reads what you wrote. They are told about it by someone who read it. Told to me by patio11.
Some Kipling:
Ancient, effortless, ordered, cycle on cycle set,
Life so long untroubled, that ye who inherit forget
It was not made with the mountains, it is not one with the deep.
Men, not gods, devised it. Men, not gods, must keep.
How do the rhythms of work change when anyone can build a proposed product change? What software is needed to support this?
The most important skill.
## The Social History of the Code Machine ### How technology evolves values and tacticsIāve been reading The Social History of the Machine Gun, which tells the story of the introduction and adoption of automatic weaponry to the battlefield.1 Iām not really a gun person, but I found it fascinating because it is a real life story of how a new technology challenges the values and assumptions of people and institutions. The life and death stakes add weight to the resistance of key leaders to adapt to the implications of the new technology. It caused me to reflect on how AI is changing software development and gave me some practical ideas on how teams and people should be adapting to get the most of the technology.
What is about to follow greatly simplifies large periods of military history, but I believe is a directionally correct description of John Ellisās central argument.
Prior to the deployment of the Gatling Gun, the decisive charge was the center of most military operations. The goal of an army was to time their decisive charge to overwhelm their opponents, break their lines, and take the field. This is how Napoleon fought and not altogether different from how Julius Caesar fought.
As guns ā muskets and cannons ā were introduced to the battlefield, they were introduced in service of the decisive charge (again, radically simplifying). The purpose of lining up lots of men in well ordered lines and firing muskets was to concentrate enough firepower to soften up the enemy ahead for the bayonet charge to come. So central was the decisive charge to battlefield tactics that one late 19th century British Army Captain was quoted in the book as saying that āguns were not as a rule made for actual warfare, but for show.ā 2
The machine gun, starting with the Gatling gun, but later the Maxim and Browning guns changed everything. Different guns have different levels of performance, but one primary source in the book notes that an early machine gun allowed a single soldier to concentrate 40x as much firepower compared to existing methods. Furthermore, this firing speed was reliable; it was the same for new recruits as it was for highly drilled veterans.
Over the following fifty years, in fits and spurts, the ability to concentrate firepower begins to change warfare. At first, machine guns are primarily used in defensive contexts. There is ample evidence in colonial conflicts that charges are useless against them, even in (previously) overwhelming numbers. Then in the Russo-Japanese war, the Japanese pioneered the use of covering fire to execute offensive maneuvers.
Despite these examples, militaries around the world are reluctant to take the evidence in front of them to its logical conclusion and reorganize around the new weapon. As late as 1915, the British Army is placing heavy emphasis on bayonet training and telling its soldiers: āThe bayonetā¦ is the ultimate weapon in battle.ā In Ellisās view, it is the machine gun more than anything else that causes the First World War to turn into a war of attrition and itās only after the war that a true reimagining of tactics begins.
So why were militaries so slow to adopt new technology when the stakes were so high? Ellis makes a persuasive argument that adoption of the machine guns and the tactics enabled by them was hindered by the values of military leaders and the institutions they maintained. One quote from the book in reaction to a demonstration of the Gatling gun: āBut soldiers do not fancy itā¦ it is so foreign to the old familiar action of battle ā that sitting behind a steel blinder and turning a crank āthat enthusiasm dies out; there is no play to the pulses; it does not seem like soldiers work.ā
The new weaponry and the changes in tactics required conflicted with their sense of what it meant to be a good soldier. They couldnāt let go of orderly lines and courageous charges, even under pain of death.
Iām not a military historian, but I am a software creator. While reading this book, Iāve been thinking about AI in general and software development in particular. For at least the last 15 years (my entire career), the assumption has been that code is expensive to create and must be done with extreme careā¦ and that isnāt the case anymore.
Itās easy from the perspective of 2025 to look back at the military elites of the 1890s with their uniforms and funny facial hair and laugh at how backwards they are. I struggled at times to fully believe the stories in the book. Who has such an emotional attachment to how a victory is won?
Itās harder to realize that these were accomplished, intelligent, competent men who had these habits drilled into them and who had literal victories to their names. The values that made them successful had become second nature to them and natures are hard to change.
So how can we learn from their experience?
If I took one thing away from this book itās that our values bleed into our work. Timeless values like remaining disciplined under pressure are expressed in actions like marching in a straight line and we become attached to those actions rather than the values. When technology changes those actions, it feels viscerally wrong to us. I see a lot of this in the discussion around vibe coding. We should be prepared for this feeling and seek to be curious rather than judgmental. Itās never a bad time to reflect upon your essential values!
A second take away was the interaction between values, tactics, organizational design, and training. Unlocking the power of the machine gun required changes in:
Values (e.g., the understanding of what made a good soldier)
Tactics (e.g., machine guns are used differently than other weapons)
Organizational design (e.g., increasing the number of machine gunners in a unit)
Training (e.g., giving units time and resources to master the new technology)
To be effective, these changes had to happen together. This should make intuitive sense. Changing your tactics will be ineffective if you arenāt trained on the tools youāre using and youāll never invest the time in training on something you donāt value.
At the margin, all of us probably experiment too little, but this is even more true now. Throughout the entire book, there was only one anecdote I can remember of a unit overestimating the capabilities of a machine gun and hundreds of people who underestimated it. Often there were pockets of experimentation from outsiders or units operating in atypical circumstances, like the previously mentioned British colonial and Japanese units. Central commands were quick to discount these experiences rather than seeking to understand them.
Taking my own advice, hereās a proposal for what the software team of the future looks like:
Using an agent, (virtually) everyone in the organization has the ability to code, proposing changes to the product. Sales, customer support, marketing operations and more are all attempting to improve the product.
This may even extend to people outside the formal organization ā for instance, customers may be given the ability to propose product changes that first go live only on their account and then are adopted more broadly.
A relatively smaller set of people are tasked with managing the scalability, design, and strategy of the product. Theyāre reviewing working prototypes and thinking about the second order implications, a blend of executives and hands-in-the-code architects, designers, and PMs.
Experimentation with these prototypes becomes much, much more common. New ways of starting, assessing, and sunsetting experiments are needed.
All of this will be heavily mediated by AI agents that both improve the output of the ānon-technicalā team and give leverage to the keepers of product quality.
Despite heavy use of AI, attention to detail and the ability to get into the weeds to make something great will continue to be prized ā if anything, it may become even more important.
All-in-all, it becomes more like a well maintained and opinionated open source project than the standard āthree-in-a-boxā PM / Designer / Engineering lead.
Shoutout to Jordan Schneider whose essential ChinaTalk podcastbrought this to my attention ↩
Ellis does note that this was an extreme position, but the Captain in question was an advisor to Hiram Maxim, one of the early machine gun innovators. ↩
I spent this week improving my HeyRecap agent. Some of the biggest benefits in performance I saw came from improving the tools that my agent had access to. The intelligence of the model was almost never the limiting factor for my use case; instead, flawed design of my search / read features was leading to bad output. Building agents means building tools for agents. More to come on this.
The optimal UI for AI collaboration are coactive surfaces where both the human and the AI are writing and reading from the same set of materials.
I think David Shor is wrong about reading and writing. I do think video is going to raise in importance, but in the age of AI, reading and writing closely is going to be even more important.
## Middlemarch Book Notes ### Up to a certain pointThe first book in ~2.5 years to crack my booklist. This book has more character voices than any book Iāve read in a long time. I can hear so many of them: Dorothea, Celia, Fred, Caleb, Mary, and especially Mr. Brooke in my head.
A rebuke to the idea that cellphones ruin the plot of movies. Almost all of the driving conflicts in this story are about the things people cannot bring themselves to say. Dorothea reminds me of George Bailey: āThe growing good of the world is partly dependent on unhistoric acts; and that things are not so ill with you and me as they might have been, is half owing to the number who lived faithfully a hidden life, and rest in unvisited tombs.ā
I was blown away by the self reliance of the main characters. They are so reluctant to blame others for their problems and/or lash out at them. I wonder if this reflects how society has changed?
I expect someday Iāll read it again. See also Markās āreflections on it.
## Good tokens 2025-07-30 ### Pork buns, royal claims, and AI insightsācatch up on this week's eclectic mix.That became my yardstick: Iād ask, āIs this dish good enough to come downtown and wait in line for? If not, itās not what weāre after.ā A chef can go years before getting another dish like that. Weāve been lucky: Hits have come at the least expected time and place. Iāve spent weeks on one dish that ultimately very few people would care about. And then Iāve spent 15 minutes on something that ends up flooring people like the pork bun.
David Chang on strange loops and food. āIs this good enough to come downtown and wait in line for?ā is going to be my measuring stick for all future projects.
This one is long, for which I apologize. Itās summer time āļø
A Roman trebuchet was called an onager because of the power and danger of its kick, like that of a wild ass. I have a name for my next company. Via ACOUP.
The only place in California that can serve alcohol until 4 am is the VIP club in the Intuit Dome. Via my friend Graham.
Anyone who does not believe in miracles is not a realistā ā Audrey Hepburn (via the Browser)
An emerging pattern Iāve seen with AI start ups is connecting software with physical processes in order to create a durable edge.
## Good tokens 2025-07-03 ### Gettysburg reflections, VTOL wonders, and literary musingsThe third day of the Battle of Gettysburg was 162 years ago today. An amazing reminder of the capacity of America to change and flourish. Glory, glory, hallelujah. His truth is marching on. Happy early birthday, America. I continue to love you and believe in you šŗšø š
Having no experience did not stop me from making a VTOL with world-class range and flight time, all in 90 days.
— Tsung Xu (@tsungxu) June 9, 2025
This is the thing I'm most proud of building to date! pic.twitter.com/iUtNeZqFAw
"The best revenge is to be unlike him who performed the injury" āMarcus Aurelius (via The Browser)
An entrepreneur is someone not limited by the resources directly under their control.
## Claude Code Thoughts ### Here we go againI finally had the space to really test Claude Code as a part of my development workflow last week. I was blown away ā itās as good as advertised. For my coding productivity, itās the biggest step change in productivity in the past year and continues to build my conviction that weāre headed to a world where well defined coding projects are cheap to implement.
My test project was related to this blog. I had chosen SendGrid as an email provider and like a month after I shipped email subscriptions SendGrid got rid of their free tierā¦ and Iām not paying $20 a month to send a handful of emails per week. So I needed to migrate and I had 60 days to do it less I disappoint my adoring public.
I used Claude Code via the Cursor Terminal / Cursor extension. Because of Charlieās advice, I used Claude Opus for the coding.
To begin with, I gave Claude the background:
Where my email service currently lives
Where my email templates currently live
That I need to move from SendGrid to Resend
Claude Code took it from there ā it did a review of my code and concluded this wouldnāt be a particularly impactful change. It reviewed the Resend documentation, outlined the files that needed to change and came back to me with a plan, which I went along with.
Along the way, I skimmed the code changes to make sure they werenāt far reaching in a way that didnāt make sense to me.
The first version of the plan worked but ran into Resendās rate limits, so we had to switch to the batch version of the API. I knew about the batch version of the API from a previous project, so I could provide it with that link. Without that, itās possible that the project wouldāve gotten stuck there.
All In all, once I was set up, it was incredibly painless ā the equivalent of slacking the superstar engineer on the team about a project and suddenly itās done.
I ended up using Opus via the API rather than Claude Sonnet via my Claude account. My experience using Opus via Claude Code will probably convince me to set up to the $100 a month plan here soon ā Iāll easily hit that in API fees working on Recap this week. But itās super confusing that my $20 a month plan gives me access to Claude Opus via the Chat App but not via Claude Code and that I have to log out of my account in Claude Code to use my API credits. Just let me stay logged in and use my API credits!
Setting up Claude Code in Cursor was a pain. I had already installed it via the terminal and had to uninstall it and reinstall it via the Cursor command line. For some reason, the Claude extension isnāt available in the extension marketplace any longer, which is confusing to me.
Claude Codeās auto updates have been failing on me ~1x a week and then I have to uninstall and reinstall. No idea why this is but itās annoying!
This was a really comfortable way for me to code. Iām a vibe coder at heart and integrating more deeply into Cursor / Terminal takes a bunch of the copying and pasting out of it. My loop now looks like: 1. Use the Claude App to define what Iām building (using Harperās scoping prompt) 2. Pull that into a spec (the overall project) and a plan (delivery milestones with a definition of done). Claude writes this and I edit it. I put it in a /prd folder. 3. Have Claude Code read this and create its list. From there we are off.
I think this is going to be a pretty large change for developer tools companies. Previously, if you could get a startup to build around your technology early, you could keep them for a long time simply out of inertia, giving you a fair amount of pricing power. It was expensive to prioritize switching services over feature development and teams would need to feel a fair amount of pain to make a change. Nowā¦ not so much? You still have to do the QA of the change, but the coding should be more or less free for well documented services.
Yet again, Iām going to have to re-evaluate what AI procurement. I have been on the ChatGPT pro plan to have access to Deep Researchā¦ but it might be time to change that.
## How to launch a beauty ingredient ### Lessons learned from an adventure in the world of beauty and specialty ingredientsOver the past several years with Macro Oceans, I found myself in a position I never thought Iād be in: using kelp to launch novel ingredients into the beauty industry (see here and here). Prior to this, I was a Big Tech product manager, totally comfortable in API docs, but totally lost in Sephora.
Suddenly I found myself neck deep in a world I had no experience in: using biotechnology to create beauty ingredients. I started looking around for people who had gone through launching ingredients like thisā¦ and it was rarer than I expected! So I write this post as a gift to past me and (hopefully) current you: what I wish I had known before I began about turning beauty science into beauty products.
If you find yourself with a particular ingredient technology that you think could be the next big thing within beauty, but donāt know how to turn technology into a product, then this post is for you. [0]
There are four primary challenges that have to be overcome to bring a new beauty ingredient technology to market:
Craft your competitive edge: Figuring out what the value proposition is going to be for your product.
Benchmarking: Showing that your ingredient is better than the alternatives
Capturing value: Figuring out how to bring this product to market; how to decide whether or not you should build a consumer brand or go business-to-business?
Safety and regulatory: Navigating the safety and regulatory hurdles in order to bring this novel ingredient to market
Most of my examples below are bioactives, but the same basic principles apply to functional ingredients or colors.
The first and most important question you need to be able to answer is this: why is your ingredient better than whatās available on the market today? This is probably obvious; itās the place where all great products begin.
I like to start with the end consumer. I imagine a shopper at a Sephora walking through an aisle and picking up a product powered by my ingredient. What is that product? Why did they pick it up? What was next to it? What made mine stand out?
To be viable, your ingredient needs to be good enough to stop this person ā a busy person, with a job and a grocery list ā and get her to pick up a box and decide to buy it. If your ingredient consistently passes this bar, itās a winner. If it doesnāt, then it isnāt.
Picturing this scene helps me articulate the benefit of my product. What does it do? If I were standing next to our imaginary shopper, how would I explain the value of this ingredient to her? This is my key benefit.
Now right now, you might be saying āactually, my ingredient isnāt for consumers, itās for formulators.ā That might be true! Iāve launched ingredients like this. But even in this case, I like to imagine this scene because itās the one your brand customer is going to be thinking about when they assess your ingredient. Do you bring them closer to being selected or not?
I try to pick a single benefit to focus on at the beginning. While many ingredients do several useful things, every extra promise increases the proof you need and muddies the message.
Sustainability and a more regenerative supply chain was a major inspiration for our work at Macro Oceans and a big part of our story. However, my experience within the beauty industry was that sustainability is a great complimentary message, but doesnāt drive sales on its own.
This makes sense when you consider it. People are hiring an ingredient to do a job: soothe their skin, strengthen their hair. Theyāll happily choose a more sustainable alternative if it works as well as the existing options, but most wonāt pay a performance penalty to switch except in extreme circumstances.
For this reason, I like to lead with performance and then use sustainability to drive interest and create a brand halo.
At the beginning, my key benefit is a hypothesis: what I think my ingredient could do better than anything else. As I go through benchmarking Iām looking to prove (or disprove) its value by comparing it to the āgold standardā ingredient currently in the market.
Here's what this means in practice, while I was at Macro Oceans, we launched an ingredient with hydrating properties. The dominant ingredient for this use case is hyaluronic acid. To be a viable hydrating ingredient, we had to at least be in the vicinity of hyaluronic acidās performance ā and ideally noticeably better.
I canāt stress this enough: itās not enough to simply measure the benefits of your ingredient. You need to benchmark it against the thing that people are using to solve this problem today. Doing this gives you a relevant comparison point for your ingredientās performance. Speaking from personal experience, having this data available makes a night and day difference in terms of driving demand.
In practice, benchmarking and defining your value proposition are iterative. You start with an initial hypothesis of value, do some measurement, and then your hypothesis may shift. This is normal.
There is a list, but it isnāt written down. As an outsider to the beauty industry and someone not trained in cosmetic science, I would start by asking ChatGPT: āWhat are some of the best ingredients for [my use case]? How is their effectiveness measured?ā
Then I would take the list it gave me and compare it with the ingredient lists of products in the real world. As I went, Iād add or remove ingredients based on what I was seeing.
Finally, I would take this list and review it with cosmetic chemists I respect. By the end of this, I would have an idea of the most relevant ingredients to compare to.
Once you have a comparison point, you need to figure out how youāre going to measure your ingredient versus whatās in the market.
For me, this was often the most difficult part of the process, requiring a mix of vendor outreach, working with the more technical members of our team or external experts, and reading about different test methods in academic papers.
I wanted someone to have a series of product stage gates they could take me through that would help me derisk our ingredient iteratively, but never really found the person or organization I was looking for.
In the end, we created it for ourselves with a fair amount of external input: first, a series of lab tests that were (relatively speaking) faster and cheaper, followed by secondary tests that were longer and more expensive, followed by a 30+ person clinical trial. The goal here is to derisk quickly and avoid costly mistakes.
Itās critical to do benchmarking at realistic use rates for the respective ingredients ā both your ingredient and your competition. This gives you a āpound for poundā comparison between the ingredients, the way they would be used in an actual product. It also helps you speak to cost (more on this in a moment).
Extending the hyaluronic acid example, there are hyaluronic acids with suggested usages as low as 0.1%. Meanwhile, our ingredient had a suggested use rate of 2-4%. So a fair test between the ingredients is the hydrating performance of hyaluronic acid at 0.1% vs. the hydration performance of our ingredient at 2%. If we are outperforming hyaluronic acid at our suggested use rate, then we are on to something interesting and if weāre underperforming it, we arenāt.
If youāre starting from scratch, you may not know your use rate. In this case, I think it makes sense to test a handful of different candidate use rates, see how they perform, and use that to guide your thinking.
With the use rate, you can now bring cost into the equation. Even if youāre going to be selling this ingredient to contract manufacturers by the kilogram, you should assess your ingredient in terms of cost-in-use. Doing this shows how much expense it adds to the final product and takes your āpound-for-poundā comparison and makes it a ādollar-for-dollarā comparison. Ideally youāre showing more performance per dollar than your competition.
Hereās an example that shows why this matters:
Product A: $1,000 per kg at a 0.1% use rate \= $1 cost in use
Product B: $250 per kg at a 2% use rate \= $5 cost in use
All else being equal, Product A adds less cost to the product in the bottle than Product B despite being more expensive on a per kg basis.
With a proven ingredient, you need to decide how to best capture the value created by the new ingredient: selling to consumers or selling business-to-business.
The great thing about selling directly to consumers is that you own your own destiny. Youāre not waiting on another business to adopt your ingredient and integrate it into their product roadmap. You also get the opportunity to educate consumers directly about the unique benefits that your product provides and why they should choose your product. If this goes well, it can go incredibly quickly and be quite lucrative. K18 is a testament to this.
The downside to selling directly to consumers is that itās competitive. New brands pop up frequently and many consumers like to experiment with new products. It can be difficult to stand out in the first place and even more difficult to sustain this over time. Additionally, youāre going to have to build out a consumer brand with a consumer friendly website and marketing capabilities, a task that can be daunting if youāre most comfortable in the lab.
The great thing about selling business-to-business is that you have a broader set of value propositions to choose from. A five cent improvement in cost-in-use for equivalent performance isnāt a big deal to the end consumer, but could be a meaningful improvement to a business buyer. Itās also sticky revenue; once a brand puts an ingredient into a product, they tend to stick with it for the life of the product. Reformulating is a pain!
The downside of selling business-to-business is that it takes a long time to scale up. Most customers are going to take at least 12 months to adopt a new ingredient and even longer to ramp up their purchasing. The business side of the beauty industry is complex and can be difficult to navigate. This makes for a long waiting game and this additional time creates risk for your product: your sale could get delayed because your point of contact changes jobs, has a baby, or loses their funding ā all real things that happened to me.
Ultimately, I think the test is this: if a typical consumer can see or feel the difference between a product using your ingredient and the other products on the market right away, then it makes sense to go directly to consumers; if not, then youāll want to go business-to-business, no matter how much better your benchmarks are.
At Macro Oceans, we ended up deciding to go business-to-business. For our company, strategy, and ingredient profile, I think this was the right decision ā we never had breakout performance in a consumer facing category ā but we really felt the pain of convincing brands and contract manufacturers to adopt our ingredients.
Let me say upfront: I canāt give safety or regulatory advice for your specific product. But I can share what helped me stay sane through this part of the process.
Safety and regulatory work caused me more anxiety than any other part of the work ā not because it was technically harder, but because I worried about missing something critical. Another thing that made it challenging is that there is a mixture of hard and soft requirements. Some things were obviously required, while others were requested by specific customers as a part of their policy, but not legally mandated.
I build my working list in a way that was really similar to creating my list of āgold standardā ingredients:
Asking ChatGPT and searching online
Reviewing competitor SDS and sell sheets
Validating with a regulatory consultant
Talking with potential customers
With the set of requirements and timelines in hand, I did my best to align the required work with different stages of the product development lifecycle. Things like toxicology data or HRIPT testing came early, because they were required for benchmark testing on humans, but INCI registration came later because I knew I could sample my ingredient with potential customers while I was waiting for the committee to approve my application. Still other things ā like China registration ā could be put off indefinitely because we knew that market wasnāt our focus early on.
This helped me avoid two potential pitfalls: wasting money on non-viable ingredients, and delaying launch because I started critical tests too late.
Especially as we began engaging larger customers, it was pretty common for us to get asked for safety data that wasnāt relevant for our ingredient. The most absurd example of this was when I was asked for the bovine content of my all natural / vegan kelp extract.
Eventually I learned that the person asking me for this stuff was working off of a procurement checklist and needed a paper trail to show that their organization had done its due diligence and that all I needed was an attestation from the company that this was not relevant in our case.
If youāve made it this far, this is obviously a topic you care a lot about! Hopefully this walk through my experience helps you on your journey.
While Iāve covered a lot of ground, in some ways Iāve just scratched the surface, skipping things like finding first customers, aligning sales and production, and how and where to tell your story. Thereās always more to say!
If youāre working on a project like this right now and could use a sounding board, donāt hesitate to reach out: hello@jdilla.xyz.
[0]: Based on my early forays into other industries like materials and food, a lot of these same lessons carry over.
Thank you to Aaron Nesser and Jesse Adler for providing feedback on this post. All its remaining flaws are my fault, not theirs.
## A taxonomy of the beauty industry ### Decoding the beauty industryTwo years ago, I left a career as a Big Tech PM to join a scrappy seaweed start up selling novel ingredients into the beauty industry.
In the process, I got a crash course on the beauty industry and the different companies and people that make the products that fill our stores and homes. Iām not going to say itās inscrutable, but it did take some time to get my head around.
In general, I think people write down stuff like this less than they should. Itās not proprietary, it doesnāt take much time to do, and it can be immensely helpful to people trying to learn a new field.
In this spirit (and because I love taxonomies), here is a simple overview of the different players in the beauty industry, the types of people that work at them, and some notes on what motivates them. Iāve put specific examples of these types of players in parentheses below because it always helps me to be able to match a real example to the type.
Of course this is a simplified view of the industry, ignoring some of the ways that organizations in the space play across these boundaries or some of the other roles involved (e.g., regulatory consultants), but I think itās useful as an initial primer.
Brands (Drunk Elephant): Brands market and sell products to consumers either directly through their websites or through retailers They can be majors (EstƩe Lauder) or independent (Dieux). Most brands do not actually manufacture products they sell themselves, but do this through Contract Manufacturers. Many of them also outsource their formulation, either to Contract Manufacturers or Formulators. Brands are motivated to grow revenue and build consumer loyalty by launching distinctive, on-trend products that balance performance, cost, and storytelling.
Retailers (Sephora): Retailers sell products that brands make to consumers. They typically do not make their own products or if they do have their own line, itās just part of what they carry. They give consumers a central place to browse products across brands and categories and create value through their selection. Retailers aim to maximize sales and customer retention by curating compelling assortments that drive traffic and basket size across multiple brands.āØ
Contract manufacturers / CMs (Allure Beauty Concepts): Contract manufacturers make products on behalf of Brands. They buy ingredients and packaging and then turn those into products that people buy for Brands. They frequently hire cosmetic chemists in house to create formulas for the brands that they work for, but will also take formulas Brands have created themselves or through Custom Formulators.Ā Contract manufacturers are focused on keeping their production lines full and profitable by securing recurring brand partnerships and offering value-added services like formulation.āØ
Formulators / Custom formulators (KKT Labs): Formulators are cosmetic chemists that sell Brands a formula that they can take to a Contract Manufacturer and have that made into a product consumers can buy. They are usually independent consultants, but sometimes small teams. Brands sometimes prefer to work with a Custom Formulator because then they own the formula and can shop it across multiple Contract Manufacturers to try and get a better price (when the CM formulates on behalf of the Brand, it is typical for them to own the formula rather than the Brand). Custom formulators are motivated to build trust and repeat work by offering differentiated formulas and acting as an outsourced R&D partner for brandsāØ
Suppliers / Ingredient manufacturers (Mibelle): These companies make ingredients for use in beauty products. They sell to Contract Manufacturers, but have to influence Brands and Custom Formulators in order to drive sales. They typically work with Distributors who represent their products to these different groups. Ingredient suppliers seek to grow sales and adoption by getting their ingredients specified in hero formulas and building trust through efficacy data, claims support, and education.āØ
Distributors (Catalyst Technologies): Distributors sign distribution relationships with Ingredient Suppliers to represent and sell their products to Brands and Contract Manufacturers. They try to have a balanced portfolio of different ingredients that makes them more relevant to a Cosmetic Chemist than any one Ingredient company could be on their own. They act as a salesforce for the Suppliers. Distributors are driven to increase sales across their ingredient portfolio by being a trusted, convenient resource for cosmetic chemists and formulators.
## Book thoughts: Hell Yeah or No by Derek Sivers ### Not a book reviewSuper easy to read.
Has some moments of real clarity and brilliance and some moments that seem self indulgent.
A great book to pick up in the middle of a life change. It isnāt going to tell you what to do, but will get you thinking. Will absolutely stay on my bookshelf.
The three chapters that resonated with me most this time around:
## Good tokens 2025-06-14 ### Frugality, flavor profiles, and Einstein's eureka momentMore Romans were killed at Cannae than Americans in Vietnam or British on the first day of the Somme ā The Rest is History
āTo create anything worthwhile, you have to put God in it.ā ā Nabeel Qu
## Good tokens 2025-05-27 ### AI Fantasies, Slate Trucks, and the Art of CompromiseCommercial success demanded blending science and marketing; a steelmaker had to recognize not just the value of a new alloy, but its potential use. Benno Strauss, of the Krupp Works, later spoke about recognizing the potential of his stainless steel in plumbing, cutlery, medical equipment, and mirrors. He, like Brearleyāwho realized his stainless steel would be useful in spindles, pistons, plungers, and valvesāwas focused.
One step back, two steps forward
āResearch on third-grade retention policies [holding kids back in 3rd grade] has found that students who are retained tend to have better long-term outcomes than those who are notā from this article on the Mississippi Miracle.
Trade laws of nature?
The distance elasticity of trade (the rate at which trade between two cities drops off as they get farther away from each other) seems to be the same today as it was in ancient Assyria.
The fact that exposure therapy works with phobias (e.g., if youāre afraid of airplanes, the cure is actually getting on a plane and seeing that it works out okay) makes me more sympathetic that the idea that one should act brave in order to become brave.
## Good tokens 2025-05-16 ### From LLM features to the history of tomatoes and pricing tacticsIn an attempt to make my writing more LLM friendly Iāve added an llms.txt file and a /llms feature where anyone ā you included ā can copy ~all of my writing for use in your chatbot of choice. Like most things around here, I built this for myself, but perhaps it will be useful to you too!
āMaybe all one can do is hope to end up with the right regretsā ā Arthur Miller (via The Browser)
## The hobby selection 2x2 ### Picking Hobbies: Balance, Burnout, or Boost?
When I read Range in 2019, one of the things that stood out to me was the role that hobbies have in the lives of high performers. As an example, Nobel Prize winners are 12x as likely to have a creative hobby than their scientific counterparts.
This really stuck with me. I became much more willing to engage in work outside of work. Side projects, hobbies, explorations: I welcomed all of these in my life, sometimes even when I didnāt feel like I had time for them.
On the whole, Iād say this has been a positive thing. Most of them havenāt gone anywhere, but some of them have, bringing people and opportunities into my life that I wouldnāt have had if I hadnāt stretched to take them on. Moreover, some of them seemed to provide their own energy, and even a balancing effect to my life. But this wasnāt consistent. Other times, I felt exhausted or burdened by them, beyond the normal speed bumps that come with typical work.
What was the difference? I didnāt really know.
Then earlier this year, David Epstein (the author of Range) posted about this paper in the Journal of Vocational behavior that explained it for me. The authors compared the impact of hobbies on self efficacy, your belief in your ability to do things, and found a consistent pattern when analyzing the seriousness and intensity of the hobbies.
Put simply, if a hobby is very serious and very similar to your core work, itās going draw you down. This should make intuitive sense: itās drawing on the same energy you need for your day job.
But this isnāt the end of the story. The authors found that serious hobbies that are different than from your core work donāt have this impact. Having something you pursue seriously that is outside your focus can increase your self efficacy. Speaking personally, I find that hobbies like this help me take setbacks in my day job in stride by reminding me that itās just one part of who I am and give me an outlet for my creativity when itās stifled at work.
If you are going to do something in an area thatās close to your focus area, then it needs to be more casual. This is what makes hack week projects so fun: the chance to explore a familiar domain with fresh eyes and fewer constraints. The playfulness of it is the point.
Overtime, Iāve come to visualize this as a 2x2. Iām keeping it as a reminder for when I select my next project: what role do I want this to play in my life? Is it meant to build mastery? Offer escape? Recharge me? Stretch me?
If itās not my main focus, I want to be intentional about how it fits. Does it complement my work or quietly compete with it? Does it give me energy, or siphon it off?
## Notes on "How Will You Measure Your Life" ### Revisiting Wisdom: Balancing Life's Scores and PrinciplesI keep a semi-secret list of great reads that I periodically revisit during a period of change or when the spirit moves me.
One of my favorite on this list is Clayton Christensen's "How Will You Measure Your Life", which I revisited this week.
Some things that stood out this time:
Over the years Iāve watched the fates of my HBS classmates from 1979 unfold; Iāve seen more and more of them come to reunions unhappy, divorced, and alienated from their children. I can guarantee you that not a single one of them graduated with the deliberate strategy of getting divorced and raising children who would become estranged from them. And yet a shocking number of them implemented that strategy. The reason? They didnāt keep the purpose of their lives front and center as they decided how to spend their time, talents, and energy.
And:
When I was a Rhodes scholar, I was in a very demanding academic program, trying to cram an extra yearās worth of work into my time at Oxford. I decided to spend an hour every night reading, thinking, and praying about why God put me on this earth. That was a very challenging commitment to keep, because every hour I spent doing that, I wasnāt studying applied econometrics. I was conflicted about whether I could really afford to take that time away from my studies, but I stuck with itāand ultimately figured out the purpose of my life.
And:
The lesson I learned from this is that itās easier to hold to your principles 100% of the time than it is to hold to them 98% of the time.
I think this is the first time I've read this one closely since I've had children. At the very least, the first time since my youngest one was born. The grind of parenting is much more real to me now than when I've read this previously. I've described my life recently as a "5 on 3 power play where I'm the 3 and it doesn't reset when a goal is scored." It's good to have the reminder to keep some of my energy for my children and not to spend it all elsewhere.
## Good Tokens 2025-05-09 ### AI Fantasies, Intensive Learning, and a Coke-Free AmericaCoca Cola became cocaine free in 1929! Suddenly American history makes more sense. No wonder so much got built between 1880 and 1930. No wonder the market crashed! From the AP via Stan Veuger.
Like a lot of parents, weāre trying to teach our toddlers how to swim. For about a year, we did weekly lessons with my son and daughter and they improved, but ever so slightly. Then over winter break, they did much more intensive work, something like 10 lessons in 12 days. The improvement was dramatic, more than a year of weekly lessons. I wonder what other skills are like this? Language seems like it; immersion seems to outperform regular practice. Is things that require rewiring your brain? I wish I understood this better.
## Good tokens 2025-05-02 ### Autonomous rides, loving your job, and lessons from show runningWaymo or way less than youād expect?
Waymo is now carrying 250,000 passengers/week in autonomous cars. Transit agencies are falling behind on automation, which can improve service & save money. It's exciting to see DC's Metro taking a step forward in automating its existing lines. We need to be talking about autonomous buses, too.
— Yonah Freemark (@yonahfreemark.com) April 24, 2025 at 6:29 PM
[image or embed]
Misery loves company
I remember feeling a version of this when I was at my last job. I really loved my job, my boss, my colleagues. and it was hard to talk about it without sounding contrived, especially eg in a group setting where everyone's going around sharing horror stories. but it's the truth!!
— Visakan Veerasamy (@visakanv) April 29, 2025
This is 100% true of being a parent. My typical experience of being a father is 95% āI canāt believe I get to hangout with these three little people that I loveā and 5% āIām so tired I canāt see straight.ā But itās almost impossible to communicate the 95% to someone besides my wife or my parents. I donāt know why this is, but itās true!
The 11 laws of show running
Very applicable to new product development.
Are you strong and secure enough in your talent and accomplishment to accept the possibility that other people - properly empowered by you - can actually enhance your genius... or will you cling to the idea that only you can be the source of that genius?
Also, things I learned: āa second definition of "nice" is also "precise and demanding careful attention.ā
## Good tokens 2025-04-25 ### Skyscrapers, blog nostalgia, and AI bottlenecksNew York State of Mind
Manhattan, 1931. A city without skyscrapers, save for a handful in the Financial District and the brand new Empire State. A city of 3-5 story buildings.
— šššš šÆ (@atlanticesque) April 20, 2025
Yet more people lived there in 1931 than today. pic.twitter.com/yYKkXzIQaZ
I love Chris Ryan
Chris Ryan on The Press Box talking about the early blogosphere. Three things I loved about this:
Things Brian Potter has learned
Number 30 was my favorite:
Whenever thereās a major bridge incident in the US we hear stories about the USās crumbling infrastructure, but the worst bridges in the US are steadily getting fixed. Between 1992 and 2023, the number of US bridges in critical conditionĀ declined by more than 70%.
Read the whole list here.
AI bottlenecks
An exploration on where the value from AI will come from that also starts to articulate specific bottlenecks that (currently) AI faces in improving R&D work. Somewhat related to my reaction to Situational Awareness, I suspect that more of these bottlenecks exist than people think. Coding might be a unique application for LLMs: relatively closed loop, fast feedback, lower diversity of tasks.
Too online
From No Honor Among Mutuals:
Self-importance, contempt, and arrogance is rewarded online. Virtue rarely is. In this way, technology is inverting many of the incentives for developing character.
Bias hacking for progress?
āOnce you put that first stake in, theyāll never make you pull it up.ā ā Robert Moses, from The Power Broker.
Iāve seen this same dynamic in all sorts of projects. Creating the impression that it is happening unlocks funding that is unavailable before it has begun. It occurs to me that this is a way of hacking peoples sunk cost bias to get things done.
## Chattanooga travel notes ### Trains, aquariums, and wondering about the purpose of a city
Iāve been hearing whispers in the parents-of-young-children circles about Chattanooga since we moved back to the Atlanta area 3 years ago.
We would discuss family friendly travel destinations and Chattanooga would come up, but I never really got it. Whatās in Chattanooga?
Theyād say things like ātheyāve got the Chattanooga Choo Chooā or āthereās a great aquariumā. My children like a train and some fish as much as anyoneās, but I struggled to see what would make this worth the 2 hour drive.
This spring break, we finally went for it, and now I get it. Itās a beautiful, accessible city with a ton of options for things to do with children under 5. Not many of those things are unique to Chattanooga per se, but Iām not sure how much that matters at this stage.
I have to say, I was surprised and more than a little bit charmed. The city itself is like a combination of Pittsburgh and Raleigh. The downtown urban core clearly came of age in the ~1880s to ~1920s and was formerly industrial. Thereās a ridge above the city and a river that runs through it. Itās small. You could walk across it in an hour or two, even with a stroller.
The city is clearly prioritizing tourism. Thereās a free electric bus that connects the city and makes it easy to get around (and for children the age mine are, it is an activity in an of itself). Itās clean and well marked. The aquarium and childrenās museum are new and have a lot of local support. Thereās a hint of Disney World about it, but itās fun (to be fair, the North Shore neighborhood seemed a lot more vibrant, like people actually live there).
The Chattanooga Choo Choo, a classic steam engine in the old train station that kids can climb up on and play in. Things I learned: Chattanoogaās claim to fame is that it connected the Northern and Southern railway systems in the 1880s.
Thereās a great aquarium, at least as good as Atlanta or Monterrey Bay.
Great parks ā Coolidge Park was my favorite. Itās got fountains and a carousel. It used to be that you could walk from the Riverfront (where the aquarium is) to the North Shore (where Coolidge Park is) but thatās currently closed for renovations.
Rock City is just outside of town. This is probably the most unique Chattanooga thing that we saw. Itās tough to describe, but worth it. Plenty to see and do ā and an incredible view of the valley.
Childrenās museum. Is it just me, or have these gotten way better since I was a kid? Basically everywhere I travel with my kids has one (Pittsburgh, Hendersonville, apparently Chattanooga). Do they actually exist year round or are they pop ups that get set up when I book a trip?
Walking through the beautiful old train station made me a little bit sad. Itās a gorgeous building that clearly isnāt being maintained. Itās an elegant building with a sense of place, but the big central corridor is just empty. It reminded me of Buenos Aires: all those beautiful buildings slowly falling into disrepair.
This sadness seeped into the rest of the trip. Chattanooga is clearly doing great. Thereās a lot of investment in downtown. Itās safe and clean. Many of the restaurants and stores are obviously new. My family will definitely go back.
And yet, I couldnāt help but noticing a lot of empty storefronts. Not enough to detract from the experience and all well maintained. But one in four, in some places as many one in two are empty.
Observing this in the background of Liberation Day and all the discourse that followed had me wondering: how prosperous can a city be if it is only a tourist destination, rather than a place that people live and create?
## Good tokens 2025-04-18 ### Oxygen-rich dinosaurs, hidden Christians, and the SynBio bustIf you want to get my posts via email, now you can: jdilla.xyz/subscribe.
The air dinosaurs breathed had more oxygen in it than the air we breathe. Jurassic Park couldnāt have happened because the dinosaurs would asphyxiate. Via Lyman Stone.
The combined number of hidden Christians in China and India is estimated to be 120 million, large enough to be the fifth largest religion in the world. Via Diarmaid MacCulloch.
āA personās success in life is determined by having a high minimum, not a high maximum.ā ā Donald Knuth via Mark Larson.
From Stay Sassy:
I noticed that many parents of young children, despite having significantlyĀ moreĀ on their plates, seemed to get burnt outĀ less. I even noticed this in myself, and didnāt have a real way to explain it ā my first kidās birth coincided with the busiest working period of my life (do not recommend), but I found that I had a more positive attitude towards work, for no reason that I could really explain. When I read this post, it all clicked ā when you have young children Mission Doubt entirely disappears because you need to feed them.
I have also noticed that having a baby in the house is good for my creativity and relationship to my work. While I have less time for work, what I get out of that time is much, much higher quality.
Whatās behind the SynBio Bust? by Sarah Constantin
Risk of Ruin Goods by Uri Bram
Molson Hart on how tariffs and manufacturing in the USA: part 1 and part 2
How restricting car traffic changed air pollution in Paris
this is actually incredible pic.twitter.com/VdcZU94bqd
— sam (@sam_d_1995) April 12, 2025
The myth of the infrastructure phase. Applications and the infrastructure to power them are built in tandem.
Thereās no speed limit.
āThe first principle is that you must not fool yourself and you are the easiest person to fool.ā ā Richard Feynman
## Book notes: Christianity: The First Three Thousand Years ### A long walk through the history of my faithAn incredibly ambitious book by Diarmaid MacCulloch. He really does cover it all, from the Council of Nicaea to the intricacies of African Christianity in the post-colonial era.
I picked this one up mostly to get a better understanding of the 200 years after the birth of Christ and then stuck around for the rest because Iām a completionist. Despite growing up in the church, thereās a bunch of tradition outside of the gospels that I wish someone would put in one place. As an example, Thomas went to India! I continue to think that this would make a great podcast.
After listening to this book (I did the audiobook ā no way I was ever going to get through this one in print), I think I have a better understanding of why: ātraditionā is a friendly stand in for āsources outside the Bibleā and discussion of these makes people uncomfortable. Iām not particularly sure why, but this is a theme of the book ā what goes into the cannon of scripture gets people riled up!
Some other topics like this:
What is the nature of Jesus? Fully God? Fully man? How exactly can one be both?
What is the trinity? Is the trinity one or three?
What is the human role in salvation? To what degree are we choosing Christ or being chosen?
One of the benefits of the breadth of the book was seeing these topics reoccurring throughout times, places, and cultures. It made me feel better about the parts of my faith I simply donāt fully understand.
I was surprised at how much of the story I already knew. Between The Rest Is History, Literature and History, and Dominion, I have a pretty workable understanding of the story of the beginnings of Judaism and Christianity. I'm also surprised that less of this is taught in churches.
There parts of the history, say from between 500 and 1700 in Europe, where centering the history on what was happening within the church rather than what was happening with political leaders clarified the story for me. A reminder that the separation of religion and state is a modern one. It also helps explain why religious details that seem uninteresting in 2025 were all consuming. These were the organizing principles of society, in the way that gender, sexuality, and citizenship are today.
I gained a deeper appreciation for Eastern Christianity and Syrian Christianity in particular. I might be mangling some of the details here, but the book mentioned that one of the Syrian churches has chants that seem to be unchanged since the early 100s AD. Isnāt that an incredible lineage? Augustine also seems more important to me now.
My next book in this area is going to be on the early Christian sources. Suggestions are welcomed!
## Good tokens 2025-04-03 ### Apollo shirts, torpedo bats, and the ancient life of mushroomsApollo by James Edward Dillard.
Paul Skenes is the youngest Opening Day starter in Pirates history since 1893. pic.twitter.com/3Aj7bqgmJL
— Pittsburgh Pirates (@Pirates) March 27, 2025
A single fungal organism can ~live for thousands of years~ and ~span over miles~. Their vast underground webs are largely invisible to us but ~communicate impossibly complex information~ we barely know how to decode. They are among the oldest life forms on earth, ~predating plants by more than 300 million years~.
Do not end the week with nothing by Patrick McKenzie
The Internet of Beefs by Venkatesh Rao
How to run major projects by Ben Kuhn, via Mark Larson
The Decline of Industrial American Science was an interesting read. It actually made me think of this conversation about stagnation in beauty ingredients thatās happening in the beauty industry. I wonder if theyāre related!
Uri Bram on 80/20 weight loss: āStill, as with ~other entries in this series~, the people who care deeply about stuff are often unwilling to write up an 80/20 version of it, so you get me instead.ā Petition for Uri to become the 80/20 guy. This is a good lane for you, Uri!
How to be good at dating <- applicable to things besides just dating! Breaking the problem down and then actually changing behavior to get different results works surprisingly well provided youāre willing to do it. Often success doesnāt come to us the way we want to receive it. I canāt remember where I read this, but somewhere someone posted about how Harry Potter ruined a generation of children because he just wakes up one day and finds out that heās this ridiculously special wizard, when in reality itās Hermione we should be admiring because has to work to be great. If you know who wrote this and can point me to it, come find me!
## Hyman Rickover celebration hour17. If you canāt say ānoā easily, you canāt be trusted.
— Joe Hudson (@FU_joehudson) December 22, 2024
Loving this deep dive on Hyman Rickover by ChinaTalkand Charles Yang. A couple quotes:
Rickover spent an inordinate amount of time focused on interviewing personnel ā he made the final hiring decision for every naval officer who applied to serve on a nuclear submarine until he retired
Another Rickoverian approach was his famous āQuaker meetingsā. When disputes arose between the Naval Reactors and the contractor, or when trust had become frayed over too many disagreements and miscommunications, Rickover would send his staff and the contractor staff to a retreat location for a weekend, a week, or however long as needed. They would meet with no parliamentary procedures or formal meeting agenda and simply talk out their issues until they could ādeal with each others as individuals, not as spokesman for either organizationā and come to a consensus on a path forward and build mutual trust.
Some other themes that stood out:
Parallel tracking wherever there is risk (technical risk, bureaucratic risk)
A focus on consensus building
Focused on finding talent
Demanding of that talent
High trust relationships
Iām repeatedly struck by how the leadership style of the generation of Americans that lived through the Second World War. It seems more pragmatic, demanding, and less hierarchical / political.
## Book Thoughts: The Shadow of the WindSet in a noir version of post Spanish Civil War Barcelona where somehow everyone is one degree removed from a brilliant yet fantastically unsuccessful author.
It's alternate name, The Cemetery of Forgotten Books is a better one, I think.
The plot was a little just so, still I enjoyed it enough not to put it down. A big plus was the chance to travel to Barcelona through the book.
Some quotes from the I enjoyed:
āThe moment you stop to think about whether you love someone, youāve already stopped loving that person forever.ā
āThe nurse knew that those who really love, love in silence, with deeds and not with words.ā
## Good Tokens 2025-03-21The making of Richard Scarry. Cars and Trucks and Things That Go has been the book of the year in my house, so I loved this behind the scenes look at how the book and its author came to be. Even better that much of it was written in my beloved Switzerland.
Let their be more biographies of failure by Henry Oliver
So what are the lessons we can learn? It doesnāt always help to be right. Ideas arenāt easy to implement without the right combination of technology, attitudes, and luck. The work is whatās important, not the result. Maybe the cranks who fill their houses with cart loads of ephemera arenāt so crazy. Donāt make political trouble. Get a PR department. Have a partner who can do these things if you canāt. Be in the right place at the right time. Donāt get cynical, or as Churchill said, donāt let the bastards grind you down. Keep working. Philosophical and ethical beliefs matter a lot to what work you do and how you do it. Donāt be so pragmatic you end up being a conformist. Conventional schooling isnāt always the best approach for your children. Worry less about imaginative young people becoming lawyers. Being bored might give them the opportunity they need to have their big idea.
A great critique of Seeing like a State from Slate Star Codex. Iām like 1/3 of the way through the book and fully buying Scottās arguments. Now I feel like someone has revealed the magicians trick.
A way to think about which jobs are most likely to be automated by AI: time horizons, legibility, humanity, and trust.
The model is the product. Iām not sure this is correct but the hypothesis is clear and it made me think. Iām not sure Iām ready to bet against generalist scaling, but this was a compelling case that specialized models effectively are the application layer for AI.
Cars have pop up ads now.
The importance of serendipitous meetings: Silicon Valley companies will cross reference each otherās patents more when their employees frequent the same coffee shops. Iām reminded here of Austinās 3 types of luck and the fact that serendipity can be encouraged.
All innovation (particularly social innovation) should be presented as a return to tradition.
## Good tokens 2025-03-14Individual posts now have related posts at the bottom of them, leveraging the infrastructure I built for Search. Iām excited to see what serendipity this sparks. Let me know what you think!
Life expectancy for dogs has been growing faster than life expectancy for humans, via my fascinating friend Uri.
Henrik Karlsson on Constraints. āHave you walked face-first into the wall to see if it is a chalk line?ā We are all trapped in our preconceptions of what our ideal life needs to be like.
The skill of troubleshooting by Autodidacts via The Browser. A great deep dive into how to be a better troubleshooter and also a fantastic example of a meta skill that we probably all have under developed, reminiscent of Oliver Trimboliās work on Listening. A favorite quote from the post: āTreating a system like the enemy makes it one.ā
Will Boom build a Supersonic Airplane by Brian Potter.
Startup Strategy for Commodity Products by Austin Vernon.
Taylor Swiftās security mindset
## Good tokens 2025-03-07Taylor Swift's security practices around her songs before they're released are literally airgap. Ed Sheeran is one of her closest friends in the world. She didn't override anything for him. Because security is a systemic risk and he is a target.
— SwiftOnSecurity (@swiftonsecurity.com) March 9, 2025 at 2:34 PM
[image or embed]
D.A.R.E. anti-drug interventions in schools may have increased drug use among suburban students. Drug Library via Atoms vs. Bits.
The beauty industry faces many of the same materials food, textiles, and industrial applications do. Xanthan Gum is a clean beauty ingredient and also commonly used in fracking.
Hobbies promote self efficacy (the belief that you can do hard things) when:
they are very serious but dissimilar to your work (e.g., a scientist that is a committed rock climber)
they are not particularly serious, but similar to your work (e.g., coding on a side project for a software engineer)
Hobbies begin to reduce self efficacy when they are too serious and too related to work (e.g., intensive blogging for a technical writer)
When written out this is pretty intuitive, but nice to see on paper. Making a personal commitment to find at least one hobby that is truly dissimilar from my work by the end of the year next year. From the Journal of Vocational Behavior via Range Widely.
The Pennsylvania Amish originally hail from Switzerland.
Humans are not the only animal to domesticate other animals. Black garden ants keep aphids in a manner similar to how humans keep livestock. Via Kevin Kelly.
Weāve cracked the code on Roman concrete. Damascus steel next?
Simon Wilson on how to code with LLMs. His point around manual testing certainly matches my experience!
You donāt own the story, but you do own the execution.
A common thread I noticed between Paul Grahamās The Origins of Woke and Tanner Greerās The Euro American Split: the importance of generational change in culture change. We are always either rebelling against or seeking the approval of our elders.
Having a new child is like meeting your spouse, having your first date, falling in love, and marrying them in the span of a moment.
Being boring is a choice.
A half baked business idea: The Anti-Recruiting Firm: You identify the lowest performers at your company; we land them plum roles at your rivals, simultaneously improving your productivity while tanking the competition.
## Good Tokens 2025-02-28Iāve made some changes to the blog:
Iāve added search. Increasingly this blog has become my scratch pad for ideas that are worth saving, a more public version of my various notes apps. Occasionally I look back for something but I canāt find it. Eventually this drove me crazy, so I added keyword search. Try it out and let me know if it works for you.
The home page is now a landing page rather than my blog directly, with the idea that itās a softer landing spot for people who are searching for me for the first time and itās less embarrassing when Iām in a dry spell.
Iāve fallen in love with Literature and History by Doug Metzger. I picked it up because it covered the Epic of Gilgamesh which is one of those things I wanted to know more about but never really wanted to read a book about. That episode was good, but the Homer episodes about the Iliad and the Odyssey were next level.
A decision coach who will make decisions for you for $247. Iām definitely going to try her out for something!
āMy hack to-do list is empty because I built everything.ā ā Harper Reed on his LLM workflow.
Product 101: My thoughts what you need to know if youāre starting a product role. I wrote this last year for a friend of a friend and have found myself sharing it periodically. Now Iām sharing it with you!
I think Iām in love with TimothĆ©e Chalamet.
## The difficulty level of childrenLast week I quoted Greg LeMond on cycling.
I was thinking about doing great work when I posted it, but it also applies to kids. Having your first child feels hard (fun, but hard), until you have a second one. Then having one seems easy in comparison, but having two seems hard. This continues with child three, which is all I can speak to from personal experience, but I assume it continues onward from there.
It also runs the other direction. If you have two kids, and one kid is away (with a grandparent), it feels like having zero kids.
## Good tokens 2025-02-21Matt Holden on Fuzzy computing.
The Polymerist on the Courtship of Sampling. Making a self reminder to write down my thoughts on this one at some point.
The only right protected in the main body of the US Constitution is the right to intellectual property (remember that the Bill of Rights were amendments added later). It seems fitting to me that in the American Brain, the right to own ones ideas is even more primal than the right to free speech. Also, the first patent examiner for the United States was none other than Thomas Jefferson. Via ChinaTalk.
North Pole temperatures were a full 36 ĀŗF above their seasonal average earlier this month. Uh oh!
People need a certain amount of structure to create bonds. You need the conference schedule to create the interactions that lead to friendships. Itās like the coral for the fish.
āIt never gets easier you just go fasterā - Greg LeMond
Culture and values are underrated for explaining the behavior of people and nations. This point was made by Prof. David Kang on ChinaTalk when describing why Asia doesnāt exhibit the same international relations patterns as Europe, but it also made me think of the book Iām listening to Christianity: The First Three Thousand Years. When someoneās values and their material self interest align, you always know what theyāll do. When they conflict, though, often peopleās values win out over what would narrowly benefit them.
## Good tokens 2025-02-07Worth your time
A deep dive on how a clothing brand prices their clothes.
Ethan Mollick on Deep Research. Iāve been working with it quite a bit this week and generally speaking been impressed with it.
āAll great work is preparing yourself for the accident to happenā ā Sidney Lumet, via The Browser
An underrated truth about the modern world is that everyone reads their @mentions and sees who likes their posts
The Luka trade is a reminder that even at the highest levels, mistakes genuinely do happen. You would think that evaluating the market for one of the 5 best basketball players in the world would be an efficient marketā¦ but it appears that it wasnāt!
Reflecting on the ROI of marketing efforts Iāve done recently
Print isnāt that useful unless itās with a writer with a voice (e.g., a substack)
Audio and video really make an impact. You want to be inside someoneās EarPods.
Speaking at trade shows is helpful in expanding your network
Asking good questions is more important than ever
On the modern internet, one should never be self congratulatory. Itās totally okay to accept compliments from your audience, but the moment you start saying or implying youāre great at something or have it figured out, you begin to sew the seeds of your downfall.
## What if OpenAI actually does have a moat?My bias is to think that the most useful AI is the one that has your personal context ā your documents, your emails, and all that stuff.
But despite the fact that ~100% of my useful documents are with Apple or Google, I almost never find myself using Apple AI or Gemini, and instead copying and pasting into Claude or ChatGPT. I have meticulously organized project folders in Google Drive, yet Iām never asking Gemini about them.
Does this mean:
My intuition is wrong - model quality is actually more important; over time, our data will move to where the models are. Owning the user was less valuable than we thought it was.
My intuition is right, but itās still too early; Apple and Google are big, but eventually theyāll catch up and Iāll start using them
Apple and Google are failing in the way that companies fail when paradigms shift and new competitors will rise up to take their place
Thereās a stickiness in how models work. Iāve developed a feel for how to use Claude / ChatGPT, what theyāre good at and when theyāll bite me, and thereās a lock in quality here. I did feel a sense of exhaustion at having to learn DeepSeek r-1 despite being excited to have a new player on the scene.
Iām the problem. Iām not as exploratory of a consumer as I think I am and actually thereās a low friction way to use Gemini for these tasks that I havenāt figured out yet. I donāt think itās this one because itās pretty common for me to choose Chat GPT 4o when I want an answer fast vs. Claude when I have a project already set up with my key documents in it vs. o1-pro / Deep Research (or one of the other models now) when I need a higher quality answer
Something else Iām not thinking of?
Here is what Claude thinks:
## How GLP-1 drugs change consumer spendingThis is an interesting reflection on AI adoption patterns. I think the stickiness factor plays a significant role - there's a real cognitive investment in learning how to effectively "speak" with a particular AI model, understanding its strengths and quirks, which creates a natural barrier to switching even when alternatives might offer theoretical advantages in terms of data integration. This parallels how humans often stick with familiar tools and workflows even when "better" options exist, because the familiarity itself has concrete value.
I think GLP-1 drugs might be the most underrated technological change currently underway. We're talking about big shifts in what people eat, how they look and feel about themselves that run counter to the last ~50 years of consumer trends.
With this in mind, I found this Cricana report to be eye opening.
First, I love that the report using real transaction data - this isn't what about what consumers are saying, but what they're doing compared to what they were doing before.
Spending went up on:
š« Drinkware and water filtration (+28%)
āļø Fitness trackers and wearables (+183%)
š§“ Skin care (+12%)
š Lip care products (+14%)
And down on:
š„ Refrigerated salad dressing (-19.5%)
š„© Dried meat snacks (-17%)
š· Alcoholic beverages (-10-12%)
My simple model here is that people drift away from unhealthy foods and towards healthier ones. As they do this, they start getting out more and doing more things (see the spending on wearables, cosmetics). They're also spending more on the things that mitigate the downsides of the drug (mints for bad breath, tea to soothe stomachs).
The entire thing is worth a read.
(Hat tip to Dan Frommer whose New Consumer newsletter flagged this for me)
## Good tokens 2025-01-31Rohit on AGI.
Noah Smith on China Talk. Interesting throughout but what I enjoyed most was him talking about the types of posts he writes: 1. Things he already understands that are topical (he can just sit down and write) 2. Things where he is going out and doing the research because he thinks something should be better understood 3. Things he is passively interested in and just collects links as he goes, so when it becomes topical, he can go back and find those things and quickly write them. I think I can adopt some of this into my own work.
The short case for Nvidia, but also a fantastic explainer on how different approaches to AI work. I now understand what makes Groq special.
Goals:
## First impressions on OperatorIāll never forget standing in your kitchen in Oakland, watching you juggle a toddler and two babies while making margaritas and asking you about your next venture - what problem you wanted to solve. You looked at me replied āI want to bring commercial super-sonic flightā
— Karl (@KarlGoetze) January 28, 2025
I tried using Operator on a couple of tasks.
The most successful one was drafting an update to a document. We have a Partnerships document that we use as a part of our sales materials and we're onboarding a new partner. I needed to go to the Partner's website and draft some language for them to edit / approve.
I've been putting this off for a couple of weeks so I fed it to Operator. It took a little nudging, but eventually it was able to read through their site and write passable copy that I could refine and send to the partner for their review. So that's a success. Worth $200 a month? If I have 2-3 tasks like this each month, I think so.
I also tried seeing if I could get Operator to compile information for me (e.g., create a CSV file with the meeting dates and times of the Roswell City Council for the rest of January). It failed here in two ways:
This past summer, I tried out several other products in this space (e.g., AutoTab) and this is a big step beyond what was possible then... but still not there yet.
More to come here over time!
## Good tokens 2025-01-24Things I learned
āIn 2022, adults spent an additional 99 minutes at home on any given day compared with 2003.ā
— BuccoCapital Bloke (@buccocapital) January 12, 2025
Increasingly, we are living an isolated and remote existence. Great article from @DKThomp pic.twitter.com/jEeW2BI4G6
A thread on the Shroud of Turin. Worth it even (especially?) if you think the shroud is fake.
## Good tokens 2025-01-17Uri Bram on Noble Lies
Zheng Dong Wang on productivity. Iāll be reading through each of the documents on his list at some point this year.
Why did everything take so long?
Principles by Nabeel Qureshi
Pine needle tea has more than 100 percent of the vitamin C of orange juice ā Nautilus
The value of returned purchases in the United States would make it the 16th largest economy in the world ā Rohit
āGreat problems have to be discovered; often the solution of the problem is only a tiny part of the story, most of it is really about discovering the problem.ā ā From Michael Nielsen, ~Quick thoughts on research:~ (found via Zheng Dong Wang)
## Good tokens 2025-01-10A message from my sponsor:
OceanMade has announced pre sales of its Kelp Pots. These seed starter pots use kelp pulp to retain water instead of the traditional peat. The kelp pulp used in these pots is a byproduct of Macro Oceans beauty ingredient, Big Kelp Hydration. Iāve gotten a chance to see some of these up close and Iām really excited to see them coming together. Itās a small example of a big dream: using traceable, ocean farmed kelp products as an alternative to higher impact terrestrial sources. Due check them out if youāre a gardener.
Michael Lewisās story about Chris Marks, a public servant who āled the development of industry-wide standards and practices to prevent roof falls in underground mines, leading to the first year (2016) of no roof fall fatalities in the United Statesā, is fantastic. Some gems:
At the height of the Vietnam War, a coal miner was nearly as likely to be killed on the job as an American soldier in uniform was to die in combat, and far more likely to be injured. (And that didnāt include some massive number of deaths that would one day follow from black lung disease.)
And
People facing a complicated problem measure whatever they can easily measure. But the measurements by themselves donāt lead to understanding.
And
Roof bolts were indeed more efficient and effective than timber supports in preventing chunks of roof from wounding miners. But they were expensive to install. The coal mine companies had, in effect, figured out how few roof bolts they needed to use to maintain the same level of risk their miners had endured before their invention
Materials we have run out of by Ed Conway
Noah Smith on Japanese urbanism. Having zones that restrict certain activities rather than prescribe what can be done seems like a small change with a big impact.
From Zheng Dong Wangās fabulous 2024 letter:
The first awesome conclusion of the model does the eval is that we will achieve every evaluation we can state. Recall that evaluations must be legible, fast, and either a good approximation of a wanted capability or useful itself.
And:
Two years ago, ~Demis Hassabis enumerated~ three properties of problems suitable for AI: a massive combinatorial search space, a clear objective function to optimize against, and lots of data or an efficient simulator.
Chicken tikka masala originated in Scotland; kilts did not ā Jason Crawford
Squirrels are becoming carnivorous ā Smithsonian Magazineļæ¼
All of the worldās gold is estimated to fit in one 20 meter cube ā BBC
There are more people under the age of 25 today in Africa than there are in all of Europe ā Stephen Kotkin
All large scale changes should be presented as a return to the past.
I wonder what it would look like to restructure local government around an escalating set of reviews. Imagine filing for a building permit where:
The first level of the form is evaluated by AI with the ability to appeal
The second level goes to a human
The third level goes to a supervisor
The second and third levels become new evaluation cases. This already happens today at places like YouTube, but imagine bringing it to your local government.
Never confuse a single defeat with a final defeat
ā F. Scott Fitzgerald via The Browser
## LosingThe proper response to losing, I learned as a child, was to hate it. It should make you miserable, frustrated, and sad. Feeling terrible about it was a good thing, because you would work even harder to make sure you never felt that way again.
I like to think that I got this from Michael Jordan. The impression I have of him was that he hated losing more than anyone else and because of this, he could will himself to win, no matter obstacles lay before him.
I donāt know about him, but it didnāt work for me. More than once, I quit or didnāt try my hardest to avoid the pain of losing.
Now I think the best response to losing is to take it as easily as possible. Maybe not enjoyment, but bemusement if you can manage it. Analyze it, but donāt stress it. The important thing is to continue to try.
## Book notes: Second Act by Henry OliverSecond Act is a book about late bloomers. I listened to the audio book.
In my mental library, this book is part of a trilogy with Range and Talent about how to do your best work.
The book probably only gets published because itās about late bloomers, but I canāt think of any part of it that is only applicable to late bloomers.
I thought about survivorship bias basically the entire time I was listening to the book. Some of it is definitely embedded in here, but some wisdom is too.
A common theme in Oliverās late bloomers is earnestness. Earnestness to the point of being annoying to their contemporaries. I think earnestness is a quality that ages really well.
Many of the lessons I took from this book can be reduced to the sorts of things a youth baseball coach would say to me during practice. This is related to the earnestness.
The need to move through periods of exploration and exploitation at different stages of a career is a lens that will stick with me. If you think your potential is capped in your current situation, itās probably time to turn the dial towards exploration. This is not one I got in youth sports.
Another lens Iāll remember is āmaking yourself a big target for luckā. The book introduced me to Austinās types of luck:
Networks are important because of the influence they have on your aspirations. You need to be around people that expand your idea of whatās possible through words and actions.
Itās really important to (appropriately) display your work. People canāt bump into you if they donāt know you exist.
Caring is a source of alpha. Ray Kroc was one of the late bloomers. McDonaldās dominance made more sense when I better understood how much Ray Kroc cared. His passion for french fries isnāt something I share, but it makes sense that he of all people created the dominant fast food company. He cared more than anyone else!
Being a little reckless can be a good thing as you age. He cites a study (I think this one) where people who make a life change by flipping a coin are ultimately happier when it forces them to change rather than stick with the the status quo.
People who keep trying have more successes and more failures than those that donāt. Chaos and failure are not to be avoided but part of taking many chances at success. You do your best work when you do your most work. Quantity precedes quality. āYou miss 100% of the shots you donāt take.ā
Courage / not counting yourself out is underrated. Believing that you have the ability to be excellent is not sufficient for becoming excellent but it is necessary. This is increasingly important with age. Persistence and determination alone are omnipotent.
Recommended if you, like me, hope your best work is ahead of you.
## Good tokens 2024-12-20Many wonderful public servants made valiant efforts and scored some great wins, but Democratic leadership did not make it a top priority to clear out the underbrush that jams the gears of government.
From Bringing Elon to a knife fight
New rules for media; from #19:
Every second for the viewer is just that viral video where the person picks between two pop stars. Youāre always deciding what to pay attention to. The relationship between person-who-makes and person-who-consumes is paramount to long-term success, because if you are winning that game then you will be able to survive.
Eyes have evolved more than 50 times - Salon via Rohit
You can only avoid competition by avoiding good ideas. ā Paul Graham
Product market fit provides a business with gravity
- It lets you know up from down ā this helps
- But it also weighs on you; itās tough to take the business in a direction that your current product / market / customer isnāt pulling you
A key skill for the future is going to be how to work with something that is:
1. smarter than you in many / most domains
2. sometimes wrong
A surprising amount of life is figuring out the right words to say in order to get what you want
## Good tokens 2024-12-13Already my first correction of my 2024 list: it appears that 75% of the world's saffron probably does not end up in Fernet. Thanks to Tom Whitwell for the correction.
Need a gc filled with people sending esoteric pieces of fact back and forth. For instance, Saturn's rings are younger than the dinosaurs.
— rohit (@krishnanrohit) December 8, 2024
Infanticide facts from Pacific Standard:
From my friend Oriana: The more males and females of a bird species look alike (e.g. cranes), the more likely they are to mate for life.Ā The more males and females of a bird species look dissimilar (e.g. mallards), the more likely they are to be promiscuous.
The oldest bond in the world dates from December 10, 1624; pays ā¬13.61 of interest a year. From the FT via The Browser.
Ben James on Fusion. Canāt wait to read the rest of his guides.
Some thinking on how companies get penalized for trying to be more sustainable by the Green Beauty Community. I do think that one reaction to the backlash on greenwashing has been some companies pulling back from talking about what theyāre doing.
this is the japanese snow fairy š¤
— Nature's masterpiece (@nature-view.bsky.social) December 8, 2024 at 5:53 AM
[image or embed]
Write shitty first drafts is another way of saying that quality is derived from quantity.
## Coding stack - Dec 2024My coding stack as of December 2024 looks like this:
I keep a running Google doc that functions roughly like a PRD:
I have this connected to a Claude project so it stays synced. In this Claude project, Iāve got the ~5-10 most critical files for the project at hand also uploaded.
In my Claude project, Iāve got some background information about my tech stack (e.g., NextJS app, Iām using yarn, not npm, Iām on a mac) and some guidance (I want you to challenge my thinking rather than flatter me; ask questions if you have them; I prefer simple solutions).
I sit down and start chatting with Claude on whatever I happen to be working on (at the moment, uploading PDF files and extracting key data). I chat through different considerations and then create or update a file. Iām using Cursor as my text editor, but really only rarely using the AI features at this point as I was running into too many places where it would spiral off and start making changes I didnāt need and couldnāt follow. Claude feels better at staying on the task at hand. I probably wonāt re-up my subscription when itās done.
I usually code in 2-3 hour blocks. When one of these blocks is ending, I ask Claude to summarize what weāve been working on:
I then paste this into my Google doc and Iām done for the day, picking back up at the beginning the next day.
## Learning How To Learn booknotesLearning How To Learn, suggested to me my Melanie.
The first part of the book was too dense for me. But when the book gets into how the Montessori classroom works, the role of the child and the role of the teacher, it really shines. It is a continuous struggle for me not to expect my children to behave as adults.
Some gems:
## Token stream 2024-12-05Children are taught to speed up their performance by an emphasis on completing a task or test accurately within a fixed time. The young child has, in his own view, all the time in the world. And he needs this time. And he needs this time. The number of perpetually harassed mothers who tell their children to stop dawdling and get it finished, whether āitā be supper or dressing, is legion.
The action has merit above and beyond the actual physical fact of the childās accomplishment. It has the merit of allowing the child to participate in the society in which he finds himself, not at the level of an adult, but at the level of an emerging individual. The importance of a strong sense of self can be seen when we think of the tasks which will be demanded of an American child of three in a few short years. The ability to work independently, to continue to accomplish, whether or not the adult is physically present at oneās elbow at all times, the ability to initiate work because one has had previous successful experience, are important learning skills for a child. Many children are so conditioned by adults that they will refuse to attempt anything new until they have been given either explicit directions by an adult or, what is more frequent, explicit approval to do so by an adult.
The rhythm of the child is a rhythm different from the adult. The child works at a thing until he is satisfied. The teacher has no foolproof way of knowing when this point is reached. The teacher must constantly guard against over-teaching and over-correctingācorrecting a child who is unaware that he has made an error, intervening to show a child how to improve a skill he has barely learned. Respecting at all times the childās right to help himself, and to solicit help only when he feels it is needed, requires tremendous patience.
Literally worth your time:
Telling the time by sundial and history not only was custom, but also was understood as following Godās time. The idea of overriding traditional timekeeping because of the needs of the modern world seemed positively sacrilegious. āPeopleā¦must eat, sleep and workā¦by railroad time,ā wrote a contributor to the Indianapolis Daily Sentinel. āPeople will have to marry by railroad timeā¦. Ministers will be required to preach by railroad timeā¦. Banks will open and close by railroad time; notes will be paid or protested by railroad time.āĀ
From Heather Cox Richardson on the standardization of time.
Skunkworks rules by Eric Gilliam:
- Reduce the bureaucracy to almost zero. Ideally, one person should have almost complete authority over day-to-day decision-making.
- Keep the team ruthlessly small.
- Whenever possible, only take on contracts where there is enough mutual trust with funders and subcontractors to work with them with a minimum amount of bureaucracy. If funder decisions cannot be made swiftly, the project is probably not worth pursuing.
Iād add: Build ambitious things on short timelines. And a bonus quote from Kelly Johnson: āThe theory of the Skunk Works is to learn how to do things quickly and cheaply and to tailor the systems to the degree of risk. There is no one good way to build all airplanes.ā
āWhat makes a good business is industry structure.ā ā Cal Paterson on the business potential of LLMs.
What long context windows mean for how AI will change work.
AI and material design: AI_innovation. Very relevant to my work at Macro Oceans.
From Sapiens on the origins of suicide:
More typically, among todayās ~U.S. high schoolĀ students~, 60Ā percent say they have considered killing themselves, and 14 percent have thought about it seriously in the past year.
Panda related merchandise made up half of all Atlanta Zoo merchandise sales ā AJC
Between 1986 and 2018, 12.3M hectares of cropland in the United States was abandoned. Note that this figure does not include urbanization or development. Environmental Research Letters.
āIt wasnāt that Dario had the best ideas, although he had plentyā¦ he just ran 10 to 100 times as many experiments as anyone else. Thatās when I knew he would do amazing things.ā
From Dion Almaer, Principles of Developer AI Product Development. Reminds me of āquantity precedes quality.ā
My latest AI hack for getting out writing more quickly (company project docs, blog posts): * Record an audio file where I dictate as much as I can about a project * Add that + any other relevant documentation to NotebookLM * Ask NotebookLM to write a first draft for me * Edit that into the shape I want it
I still end up editing out ~50-75% but it gets me passed the empty page as quickly as possible. I estimate I turn a ~half day writing block into something I can do on a 15-20 minute walk + a little computer organization.
People only decide to buy something when they really, really care about it. Otherwise itās not worth the friction of getting out your wallet.
āThe answers are always inside the problem, not outside.ā ā Marshall McLuhan via Gordon
āIf you're not living on the edge you're taking up too much space.ā ā Stephen Hunt via Ade Oshineye
## Things I learned in 2024Pisgah National Forrest, North Carolina, USA. My favorite place I visited in 2024.
I borrowed this concept from Tom Whitwell as a way of cultivating a habit of curiosity. You can read his 2024 versionĀ here. I didnāt make it to 52 things this year, but I stayed curious.
My 2024 highlights: My third child and second daughter was born. I helped Macro Oceans scale a regenerative kelp economy and received an Emergent Ventures grant for my work on HeyRecap, a local news AI experiment.
Here are the things I learned along the way:
If you think weād have an interesting conversation about kelp, local news, our anything else, send me an email (jdilla.xyz@gmail.com). Iād love to meet you!
## Cornbread sausage dressingI adapted this from a recipe my mother gave me. It's now what I make each year. Enjoy and let me know what you think. Happy Thanksgiving!
Spray two 9x13 baking dishes with cooking spray
Cook the Meat:
Leave the drippings in the pan
Cook the Vegetables:
Cook until heated through
Combine Ingredients:
In a very large mixing bowl, combine:
Add Liquid:
You may not need all 2 cups
Bake:
If you're in tech and not in the Bay Area, you lack either judgment or ambition.
— Flo Crivello (@Altimor) November 15, 2024
Floās tweet is my favorite scissor statement of 2024. I felt the sharp edge immediately. Iāve had a career in tech and I like to think of myself as ambitiousā¦ and yet when faced with the decision about where to live, we chose proximity to family over proximity to opportunity. At least I can comfort myself that I donāt lack judgement.
Floās follow up tweet puts the issue less divisively:
where you live is the single most important career decision you'll make. You should make it with your eyes wide open
No matter what you think about living in the Bay Area, this is almost definitely true. But what if, either by choice or constraint, you find yourself outside a superstar city, but still want to have an ambitious career? What should you do then?
Here is my advice:
Keep your ambitions high. Almost every project benefits from considering how to make it 10x more ambitious; even if you donāt take that path, youāll benefit from the thought exercise.
One of the secrets of the Bay Area is the expectation that you can do something that changes the world. Merely inviting people to do great work increases the likelihood that they will do it, so make a practice of inviting yourself to do it.
Notice that impactful work can happen from anywhere. Consider that Nike, perhaps the worldās most iconic fashion company, is headquartered in Oregon. Ben Thompson, maybe the most influential writer in tech, lives in Taipei. Mr. Beast might be the worldās most popular entertainer and he lives in Eastern North Carolina. All of these are existence proof that geography isnāt destiny. Great work is never the default path so donāt waste your time worrying about what youāre missing out on.
Choose your projects wisely. Some projects benefit more from network knowledge than others. As an example, there are going to be a whole set of business ideas that fall out of what people at frontier AI labs understand that will be tough if not impossible to access from outside those networks. Thatās fine! There are many problems worth solving. When youāre choosing your work, assess the network tax youāre paying and steer towards ones where this is lower where you can. Keep the ambition high!
Every disadvantage has its advantage. Being outside the center gives you an outsiderās perspective; use it. Extending the AI example, there are going to be applications of AI that wonāt be visible to those inside the Bay Area bubble because they solve problems people inside the bubble donāt see. Enjoy being outside the groupthink that leads to Uber for dog walkers.
Be a big fish in a small pond. Most cities, towns, and regions, want to be more like superstar cities and are looking for companies or organizations of their own they can boost. You should be the one theyāre boosting! Tobi LĆ¼tke, founder of Shopify, has talked about how he was able to make Shopify into a regional talent magnet. Duolingo does this in Pittsburgh. Startups in the Bay Area have to compete with OpenAI, Google, and Meta for talent, but you can be the best possible choice in your own backyard.
Plan regular trips to the city most connected with your industry and work. Focus those trips on time spent with people as much as you can. Keep your laptop closed. Use your trip as an excuse to bring people together. Attending the right conference does this as well.
Join distributed networks. Some of these are selective (Emergent Ventures, Supra) and others are generally open (StartUp CPG). Use these as a way to meet people and make the opportunity to see them in person if you can.
Learn to build relationships remotely. Building rapport with people that you mostly know from online is different than how you would do it in person. Be more intentional about reaching out, checking in, making up inside jokes. Emojis and memes are your friend.
Set up virtual coffees. If you write a thoughtful note about a problem or topic your target is interested in and if you seem thoughtful and interesting, most people will be willing to set up a virtual meeting with you. The time demand on a virtual coffee is almost always less than an in person one.
Always be posting. Having an online presence will be more important for you than it will be for others. Practice this and use it as a magnet for your work.
There are lots of ways to do great work with focus, intentionality, and creativity. Donāt let your location stop you.
Have anything I missed? Send me an email jdilla.xyz@gmail.com. Iād love to hear it.
## Token stream 2024-11-22Smiling was once considered a sign of drunkeness. Upworthy.
ā¢š"A banana contains the same amount of radiation as a person would get from living next to a properly maintained nuclear power plant for one year"@NYT pic.twitter.com/pb4gF7m090
— James Pethokoukis ā©ļøā¤“ļø (@JimPethokoukis) November 17, 2024
A pro / con list means the answer is no.
every decision i've ever made that wasn't an instant "hell yes" has been a mistake. every pro/con list is an admission the thing isn't worth it. the path God has laid out for you is so obvious it feels like getting bludgeoned with an anvil. no analysis makes a bad decision good
— Will Manidis (@WillManidis) November 20, 2024
The essence of being a generalist is:
If you try and tell people 5 interesting things about your product / company / cause, theyāll remember zero. If instead, you tell them just one, theyāll usually ask questions that lead them to the other things, and then theyāll remember all of them because it mattered to them at the moment they asked.
Modern social media rewards information abundance, so if you find yourself with a product / company / cause that has lots of benefits, tell each of those story one at time. People are more likely to remember it and it gives you more to post.
America was supposed to be Art Deco.
## Token stream 2024-11-15Macro Oceans received a grant to use kelp for sustainable aviation fuel. From the oceans to the skies!
78 percent of Christmas hits were penned before 1990. From Canāt Get Much Higher. Also: āAccording to a report by CNN, about 52% of adults said they celebrated Halloween in 2005. In 2012, that percentage had jumped to around 72%. Over a decade later, ~the New York Times related~ that that percentage has slowly crept up closer to 75%.ā
Transplant recipients can inherit memories and preferences from their donors from Adaobi Adibe. More on this here and here.
Getting materials out of the lab by Ben Reinhardt in Works in Progress. Lots of this resonated for me in my work at Macro Oceans, even though our materials arenāt novel in the same way that say carbon fiber was in the 1960s. Thereās an interplay between unique functionality (what the material does), scale (your ability to produce consistently), and price at each step along the journey. The art is picking use cases where your unique functionality isnāt blocked by your limited scale and high price.
I know from my own experience of studying martial arts in Japan that intense study brings rewards that are impossible to achieve by casual application. For a year I studied an hour a day three days a week and made minimal progress. For a further year I switched to an intensive course of five hours a day five days a week. The gains were dramatic and permanent, resulting in a black belt and an instructor certificate. Deep down I was pessimistic that I could actually learn a martial art. I thought you were either a ānaturalā or nothing. Then I saw natural athletes fall behind when they didnāt practice enough. This, shamefully, was a great morale booster.
The global wars between ant super colonies.
"Social media basically brought us to something like an oral culture" and more from Katherine Dee.
The Marginal Revolution Podcast on Crime in the 1970s. They were somewhat pessimistic as the episode ended, but it made me much more optimistic about Americaās future. The resilience of our society is really underrated.
AR binoculars that automatically identify birds anywhere in the world. Stupid great product idea.
Product market fit provides gravity for a business. Before you have it, moving in almost any direction might be a good idea. But after you have it, youāre either going to double down on whatās possible or expand into the adjacent possible. All the moves are directly related to your current momentum.
āFor an increasing proportion of software itās more helpful to think of it as content rather than softwareā ā Daniel Kuntz
āThe difference is that I just get to be really stubborn about making things as good as we all know they can be.... But the real big thing is: if youāre going to make something, it doesnāt take any more energy ā and rarely does it take more money ā to make it really great. All it takes is a little more time. Not that much more. And a willingness to do so, a willingness to perservere until itās really great.ā - Steve Jobs
Sometimes caution is the riskier choice.
Politics is made up of both style and substance. I remember Tom Holland of the Rest Is History saying that Roman political parties didnāt break down on policy lines the way that ours do but on style: a conservative style vs a progressive style.
Prizes are an under used way of incentivizing behavior.
ABP.
## Sports gambling needs to be restrictedAt this time in 2022, I was pro-online sports gambling. I wanted the ability to make small money bets. I enjoyed traveling to New York or Pennsylvania and being able to bet on teams and players I like. I followed Georgiaās progress (or lack thereof)in legalizing online sports gambling.
Online sports gambling would be all upside for me. I know Iād never bet more than I could afford to lose. Betting $1 is more fun for me than betting $100,000 because I can afford to lose the $1; the thrill is from making a prediction and seeing the result.
I changed my mind on this somewhere in the past couple of years. I now believe that unrestricted online sports gambling is a bad idea and we shouldnāt allow it. Many people arenāt like me and the costs to them and their families outweighs whatever fun I can have.
The Zvi has done a great job at cataloging the negative impacts: large increases in bankruptcy and domestic violence; large decreases in household savings all related to the expansion of online sports gambling. Expanding access to this addictive activity causes more people to get addicted and the consequences for them are catastrophic. At this level of cost, it isnāt worth it for me to be able to put $3 on SGA for NBA MVP. [0]
At minimum, we should be restricting access and possibly even preventing the maximum amount that an individual is able to lose ā and making gaming companies liable if they go above that amount.
[0] Plus I can substitute with Manifold Markets, which is a prediction market where you can bet with free points is a great substitute for me. I just like being able to make the bets!
## Book Thoughts: Calvin and ZwingliIāve spent most of this fall with the Swiss reformers John Calvin and Huldrych Zwingli.
Sometimes when I read a biography of a thinker, I get a new appreciation for their ideas. Situating them in the actual life and time they were born in gives them a vitality you canāt get from Wikipedia.
That did not happen with these two. Iām not sure how much of that is the author, who doesnāt seem particularly interested in the theological minutiae of the early reformation, and how much is the sources, who mostly kick in when the men have already developed their ideas rather than as they were forming them, but I didnāt come away with an appreciation for why they felt so strongly about the theological issues that seem so remote to us now. Imagine having a fight over how much art is in a church!
What I did take away from the books is an understanding for why Switzerland and the US feel so culturally similar.
Here is Gordon describing Zwingli:
His calls for religious freedom were coupled with demands for liberty from tyranny, both religious and political.
This then gets exported to England through Jean Calvin and Heinrich Bullinger, Zwingliās successor in Zurich. The English and Scottish reformed communities become very influential in the culture of the thirteen colonies. In a very real way, the cultural DNA of the US comes from the Swiss Alps.
One other thing that stood out: there was clearly a warmth and a charisma to the two men at the time that doesnāt translate. Both had large followings and deep relationships that inspired people to follow them through hardship. Calvin, as an example, taught lectures that were well attended and created many acolytes. They arenāt the austere caricatures that are passed down to us.
## Token stream 2024-11-01Public Service Annoucement
Macro Oceans joins forces with Everything Seaweed . š Letās make waves together š.
Lebron and Bronny James are the highest scoring father and son duo in NBA history without Bronny ever scoring a point. Theyāve also outscored Chris Bosh and Dwayne Wade combined.
GOAT: The Gospel of Goatse. The world is going to get really weird. Iām here for it.
Risks are tackledĀ up front, rather than at the end.Ā In modern teams, we tackle these risksĀ priorĀ to deciding to build anything.Ā These risks includeĀ valueĀ risk (whether people will buy it),Ā usabilityĀ risk (whether people can figure out how to use it),Ā feasibilityĀ risk (whether our engineers can build what we need with the time, skills and technology we have), andĀ businessĀ risk (whether this solution also works for the various aspects of our business).
Where does energy, in the sense of human vigor, come from? Sadly, no concrete answers were given.
A great article on creative discontent from Celine Nguyen. One of my favorite David Halberstam quotes is āBeing a professional means doing your job on the days you donāt feel like it.ā He wrote one of my favorite books, Breaks of the Game, which is so good because itās insanely well sourced. I imagine him getting up, heās got a cold and itās rainy. He would rather sleep in, but he pours a cup of coffee, gets in his car, and goes to have one more conversation with one more soruce. This one reminded me of him. Plus it lead me to the Ogilvy company principles:
Dogged determination is often the only trait that separates a moderately creative person from a highly creative one. Never give in, never give in, never, never, never, never. Persistence and determination alone are omnipotent. Before them, obstacles vanish into thin air and mountains crumble into atoms.Ā
And:
Conclusion: We are what we repeatedly do. Being very good is no good. You have to be very, very, very, very, very good.
How to build an LLM judge by Hamel Husain. Notice the role of the taste maker?
When I was a PM on the YouTube Creator team, a super common interaction I had with top creators when something like this:
- Me: How did your channel get started
- Creator: I just started one day and it blew up almost from the first video
- Me: Really?
- Creator: Yeah.
- Me: Wow, thatās crazy.
I must've had this exact conversation at least 5 times. There's a lesson in there somewhere.
## This week's token streamThere are more deaths from alcohol each year in the US than all illicit drugs combined ā Charles Fain Lehman via the Ezra Klein Show
How Density Zones can solve Americaās housing Crisis ā Agglomerations
A Character.ai characterās (potential) role in a teenagerās suicideā NYT.
These newly-discovered Nazca lines depict very figurative rather than literal shapes, but they also depict ritual, and maybe even lawgiving or war.
— CrĆ©mieux (@cremieuxrecueil) October 11, 2024
These lines are weird precisely because they had to be distinct to suit their purpose.
I'll come back to this. pic.twitter.com/RY6pdTUZ3q
One thing I notice is that super early stage companies have to do impossible things to make it to the next dayā¦ which sometimes leads to them being bad at assessing which super impossible things they can do and which ones they canāt. I think political progressives can be the same way. They have to by nature believe that wholesale change is possible but this blinds them to the things that canāt be changed.
With apologies to the boffins around the world who know more about this than I do, I think AGI arrived with GPT-4. Everything else is just a continuation on that theme.
The thing thatās great about Duolingo is that theyāve figure out how to turn aggressive gamification strategies towards a neutral to positive end
āA good science fiction story should be able to predict not the automobile but the traffic jam.ā ā Frederik Pohl
Great advice from This is Not Advice:
When starting out, itās easy to spend time on your strengths and ignore your weaknesses until much later. In fact the startup world often fetishizes a founder's particular strength. That is to say, if you know a particular founder is design-oriented, you expect to see some of the most beautifully designed things from them. Despite this, if you know you are capable of doing something, it can be beneficial to focus on the other things. Simply put: start with the stuff you donāt know you can do. [Ed.: I remember a startup in which we, foolishly, passed on investing that had made such fast progress. When I asked them how they'd advanced so quickly they explained that they'd exclusively focused on things they were unsure would work.]
good take from @nabeelqu pic.twitter.com/iRRmeGCSjN
— Jordan Schneider (@jordanschnyc) October 16, 2024
Iād add Alex Komoroske and Simon Wilson to the meme creator list.
## This week's token streamCreativity is not magic! Good ideas don't just come from thinking really hard! Good ideas come from arbitrage: knowing about areas and ideas and facts that your classmates don't know about. By definition, this is not going to be on your class syllabus
— alz (@alz_zyd_) October 22, 2024
The last person to be guillotined in France was put to death in 1977. The Rest Is History, Wikipedia.
The Barnum effect isĀ individuals give high accuracy ratings to descriptions of their personality that supposedly are tailored specifically to them, yet which are in fact vague and general enough to apply to a wide range of people. Hat tip to Simon Wilson who brought this to me.
āMeanwhile, those who study, pray, and commit acts of loving-kindness keep the world going.ā Cluny Journal
How young men and women (mis)understand each other. American Storylines.
Internet of Bugs, a software developer + YouTube Creator reviews o1 and sees it as a step change for AI.
Engelsberg Ideas on the missing quality of judgement. Iāve noticed that product management orgs in particular are often embarrassed by the degree to which the job depends on judgement; instead, I think they should embrace it.
How drones are changing the war in Ukraine.
When Iām in execution mode, I find myself searching for slack in my schedule to eek out just a little bit more productivity. When I look at my to do list this way, I start to see the little things I can do here and there regardless of priority. But eventually what happens is that I begin reordering my to do list away from whatās important and towards what I can do quickly. The urgent overwhelms the important.
I suspect that the Creator Economy is coming for software development. In the same way that the iphone made it so that anyone can make a video, LLMs are making it so anyone can make software. This makes it even more important to be focused on who the software is for and how it integrates into the life of the user. Nicheness is even more important.
Meaning comes from cost. If itās free to do, you wonāt feel ownership of it.
## AI and the Pyramid of SuccessOne of the things that Iāve noticed about AI tools is that theyāve changed the way that I think about the role of intelligence in success.
In 2018, if you had sat me down and pressed me on the top qualities required to be successful, I probably wouldāve had intelligence first. The ability to figure out the answer, it seemed to me, was probably the most important single quality to have if you could only have one.
In a post GPT-4 world, though, Iām no longer sure this is true. When I think about my kids and the qualities they need to develop, yes, I want a minimum level of intelligence, but Iām more interested in curiosity, initiative, earnestness, industriousness, judgement / taste, courage, and playfulness.
Iām struck by how many of these qualities are sitting in John Woodenās pyramid of success.
Perhaps I was overrating the role of intelligence all along. Now with this frame, when I think about the most successful people that Iāve observed up close, alongside intelligence and the ability to process information is a whole lot of earnestness and industriousness.
Looking ahead, it seems to me that what is increasingly scarce and valuable isnāt the ability to breakdown what needs to be done, but the ability to get up off the couch and go do it. To consider the result that you get, and then go try again. To create trust with others so they help or at least don't hinder your progress. When I think about preparing for the world of the future, I think about thing a design studio, a Montessori classroom, or a basketball team.
## This week's token streamThings I learned: āthe Milky Way builds between two and six sun-size stars a year.ā Quanta Magazine.
My friend Alex Komoroske on Lennyās Podcast. Alex is in the top 5 most influential people on my career in the past 5 years.
Henrik Karlson on how Jesuits and Montessori schools teach and scale culture.
The Swiss border is changing due to climate change.
āBut the truth is that kids are more like artificial neural networks ā theyāre at a subtly different point in mind-space, theyāre good and bad at different things than adults are good and bad at.ā The Psmiths.
The story of how Dr. Zhivago got published. One of the most haunting books Iāve ever read. A reminder that civilization can collapse before your eyes.
How Jason Crawford chooses what to work on.
Speed Matters. āBeing 10x faster also changes the kinds of projects that are worth doing.ā
## This weekās token streamHow to make millions as a professional whistleblower. What a weird and interesting career path.
Itās time to talk about Americaās disorder problem. One of the things that stood out most to me when moving back to the US from Switzerland was the amount of disorder that we tolerated as a society. This tolerance for disorder might not be entirely bad ā America is nothing without its weirdos ā but Iām not sure we realize the degree to which it is a choice.
Inventing on principle. Fantastic and through provoking talk about what motivates innovation. It has me wondering what principles I can commit to in this way.
How to succeed at Mr. Beast Productions. Includes a great 101 description of how YouTubeās algorithim works + a lot of tenacity.
How I failed. The CEO of OāReilly Media talks candidly about the biggest lessons heās learned along the way. Rare to get this much candor in one of these.
Meta smart glasses lead to real time doxxing. I donāt see anyway we can expect to unrecognized in the future. Better to accept it.
## We're past peak obesityRelative to trend US obesity is down about 7 percentage points (~20%) among graduates (the group most likely to be using GLP-1 agonists like Ozempic).
— Rob Wiblin (@robertwiblin) October 4, 2024
Incredible stuff.
We may have passed peak obesity: https://t.co/GRWsl6E52V https://t.co/9zTEelFxVr pic.twitter.com/a3zjjIUSnX
Increasing obesity has been a fact in the United States for my entire life and it isn't any more. Pretty incredible.
## This week's token streamHow the psychiatric narrative hinders those who hear voices | Aeon Essays - an exploration of the āTargeted Individualā community, people who hear voices in their heads. Weird, wild, and a bit scary.
How to beat AI at Go - humans are able to beat the best AIs at Go by finding failure cases they arenāt prepared for. This is the future of warfare.
Palmer Lucky profile: such a great reminder that anything is possible with hard work and determination. Similarly, Casey Handmer on how entrepreneurship has changed the way he thinks.
## ScaffoldingIām really coming to appreciate the value of scaffolding in product development.
What do I mean by scaffolding? The structure that allows you to build the product effectively.
Some examples:
You set up scaffolding to help you build. It doesnāt need to be pretty, but it needs to be fast, cheap, and effective. At the end of the project, you take it down. Or maybe you incorporate it into the structure of the product, improving it to make it fit for purpose.
Sometimes the scaffolding feels like a distraction. Iām going to build a whole separate structure just to help me build? Only if you want to build it well.
The best projects Iāve worked on outline the scaffolding early. These are the support structures weāll need to do good work fast.
## The AI that makes the AIOne of the grand challenges of artificial general intelligence is developing agents capable of conducting scientific research and discovering new knowledge. While frontier models have already been used as aids to human scientists, e.g. for brainstorming ideas, writing code, or prediction tasks, they still conduct only a small part of the scientific process. This paper presents the first comprehensive framework for fully automatic scientific discovery, enabling frontier large language models to perform research independently and communicate their findings. We introduce The AI Scientist, which generates novel research ideas, writes code, executes experiments, visualizes results, describes its findings by writing a full scientific paper, and then runs a simulated review process for evaluation. In principle, this process can be repeated to iteratively develop ideas in an open-ended fashion, acting like the human scientific community. We demonstrate its versatility by applying it to three distinct subfields of machine learning: diffusion modeling, transformer-based language modeling, and learning dynamics. Each idea is implemented and developed into a full paper at a cost of less than $15 per paper. To evaluate the generated papers, we design and validate an automated reviewer, which we show achieves near-human performance in evaluating paper scores. The AI Scientist can produce papers that exceed the acceptance threshold at a top machine learning conference as judged by our automated reviewer. This approach signifies the beginning of a new era in scientific discovery in machine learning: bringing the transformative benefits of AI agents to the entire research process of AI itself, and taking us closer to a world where endless affordable creativity and innovation can be unleashed on the world's most challenging problems.
From The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery, full paper available here
I'm interested to read this one more closely and see the degree to which it does (or doesn't) rely upon having experiments that the LLM can execute without human intervention. Either way, an interesting result, but my hypothesis is that "places where the LLM can verify a result" is going to be the limiting factor.
## Book Thoughts: Between Two FiresLink to the book is here.
The first 10 chapters of this book are among my favorite reading experiences ever; terrifying, mysterious, creative. I wondered how the author was going to keep up that pace for the rest of the book.
Ultimately, he didn't. I'm not sure if it was possible for him to. Part of what made the beginning part of the book so enticing was wondering if miracles were actually happening or if they were just coincidences. At some point, the author had to make a choice and from that point forward everything got less interesting.
## California's changing climateA fantastic piece by Paul Kedrosky on how California's Atmospheric River is changing and the implications of these changes.
Lots of great stuff in it, but this was totally new to me:
## Time spent with childrenA predator-prey model is a mathematical representation of the interactions between two species: a predator and prey. It is often modeled as wolves, sheep, and grass. The most common model is the Lotka-Volterra, which consists of two differential equations. There are two stable equilibria: one with predator and prey in approximate balance, and one with both extinct
and want to know the insane fact about this stat?
— Lyman Stone ē³ä¾ę° š¦¬š¦¬š¦¬ (@lymanstoneky) July 2, 2024
it's probably not true since the stat being quoted is "childcare time" not "time with kid" https://t.co/l3U4ibUkT3
Posting this since this stat was one of the 52 things I learned in 2023
## How will agents interact with the world?Lattice made a splash this weekwith a pretty crazy announcement about adding AI workers to their platform. It was shambolic and theyāve since walked it back.
While Lattice did this poorly, I think that the question of āhow do we integrate Agents into the world?ā is an interesting place to dig and experiment right now.
As an example, if you believe in agents, it seems pretty clear that agents are going to need to be able to pay for things subject to certain rules. So... what does it look like to give an agent a credit card?
I could just give it a credit card in my name, but that seems a little risky, and if things go wrong, whoās going to make that right?
But if I hire an agent created by another company to do work for my company, who gives the credit card to them? Is it the creating company? Do they then invoice me after the fact?
It's possible that this looks exactly like how businesses give workers credit cards... but maybe not? It might be better to know that this is the card assigned to system X by entity Y. The entity that is ultimately on the hook for the spending even if things go wrong might want to be able to track that; the credit card issuer might also want to know which of its clients are giving Agents these abilities as the patterns of spending, real and fraudulent, might look different. This transparency probably helps the system overall.
Another example is account creation. There are probably types of services where we want non-human actors to be able to create an account. We could have them pretend to be human, but it might help to let them ask for agent access to a service. This is probably different from API access; in some cases, it probably helps for them to see exactly what I see in the system.
Zooming out a bit, it seems to me that people get really upset when something pretends to be a human but it is actually AI. It also seems likely that weāre going to want to give agents more ability to act in the world and be productive. Yet the systems we have today that are essential for productive work assume human actors or computers acting on behalf of humans (programmatic access), but nothing in between. If weāre going to capture the value from agents, our systems are going to have to adapt.
## How children refer to adultsWhen I was growing up, I never used first names with adults. The adults in my life were "Mr. Knabe", "Mrs. Stanley", or "Dr. Woods".
Adults reinforced this norm as well. When I met my parents friends, they introduced themselves ā in a friendly way ā as "Mr. Brinker" rather than Chris. The same with teachers ā I had "Mrs. Bryson", not "Deborah".
My parents wouldāve corrected me had I tried something else. Iām sure they probably did at some point, but I donāt remember it happening. It wasnāt notable, itās how the world was. In lots of cases, I'm not even sure I knew the first names of my parents friends until I graduated from college and then someone like Mr. Hehn would say, "please, call me Gunther" in a way that communicated I was now an adult too. This made me feel proud. The only exceptions I can think of here are my Pastors (Jerry) and family (Aunt Julie and Uncle Bert).
As far as I can tell, this has completely gone out of fashion.
With my kids, 2 and 4, no adult uses their last name. My friends introduce themselves as Mr. Jon and Ms. Veronica, not Mr. and Mrs. Flash. I do this too ā I introduce my friends to them as Mr. Graham and Mr. Ted not Mr. Rowe and Mr. Strong. Even my daughterās teacher is Ms. Heather not Ms. Jones. I assume this will change as they enter the formal school systemā¦ but who knows!
This new behavior is so consistent that if an adult that I knew well introduced themselves to my child as Mr. Banna instead of Mr. Rami, it would seem overly formal, like wearing a tuxedo to an office.
This doesnāt bother me on a moral level but I am intensely curious about it. When did it change? Why? I assume itās related to the broader decline of formality in our culture, the way that the hoodie has replaced the sports coat for menswear.
But what is driving this? Is it a desire to be youthful? Relatable? A way of communicating that adults and children are on the same level? As weāve made this switch, what have we given up? Anything? Nothing? Does this change how children perceive adults? Does it change how children perceive themselves?
Iād love to hear a theory of the case here.
## Reactions to Situational AwarenessMy first glance reactions to Situational Awareness: The Decade Ahead by Leopold Aschenbrenner.
I enjoyed reading it a lot.
The most persuasive part of his argument to me is the relationship between compute and intelligence. This is sort of like the New England Patriots to me; I'm going to believe in it until it stops working. I see reasons why it might stop (run out of data, limited by energy / computing power available), but I don't know when or if we'll actually hit those constraints. People are pretty good at avoiding constraints!
I think he underrates the likelihood of a bottleneck somewhere that keeps us from getting to the AGI he imagines. Any individual bottleneck might be unlikely, but as long as one exists, the entire system is constrained.
Something I see Leopold do at points is assume a super AI, in his case, an automated AI researcher that is 100x as competent as today's top AI researcher. With this assumed, any AI research problem is solvable because you can scale up infinite 100x AI researchers to get around the problem. Once any AI research problem is solvable, then any problem is solvable.
What I think will ultimately happen is something like this:
Somewhat off topic: earlier this year, I read Meet You in Hell, which is the story of Henry Clay Frick and Andrew Carnegie. The dynamics of that era, with the railroad leading to a spike in demand for steel and steel leading to a spike in demand for coke were very recognizable in today's AI race.
[0]: I think GPT-4 is already this! Do you know a single person who knows as much stuff about as many things as it does? I don't. And yet it still has limitations!
## Goodbye, Bill Walton
Bill Walton is my favorite athlete I never saw play. He passed away yesterday.
In the summer of 2007, I read David Halberstam's classic book, Breaks of the Game, which is about the Portland Trailblazers of the Late 1970s who briefly shot to the top of the NBA world before falling back down to mediocrity.
Breaks of the Game is as close to a perfect book as I've ever read. The focus for the story is Bill Walton's foot, which breaks in the 1977-78 playoffs with his Portland Trailblazers set to win a second straight NBA title. This bad break cuts short a championship run, a budding dynasty, and according to Halberstam, the perfect basketball team with Walton the perfect basketball star.
With this as his lens, Halberstam tells the story of how basketball and American culture are changing, becoming more financially driven, and in the process something is being lost. If you have even a passing interest in basketball, you should read it.
I met Bill Walton through this book, but I fell in love with him when he came on to the Bill Simmons podcast. [1]
Simmons wanted to have a fairly conventional conversation about the 2007 NBA season. [0] But Walton wanted to use basketball as a metaphor for life, specifically how to create a meaningful life, especially for young people, through preparation, judgement, and irrepressible joy. He had a sense of the interplay between individual talent and team chemistry that I love most about the basketball and articulated in a way no one else can.
Over the years, I've probably re-listened to that conversation 15 times to try and capture a little bit of his approach to life. [2]
As I've gotten older, the list of athletes I want to meet has dwindled. I still appreciate the skill, but cringe at the awkwardness of approaching another grown man... to talk about what exactly? I'd rather leave them to practice their craft and enjoy their time with their families.
Bill Walton remained the exception. When I moved to San Diego in 2021, I would go on runs past his house in hopes that he might be out checking the mail and I might get a chance to thank him. For what exactly? Having a great basketball career? A wonderful approach to life? I'm not exactly sure, but I know he mattered to me. Of course, I flattered myself that he'd recognize in me a kindred spirit and a friendship would be born. It wasn't to be. With his passing, it never will be.
It's an odd thing to care this much about someone who you never saw play. If anything, it's a reminder that how we live matters and can echo out into the world to the people around us, even people who've never met us and never saw us do the thing that we were best at.
Thank you, Bill Walton, for sharing your gifts with me.
[0]: Such a great time to be an NBA fan.
[1]: At the end of the podcast, Simmons brings up Breaks of the Game and it's so painful for Walton that he has to end the show.
[2]: I also printed out a copy of John Wooden's Pyramid of Success and bought a handful of old Sports Illustrated with Walton on the cover.
## ChatGPT Mac AppI'm one of the early users here, but this app is severely underbaked, even for a beta rollout.
Forget about advanced features ā I'm having issues with basic scrolling. It's also slow. The speed of response time that's so exciting on the web isn't there for some reason.
I also miss the ability to cmd + f for pieces of text within a chat. For some reason, on the app this does a search across my chats but not within the chat I'm focused on. Not helpful!
I don't want to draw too many conclusions from a bad week for OpenAI, but it definitely seems like they've lost focus on what matters.
## What makes a political office non-partisan?The state of Georgia has certain offices that are designated as non-partisan. How does this get decided? Where is the line drawn and why? Are there any restrictions placed on the candidates when running for a non partisan office, or does it just mean that the party isnāt listed on the ballot?
If you feel like you understand how this works, let me know: jdilla.xyz @ gmail dot com.
## Friday threadsFor a couple of years now, Iāve been posting things I learned as a way of cultivating curiosity.[0]
But this year Iāve stumbled upon a different sort of thing I want to train myself to notice: things I wish I knew.
I find myself somewhat embarrassed to post these. But why? Probably because I feel like if I were truly motivated, I would be able to figure them out.
I think this is the wrong instinct. Someone out there almost definitely knows the answer to them and thereās a chance they just swing along and tell me. In that case, Iām better off. And some of the most impactful projects Iāve seen first hand have begun with someone wondering, āwhy is this the way that it is?ā
Since sifting my thoughts for these, Iāve found them to be way harder to capture. Iāve had 2-3 hit me and then disappear, only for me to be unable to locate them again. This almost never happens to me with āthings I learned.ā I wonder why that is?
Now, for my first one:
I wish I understood how individual trust is converted into group/institutional trust and how group/institutional trust converts into societal trust. I feel like I have a good idea on how an individual creates or destroys trust, but donāt think I understand how it converts for a team the size of a small company (say ~25-40 people), let alone a large company (thousands of people) or a society. Say youāre the mayor of a small city and you think a high trust society is important. Is it possible to do anything to foster this? How does it work?
My hypothesis: I assume itās some combination of credibility, reliability, and lack of self interest. So when people see society work (e.g., civic institutions function well, utilizing judgment, being able to be counted on) and that individuals arenāt profiting at the expense of the group, civic trust goes up. Butā¦ I could be wrong. If you feel like you definitively understand this, reach out: jdilla.xyz at gmail dot com.
[0]: Iāll probably keep doing that, because why not? Itās super fun.
## Friday threadsFrom my friends at Duolingo:
## Quantity precedes qualityThe people that affectionately call their Prince William "Wills" and Ā£5 and Ā£10 notesĀ ~"fivers" and "tenners"~Ā are responsible for shortening "Association Football" to just "Assoc."āwhich, when written, looks like it might be pronounced "Assock."Ā (This "Association Football" name is the same as the FrenchĀ Football AssociationĀ in FIFA!) In late 1800s England, at Oxford, there was also a fad of addingĀ -erĀ to some words. And thus, "soccer" was born. In England. š“ó §ó ¢ó „ó ®ó §ó æ
Found via Dynomight:
Quality over quantity.Ā I often worry that I write too much on this blog. After all, the world has aĀ lotĀ of text. Does it need more? Shouldnāt I pick some small number of essays and really perfect them?āØArguably, no. Youāve perhaps heard of the pottery class where students graded on quantity produced more quality than thoseĀ gradedĀ on quality. (It was actually aĀ photography class.) For scientists, the best predictor of having a highly cited paper is just writing lots of papers. As I write these words, I have no idea if any of this is good and I try not to think about it.
I hadnāt heard this before, but I do find it to be true. Creativity is a habit. The way to quality is through quantity.
## Claude ReviewYou can read my review of Phind, another LLM tool, here.
For the past several weeks, Iāve been using Claude Opus (the paid version of Anthropicās ChatGPT competitor).
For the first time, I think ChatGPT has been outclassed.
For context, I use Claude for help coding (you can read about my process for coding with GPTs here), for understanding new-to-me technical concepts, and for marketing, business, and technical writing.
Hereās what I like about Claude:
With that said, Claude has some pretty big limitations:
Despite all these limitations, given the choice between which one to work with, I am consistently choosing Claude. I wonder what this says about the stickiness of these tools?
Now for the ultimate test: have I cancelled my ChatGPT subscription yet?
No, but for an unexpected reason: the ChatGPT mobile app. Iāve been having voice conversations with ChatGPT mobile, both personally when thereās a topic I want to dig into, and increasingly with my daughter. We pick a topic sheās interested in (e.g., where do dinosaurs come from) and just riff on it. Itās perfect for a curious toddler ā she can keep asking questions over and over again. I think this is what Tyler Cowen means when he says creators are competing with LLMs for attention.
Weāll see how long this lasts ā I donāt expect that I will continue keeping two LLM subscriptions forever. Hurry up and add a mobile app, Claude team!
## Friday ThreadsHeyRecap is the next evolution of Recap Roswell, a project using LLMs to create easily readable summaries for the Roswell City Council.
I started this project with two main goals: 1. When I moved to Roswell, I wanted to get more involved with my local government, but figured I should learn about it before I got involved. To my surprise, despite the nearly $200M city budget, there was very little local news coverage to help me figure out what was happening locally. 2. I wanted to develop a better understanding for how LLMs can be used to solve everyday problems. My hypothesis here was that I could create a system that did a good enough job summarizing the meetings that it isn't worth sending a person to cover.
The first version of the project was a python script that created a summary which I copied and pasted into a Ghost newsletter. This was a great way to get started ā to my surprise, ~150 others in Roswell were also interested in this ā but it really limited my ability to customize the user experience and it was clear that it wasn't going to be able to scale beyond just my local city council. Plus the copying and pasting was tedious!
So this fall, I set out to build an end-to-end app, the result of which you can see at heyrecap.com. For users, there are two primary benefits to the new site:
For me, this was a chance to build an app end-to-end with a real, if small, audience, while getting to know my city better. I used NextJS and hosted it on Render. Summaries are produced with transcripts from Deepgram. Clerk is providing me with user authentication services (love them) and Resend with email services. My UI components are provided by shad/cn UI. I have opinions about all of these and may write more about them in the future, but for now I'll just say that it's incredible to have so many services so easily stitched together at my fingertips. More often than not, someone else has done 90% of the hard, frustrating stuff so that you can just integrate it with your project. It's awesome.
Over the next several months, I hope to explore what it looks like to scale this some. I'm not sure there is a business here per se, but I'm interested in the idea that I can run a useful local news organization as a side project using AI. We'll see what comes form it!
## JalapeƱos are less spicyFrom D Magazine:
The standardization of the jalapeƱo was rapidly accelerated by the debut, about 20 years ago, of the TAM II jalapeƱo line, a reliably big, shiny, fleshy pepper that can grow up to six inches longāwith little to no heat. TAM II peppers have become some of the most popular in the processing business.
The driver of the change was the desire for processors to have predictable levels of spiciness.
Hat tip to Mark for sharing this.
## You should go visit Pittsburgh in FebruaryTo experience the region's Fish Fries. On Fridays during lent, various Catholic Churches (usually) host fish fries. Growing up, I sort of took these for granted, but now when I go back, I see them through different eyes. It reminds me of something like Basque Cider Season, a sort of cultural coming together that is specific to a place.
## Friday threadsLetās talk about statistics. The important statistics in basketball are supposed to be points scored, rebounds and assists. But nobody keeps statistics on other important things ā the good fake you make that helps your teammate score; the bad pass you force the other team to make; the good long pass you make that sets up another pass that sets up another pass that leads to a score; the way you recognize when one of your teammates has a hot hand that night and you give up your own shot so he can take it. All of those things. Those were some of the things we excelled in that you wonāt find in the statistics.
## Resend review / friction logThinking about how cool it is that the electrons inside an atom of gold are moving at 58% of the speed of light. pic.twitter.com/tBl6v1lI6x
— Will Kinney (@WKCosmo) February 6, 2024
I've adopted [Resend] (https://resend.com/) for a project. If the Resend team is out there and stumbles across this, I offer this feedback as a gift, nit-picky as it might be.
On the whole I enjoyed it, but the lack of out of the box analytics means I'll probably search around more should I need an email service in the future.
Here are my notes:
I was listening to Fall of Civilizations podcast and they referenced the size of Lake Victoria, the source of the Nile river, in Africa as approximately the US State of Georgia.
It turns out this is incorrect. Lake Victoria is 23,146 square miles while the US State of Georgia is 57,513 square miles. However, it is about the same size as the Country of Georgia, which is 26,830 square miles or the state of West Virginia (24,038 square miles). Still incredibly massive, and it's not even the biggest fresh water lake! That honor goes to Lake Superior, which is 49,300 square miles, about the size of Alabama or Greece and holds 10% of the world's fresh water (!).
In either case, Lake Victoria has shot up my list of places to go. I'm sure I knew that it was the source of the Nile river at some point, but there's something romantic about being reminded about all that distance and all that history.
## Repackaging knowledgeIāve been noticing a particular strong point in my use of LLMs: repackaging knowledge.
Iāll use a specific example here from my work at Macro Oceans: I need to understand the research around specific polysaccharides that can be extracted from seaweeds 1) so I can sell them and 2) so I can do further product development with them. To do this, I do a pretty basic lit review. I get a lot of help from GPT in this stage understanding things that are beyond the chemistry and/or biology that I took. But I actually need to distill this into something I understand and have an intuition for, so it's pretty essential for me to take the times to put this into my own words.
At the end of this process, Iāve effectively got a 1 page memo that explains how the polysaccharides work and what their benefits are. From there, I need to turn this into: * 2 slides for the team meeting to help teach internally * 4-5 slides for a sales training deck * a blog post for content marketing purposes * a paragraph to a specific customer for a deal where it's relevant
This is where I think the LLM really shines. The context is set and hallucinations are rare. Perhaps more important, I understand the content well enough to catch things that arenāt quite right. The important part is mutating the form. In tasks like this, Iām frequently able to cut half the time more out of the work compared to what it would take me starting from a blank page.
An implication of this is that the value of proofreading and comprehension as a skill is shooting up tremendously. Iāve always been impressed with the ability of some of the senior executives Iāve worked with to read a memo or a slide deck and immediately pick out the critical issues at hand (Neal Mohan, CEO of YouTube, is fantastic at this). The difference between an average manager and the top ones on this dimension is startling. The average managers fall into nodding along but the really top tier people are engaging with and testing the material theyāre being given.
I think this comprehension skill is going to play the role in the next two decades that being a strong writer played in the previous two. Of course these two skills are closely related, but as more people move into a role where theyāre reacting to more text than theyāre creating, it will become even more important.
## Assorted LinksāThe thing that Iāve learned, that every writer needs to learn, the thing that I know absolutely, is that feeling you have that you can write something, when you know, āOkay, Iāve got it now,ā you have to write exactly then and get it on the page, because that feeling will disappear like a fart in the wind. Itāll be gone. Youāll come back to the desk, and youāll be like, āWhat was it?ā You can write notes. Thatās not going to help. You just have to sit down and write itā¦. Itās a really important thing for everybody to know, because the feeling is so convincing that youāll always be able to write it. Itās like being drunk, then sobering up, or vice versa. Youāre a different person the next day, and you donāt have it anymore, and then youāve got to think your way back into it.ā
āThere was a whole colony of experimental chimps at Holloman. Monkeys went up in balloons and in V-2 rockets. Many of them died. Chimpanzees were strapped into a rocket sled and abruptly decelerated; they were spun, tumbled, ejected from their seats, subjected to wind blasts, and slingshotted in the ābopper.ā They died, they were autopsied, or they lived but suffered injuries and were āsacrificedā and autopsied.ā
A compelling case that plants and animals without brains have memories.
āThe neuron is not a miracle cell,ā says Stefano Mancuso, a University of Florence botanist who has written several books on plant intelligence. āIt's a normal cell that is able to produce an electric signal. In plants almost every cell is able to do that.ā
On one plant, the touch-me-not, feathery leaves normally fold and wilt when touched (a defense mechanism against being eaten), but when a team of scientists at the University of Western Australia and the University of Firenze in Italy conditioned the plant by jostling it throughout the day without harming it, it quickly learned to ignore the stimulus. Most remarkably, when the scientists left the plant alone for a month and then retested it, it remembered the experience.
Found via The Browser.
## Ripple effects from ad trackingSpeculative from someone who knows the industry well: * The changes Apple made to Ad Tracking have totally ruined the playbook for DTC beauty brands, who can no longer grow using targeted ads on Facebook (this is well known) * A second order effect of this is that itās making it harder for new ingredients to get adopted, because the DTC brands were usually the most willing to experiment here. * This in turn slows the pace of innovation in the ingredients industry, especially towards cleaner / more sustainable / more natural alternatives.
I can't prove or disprove this, but it does make intuitive sense.
## More declining trust in authorityPost by @ianbremmerView on Threads
Presumably this trust is going somewhere. With congress people, I know it's common for people to say "representatives are crooks, but mine, Jane Smith, is alright." I would be the same phenomenon is happening here.
## American squalorRegulations themselves arenāt the problem, though. Germany, like much of northern Europe, is a high-regulation society, but itās also high-trust, compared to the US. Here, nice and fully functional things are built without fear of misuse. For Americans, who have both a high-regulation and low-trust society, this is all rather depressing; itās the combination that means we canāt have nice things.
I like to live here, but the reality is we are rapidly falling behind the rest of the world in liveability, especially when you adjust for our wealth. Our cities are being frozen in time by an absurd, centralised regulatory mindset, which sees human flourishing as dirty and unsafe, and seems determined to wring out the last drops of any soul from our urban spaces. A mindset that manifests as one useless La Sombrita at a time.
By Chris Arnade.
## Book Thoughts: Eisenhower in War and PeaceMy first book of 2024, Eisenhower in War and Peace by Jean Edward Smith.
One of my favorite biographies. Incredibly well paced and readable.
I think Ike has to be in the top 5 most influential Americans. Some of my contenders, in no particular order:
Ike obviously belongs in this list. As much as it pains me to say it, I think he's obviously above Lincoln, who is my favorite of the group, but just isn't on the national stage for long enough.
Here's the case for Ike:
Assorted other stuff I enjoyed from this book:
Via Marginal Revolution:
The explanation for this is pretty obvious when you think about it. Just as today, people bought tickets and their names were written on the tickets.
I never get tired of the "you were taught this in school, but it probably didn't happen" type of thing.
## Indirect driving deaths caused by 9/11Gaissmaier and Gigerenzer found that Americans flew less and drove more in the year after 9/11, which led to 1,600 more traffic deaths over that period than would otherwise have been expected.
From Range Widely by David Epstein
## Arguments about Swiss mapsIt pains me to see the warm vineyards and villages on the sunny side of the main Valais valley on the north side of Lake Geneva and the heavily-farmed sunny slopes of the north side of the Anterior Rhine Valley in the shade, while the wooded slopes on the shady side are bathed in blazing sunlight.
From the Swiss National Museum.
A somewhat trivial topic: Swiss relief maps show the sunlight as coming from the Northwest when in real life it comes most often from the South. The article is beautifully rendered. Something about it transports me to summertime in the Alps.
The culprit? Most artists draw with their right hands and write from left to right and so European maps tend to show shade on the right-hand side of the map. Via The Browser.
## Disparity in divorceI saw this yesterday afternoon and it caught my eye:
That study that found men frequently leave their wives when they get sick was retracted/corrected.
— Leah Libresco Sargeant (@LeahLibresco) January 6, 2024
Trouble was: People who left the study were actually miscoded as getting divorced.
https://t.co/I2qHTzg6eY
It reminded me of #49 in my list of 52 things I learned in 2022: If a married woman is diagnosed with a brain tumor, there is a 21% chance that the couple will divorce; if the husband has a tumor, there is only a 3% chance they will divorce, which I found via Rob Henderson.
Based on some googling, I don't think this is the exact same study, but in the spirit of intellectual honesty, I figured I should post it.
There is some nuance, but the general relationship between illness and husbands divorcing their wives no longer holds.
Congratulations to I-Fen Lin and Susan Brown, who found the error, and Amelia Karraker who handled the correction with dignity.
## A journey through Pittsburgh's neighborhoodsNew to me this week is Dean Bog who does 15 minute videos on various neighborhoods around the city of Pittsburgh.
If you have any interest at all in the city, I suggest you check them out. The Bloomfield video (I think his first one) is a good one to start with, but they're all worthwhile. As someone who grew up in Pittsburgh, I'm amazed at how much they teach me, both about the "facts" of the city and it's culture.
One thing Dean points out is that the city neighborhoods have a distinct feel because of the geography. The hills and rivers mean that two neighborhoods that are side by side on a map may not have any actual connections between each other and so can evolve totally differently.
I found out about Dean via this City Cast Pittsburgh episode where he talks about his process. I'm paraphrasing here, but one thing he does is basically walk around and talk to people until they introduce him to the unofficial mayor of the area, who tells the story of the place.
## Phind reviewI found Phind via Marginal Revolution and Tyler Cowen's recommendation. Overall, I found it to be close to ChatGPT 4, if not slightly better ā and free!
I decided to try it out because the moment last year when it seemed like OpenAI might implode reminded me again how reliant I am on ChatGPT, especially for programming.
As I've written before, I don't really program. Instead, I scope and test. My typical workflow looks something like this:
This morning, I tried doing this with Phind.
In terms of overall quality, I found Phind to be in line with ChatGPT 4. I didn't side by side test it, but in the past I've been able to feel pretty quickly when I'm accidentally working with ChatGPT 3.5. I didn't feel this difference working with Phind; if anything, it seemed to have slightly higher quality results for coding tasks.
Here are some of the things I liked:
So why do I say they haven't quite nailed it? It's not clear to me as a user how I'm supposed to use these various fields. I can tell that they're useful and I'm fine with guessing as I go, but I wish they gave me more of a model for how to help them. It would also be helpful to be able to pin or save context (e.g., my directory structure).
So my final verdict, at least after one morning: Pfind is a credible peer for ChatGPT4 for coding tasks.
## What will be the limiting factors on LLM improvementsIf you were a scale believer over the last few years, the progress weāve been seeing would have just made more sense. There is a story you can tell about how GPT-4ās amazing performance can be explained by some idiom library or lookup table which will never generalize. But thatās a story that none of the skeptics pre-registered.
As for the believers, you have people like Ilya, Dario, Gwern, etc more or less spelling out the slow takeoff weāve been seeing due to scaling as early as 12 years ago.
It seems pretty clear that some amount of scaling can get us to transformative AI - i.e. if you achieve the irreducible loss on these scaling curves, youāve made an AI thatās smart enough to automate most cognitive labor (including the labor required to make smarter AIs).
But most things in life are harder than in theory, and many theoretically possible things have just been intractably difficult for some reason or another (fusion power, flying cars, nanotech, etc). If self-play/synthetic data doesnāt work, the models look fucked - youāre never gonna get anywhere near that platonic irreducible loss. Also, the theoretical reason to expect scaling to keep working are murky, and the benchmarks on which scaling seems to lead to better performance have debatable generality.
So my tentative probabilities are: 70%: scaling + algorithmic progress + hardware advances will get us to AGI by 2040. 30%: the skeptic is right - LLMs and anything even roughly in that vein is fucked.
From Dwarkesh Patel. This is the piece that I've been waiting for someone to right. It doesn't matter if he is right, just the thought exercise of thinking through where the bottlenecks might be is really useful.
## Book notes: White Sun War
Mild spoilers ahead
A fictional account of a military history about a war between China and the US and allied forces over Taiwan set in 2028.
I generally enjoyed this book (the narrative was compelling) and found it to be an easy way to develop a feel for the general geography and challenges that a come with a Taiwan conflict.
The device for this book is really clever. The actual book is obviously about events that haven't yet happened and are in the future. But within the book, the author mentions the decision to write this as a narrative history a la Killer Angels as a way of bringing the "historical" people within it to life for a new generation. This confusing to recount but is quite clever within the book.
As of the end of 2023 (and I guess early 2024), the book feels remarkably current. There's lots in it about Ukraine and Russia that feels like it could've been written ~a week ago.
One of the most interesting aspects of the book to me was the degree to which fooling the other side's AI through feeding it false data is the key to victory. You can't beat the algorithm, but you can misdirect it.
Space Force features prominently. I wonder how true to life this part is.
For about the first ~1/3 of the book there's a lot of discussion about "bespoke algorithms", which made me chuckle. What does that even mean?
I wish I had as much faith in US industrial capacity as the author does!
The ending felt like the author decided he was done. It made me think that the most likely equilibrium for such a conflict was a stalemate where China can't quite be pushed off the island but also can't quite control it.
## US passport fact of the day## Brian Potter on the Apollo Program"In 1990, only 5% percent of Americans had a passport. Today, that number is 48%." pic.twitter.com/FWKF8IdxC2
— Devon āļø (@devonzuegel) December 19, 2023
Two things I took away from Brian Potter's recap of the Apollo Program:
Not every effort at weight reduction was solved through clever (if complicated) ideas like cold-strengthened aluminum or the common bulkhead. Much of the effort was achieved by pure brute force: parts would be fabricated, tested until failure, and then redesigned to be slimmer until they broke at exactly the required load (scaled by an appropriate safety factor).
Worth reading!
## The rise and fall of steel in PittsburghThe furnace proved that coke made from the nearby Connellsville seam of bituminous coal was uniquely able to be used in the blast furnaces that transformed iron ore into pig iron. For decades, coke made with coal sourced elsewhere proved unusable ā giving southwestern Pennsylvania an enviable competitive advantage as the second industrial revolution powered explosive demand for iron and then steel.
That technical innovation gave coal-rich Pittsburgh, which was already a successful region for energy-intensive industries like iron and glass, an overwhelming advantage in ore-based steel production.
via Chris Briem. Despite growing up in Pittsburgh, I found this explanation enormously helpful.
## Saffron fact of the dayBranca buys 3/4ths of the world's saffron, according to Eater, for their popular fernet drink.
## The dangers of coalUS coal power plants killed at least 460,000 people over the past 20 years - twice as many premature deaths as previously thought, with updated understanding of dangers of air pollution (PM2.5) https://t.co/ajJ1An3hBp
— Dr. Melissa C. Lott (@mclott) November 24, 2023
An interesting example of status quo bias. We know that coal is dirty and that it is a part of the electricity system, and these negative health effects are a part of normal operations, so we accept them.
## Mary Poppins was rightA spoonful of sugar really does help the medicine go down.
A recent study has found that a fruit-flavored medicine, taken once daily for six months, reduces a child's risk of developing multidrug-resistant tuberculosis (MDR-TB) disease by over 50%.
From News Minimalist as reported in Bloomberg.
1) How had this not already been tried? 2) How has this not already been rolled out to all children's medicine?
## The dominance of online courtship## Eiffel Tower CopyrightBREAKING: if you are looking for love in America, stay right here on the internet.
— Eric Klinenberg (@EricKlinenberg) December 14, 2023
Nobody is meeting through friends, in bars, at work, at college, or in the neighborhood any more. #ModernRomance ā¦@azizansariā© pic.twitter.com/lk36HZf6OQ
If you take a picture of the Eiffel Tower at night, you can't sell it without permission because the lights are copyrighted.
The Eiffel Towerās lighting and sparkling lights are protected by copyright, so professional use of images of the Eiffel Tower at night require prior authorization and may be subject to a fee. Professionals should therefore contact the Eiffel Tower's management company to learn about conditions for using the images depending on the case.
Via Tour Eiffel
The March 2011 9.0 earthquake in Japan was so strong that it altered the earth's axis and shortened the length of a day. From my friend Graham, more information available in Earth Sky
## 52 things I learned in 2023Brevard, NC. My favorite place I visited in 2023.
I borrowed this concept from Tom Whitwell as a way of cultivating a habit of curiosity. You can read his 2023 version here.
My 2023 highlights: I joined Macro Oceans full time, launched a low carbon cosmetics ingredient (if you want to see it in a consumer product, check this out) and built Recap Roswell, an AI-based summarizer of our local city council meetings.
Here are 52 things I learned along the way:
“Serius est quam cogitas (It's later than you think)”
A square watermelon. Putting this on my shopping list for when I visit Japan.
The Just City by Jo Walton, my favorite book in 2023.
“Do not complain beneath the stars about the lack of bright spots in your life”
Right now, I am thinking a lot about:
If you think weād have an interesting conversation about these topics or something else all together, reach out (jdilla.xyz@gmail.com) or book a meeting.
## OpenAI betSam Altman will be back as CEO of OpenAI within 60 days of today.
## How to configure your CustomGPT to send emailsHave you ever been deep in a ChatGPT conversation and wanted to email a piece of it to yourself to remember later? Or wanted to share part of a conversation with a friend?
Now with CustomGPTs, you can configure your GPT to send emails on your behalf using SendGrid for free. Hereās how you can do it.
To do this youāll need: * A free SendGrid account (you can sign up here). * An email address you want to use * Access to CustomGPTs
Note: SendGrid is free for up to 100 emails a month, beyond that youāll need to pay.
My instructions
When using the api.sendgrid.com action to send emails, always use the email address <TheEmailAddressYouVerifiedAbove@example.com> no matter what. The best sender name to use is
Name you Prefer to be called.
I havenāt tried this yet, but I suspect Iāll be able to configure the bot with multiple accounts. So Iāll be able to say email so and so from my work email or email my wife from my personal email.
If you get a chance to use this, Iād love to hear how youāre doing it. Drop me an email at jdilla.xyz@gmail.com from your next CustomGPT chat!
When Matthew and I were beginning to look at starting Macro Oceans, one of the things that got me interested in the opportunity was the mystery of it: if seaweed is so chemically rich and so easy to grow, why isn't it used for more things?
As we dug deeper into this question, I became convinced that there aren't fundamental reasons why seaweed can't be used for more things; instead, someone had to come along and make it happen.
While there have been many setbacks, false starts, and complexities over the past three years, I haven't seen anything that has changed my mind about this. Seaweed does have natural assets and it should be used for more things.
Today we announced the launch of our first product, Hydrating Marine Polysaccharides, which is our first contribution making this happen.
Hydrating Marine Polysaccharides is a bio active cosmetic ingredient, which means it's the thing that makes a skincare or haircare product make your skin feel better or hair look better. It has proven hydrating properties and a true clean beauty profile: zero waste, fully traceable to the farm in Alaska where it was grown. I'm excited to build on this in 2024!
Orthodox Jews, although they comprise about 0.2 percent of U.S. population, account for some 18 percent of so-called altruistic kidney donations (i.e., those where a living donor gives an organ to a recipient he or she doesnāt know).
From Mosaic Magazine
## Impact of ChatGPT on freelancing incomeNEW: Generative AI is already taking white collar jobs
— John Burn-Murdoch (@jburnmurdoch) November 10, 2023
An ingenious study by @xianghui90 @oren_reshef @Zhou_Yu_AI looked at what happened on a huge online freelancing platform after ChatGPT launched last year.
The answer? Freelancers got fewer jobs, and earned much less pic.twitter.com/kbtp8uDUPU
It is now so much cheaper to access intelligence on demand.
## HumanePredicting it now: Humane will be acqui-hired by OpenAI to be their hardware shop within 24 months.
## š¶ > š¶There are more pet dogs (~83 million) than children (~72 million) living in the United States.
## Open AI Dev Day reflectionsFrom State Affairs:
The Georgia Rural Health Innovation Center at Mercer University School of Medicine did an online statewide survey between January and April 2022 . The survey drew 1,651 people in the agriculture industry. It looked at their mental health and well-being. Hereās what the survey found:
- 29% of farmers said they thought of dying by suicide at least once a month.
- 42% of farmers have had suicidal thoughts at least once in the last year.
- 47% said they experienced loneliness at least once a month.
- 49% reported being sad or depressed at least once a month.
- 39% said they felt hopeless at least once a month.
Truly shocking numbers.
## Recap Roswell covered in Appen MediaMy side project, Recap Roswell was covered in Appen Media, our local news outlet this week:
āI feel like Roswell, Georgia, the United States would be healthier if we have more of those institutions, or micro institutions, where people are connecting and going through the process of deciding what they think they should do together and then, trying to make that happen in the real world,ā Dillard said.
In May, Dillard built Recap Roswell from scratch using ChatGPT, an AI-powered language model developed by the nonprofit organization OpenAI.
For each post, he creates a transcript from the video recording of the latest council meeting. Then, he runs the transcript through the OpenAI API, based on a chain of different requests which synthesizes it down to a quick summary.
āYou now are taking hundreds of hours of time and condensing it to something that someone, like yourself, can get through in 15 or 20 minutes,ā Dillard said.
I appreciate them taking the time to discuss the project with me!
## The rate of reports to child protective servicesI found these to be astoundingly high, so much so that I am wondering if there is an error in the data. From the National Library of Medicine:
37.4% of all children experience a child protective services investigation by age 18 years. Consistent with previous literature, we found a higher rate for African American children (53.0%) and the lowest rate for Asians/Pacific Islanders (10.2%).
I have no expertise in this area, but the point of the paper is trying to differentiate between first time investigations and follow up investigations, since presumably children that are investigated once are more likely to be investigated again.
In 2014, 4.57% of all US children had a maltreatment investigation... Of these, about half (2.39%) had no previous investigation in the 2003ā2014 database. After adjusting for database-first-time investigation rates as described in the preceding section, we estimate that 2.09% had a true first-time investigation.
It occurs to me that even though I'm a parent, I'm not sure what exactly would cause a CPS investigation and what wouldn't.
If this data is correct, I'm torn between being sad for a whole lot of families and wondering if we're over investigating.
## Opioid crisis fact of the dayThe opioid overdose crisis has been ongoing for over two decades in Canada and the U.S. The current mortality rate is greater than the worst years of the HIV/AIDs epidemic in these countries.
From The Conversation
## The US Armed Forces as a family businessMoreover, in some ways the military has become a āfamily businessā with over half of enlistees having a close family member with military service.
From The draft ended fifty years ago. Can the all-volunteer force survive another fifty?.
One of my less conventional beliefs is that the draft, while annoying for individual people, would be good for the United States because it ensures that the population is more broadly represented in the military.
Ironically, since the draft was ended, the military is the one US institution to see public trust in it increase.
## Lego VCYou can outperform most venture funds by buying LEGO.
— Will Manidis (@WillManidis) October 3, 2023
I analyzed the last 20 years of secondhand LEGO pricing data, and found randomly purchasing sets will match most VC's returns
if you're somewhat intentional about what you buy-- you massively outperform even the best firms pic.twitter.com/RjeuzHfAYq
I wonder to what degree this will carry forward.
## That's a big pigeon!The dodo, the giant, flightless, extinct bird was actually a type of pigeon. Imagine St. Mark's square being full of Dodos!
## Age limits for elected officials82% of Republicans and 76% of Democrats support putting a maximum age limit in place for elected officials in Washington, D.C. 82% of Democrats and 68% of Republicans favor one for Supreme Court justices.
This is from Pew Research's weekly round up.
I continue to think this is an underrated idea. If we accept age minimums, why shouldn't we accept age maximums?
## Semaglutide fact of the dayUnited Airlines Holdings Inc. would save $80 million a year if the average passenger weight falls by 10 pounds, Sheila Kahyaoglu, a Jefferies Financial analyst, estimated in a report Friday.
From Matt Levine's newsletter last week.
## 2023 Gardening UpdateIn 2022, I was finally stable enough to start an herb garden with my daughter. We started too late in the year to do anything ambitious and basically everything we planted died, except for a sage plant that bravely pushed through the winter.
In 2023, we were more organized. We did tomatoes, basil, oregano, rosemary, lavender, hyssop, and some wildflowers.
Our wildflowers did better than anything else. We actually spilled the seeds all in one spot then had a rain storm scatter them wildly and still they thrived more than anything else. Our front yard has been full of bumblebees and hummingbirds all year. Next year, we're going to forgo the lavender and hyssop, at least in the front patch where we don't get enough sun, and focus entirely on wildflowers.
The basil and tomato plant did well also. Our tomato plant would've probably produced 10-20 tomatoes except it took me too long to realize that the blackish spots I was seeing on it were a big deal. Next year, I'll be using neem oil on it from the beginning of the season.
The hardest thing about gardening is that it requires so much feel. I can't tell you how many times I looked up something happening with a plant to find that it could be a symptom of over watering or under watering... and then I had to try and deduce which is might be. Still, it gets easier with practice.
When I think ahead to next year, I think we'll most likely do: * One or two tomato plants in our raised garden * Basil in the pot I used for the tomato this year; potentially a second pot for basil * Lots more wildflowers in front of the house * I'll keep my trusty sage plant and the oregano, assuming they make it through the winter * I'll skip the rosemary (I don't use that much of it anyway), lavender and hyssop
Okay here's a ridiculous (but true) take: word processing is bad.
— Christian Keil (@pronounced_kyle) September 22, 2023
Before it, document length was bounded by the human capacity to copy long-form text.
After word processing, the tax & legal codes grew without check. And became incomprehensible to anyone but career experts. pic.twitter.com/tZMEBlQb22
Speculative, but interesting. A form of the medium is the message.
## Book thoughts: A Slave's CauseI did my best with this book, but I couldn't make it through.
A Slave's Cause by Manisha Sinha is a history of the abolition movement.
I picked it because I was interested in understanding how slavery came to be abolished.. I am still interested in this topic, but one of my goals for the year is to be more willing to put away books that aren't holding my interest, so that's what I'm doing for now. This says more about me as a reader than it does about the author and the book. I'm sure I was not her target reader!
The book is fantastically researched. It seems like the author found every single person in the historical record that opposed slavery and told their story. There are so many people who gave so much to the cause.
My biggest takeaways from the portion of the book I read are:
Perhaps I'll revisit this one in the future and if there is book written for more of a general audience that you know of, let me know as I'd happily start there.
## The implications of AI as a skill levelerEthan Mollick's fantastic One Useful Thing newsletter has an overview of a recent paper he did studying the impact of AI tools on BCG consultants.
One observation in particular stood out to me:
We also found something else interesting, an effect that is increasingly apparent in other studies of AI: it works as a skill leveler. The consultants who scored the worst when we assessed them at the start of the experiment had the biggest jump in their performance, 43%, when they got to use AI. The top consultants still got a boost, but less of one. Looking at these results, I do not think enough people are considering what it means when a technology raises all workers to the top tiers of performance. It may be like how it used to matter whether miners were good or bad at digging through rockā¦ until the steam shovel was invented and now differences in digging ability do not matter anymore. AI is not quite at that level of change, but skill levelling is going to have a big impact.
This data is invaluable, but I think the framing of it (through no fault of the author's) obscures how individuals should be using LLMs. From the perspective of studying the impact of LLMs on a population of BCG consultants, there are low performing consultants and high performing consultants. But as individuals, we are a mix of low performers and high performers depending on the task.
Therefore the implication is that we should be much less afraid of our weaknesses, especially in areas that are complementary to our strengths. The quality (skill?) of being willing to learn by doing is going to be increasingly important, since the LLM will help cover the flaws. Then the way to maximize one's impact is to pick projects where you have a relative strength (beyond the jagged frontier of AI, in Ethan's framing) and pair it with tasks where the LLM can provide complementary, replacement level support.
I can't claim to have mastered this, but over the past 6 months, I've experienced this first hand across a number of domains: programming and in my day job, about material science, chemistry, and technical writing about cosmetics.
My friend Kenneth quipped recently: "What will you do with infinite junior software engineers?", but it's even broader than that. You have infinite access to baseline expertise in basically anything. What will you do with it?
## ChinaTalk on CongressFantastic ChinaTalk episode on the role of Congress in American civic live with Philip Wallach:
The specific things that stood out to me:
ChinaTalk continues to be the podcast most likely to introduce me to something I didn't know I was interested in
## Alaska kelp harvest falls by 30%From Macro Oceans:
We spoke to half a dozen farmers across the state (if we missed you, please reach out!). All the data was self-reported and although we received updates from the same number of farmers as last year, we know of a few farmers who did not respond, so this total is probably a bit lower than reality. The results: total production in 2023 was 389,900 wet lbs, down 30% on last year.
It's clear to me that the bottleneck for the seaweed industry is customer demand.
## The etymology of filibustering## Trust and American Church attendanceIn the 19th century, americans would raise armies to support secession/revolution in latin-american nations. In many cases, the goal was to encourage USA annexation.
— Ryan McEntush (@rmcentush) September 10, 2023
This was called "filibustering", and the modern political term stems from this military act:ā¦ pic.twitter.com/Sr1xPNFKrU
So many gems in this piece by Ryan Burge, Church Attendance Used to Drive Up Trust, It Doesn't Anymore.
Trust is one of my favorite topics because I think it is one of those invisible things that makes all the difference. A high trust team can move faster and do things a low trust team can't do. Similarly, a high trust society can move quickly and do things a low trust society cannot do. Increasingly over the past 50 years, America is becoming a low trust society.
Some things that surprised me in the article:
As Burge points out, this isn't as simple of a story as it might seem. Trust and educational attainment are positively correlated and educational attainment is increasingly a driver of partisanship. However, the Republican coalition is filled with people that are less likely to believe that other people can be trusted than it was 50 years ago.
Quoting directly from Burge:
The main culprit for that growing divide is that those with low levels of education how grown more distrustful: 60% in the 1970s up to 77% in the 2010s. I think this should be ringing alarm bell for American democracy. There are lots of folks out there with low levels of education who are deeply distrustful of their fellow man.
One of the things that stood out most to me while reading Bowling Alone was the role that churches and other religious institutions played in preparing people to participate in civic life. [0] They were the training grounds of democracy where someone learns to lead at a small level, experiences what it's like, and then decides that they have the capability to take the next step. I know this is true for me; the very first times I led teams at work, I thought back to leading groups at my church in high school, what created credibility, and what destroyed it.
Burge hypothesizes that it might be due sorting, you're less likely to meet people that are unlike you and therefore are less likely become more trusting. I'm not sure if I agree with it, but I don't have a better hypothesis yet. But I do know that seeing this change is sad for me.
0: I can't link to this because I haven't imported blogposts from my old blog yet... shame on me!
## Government deficit food for thoughtBasic accounting ā the Kalecki equation ā shows that government deficits and corporate profits are on opposite sides of a ledger that sums to zero. Historically, thereās been a powerful statistical relationship between changes in the government deficit and subsequent changes in profits margins: major increases in deficits have led to rising profit margins over the next few years, and major decreases in deficits have led to falling profit margins. We have just seen one of the biggest decreases in the government deficit in history. It is very likely to be matched by a subsequent drop in profits.
From Entering the Super Bubble's Final Act by Jeremy Grantham.
I lack the experience to critique this perspective, but it is intuitive and I've found myself thinking about it repeatedly since I read the article a couple weeks ago.
## šØKoala fingerprints are nearly identical to human fingerprints, to the degree that they can interfere with forensic investigations.
## ChatGPT as a pull up bandMy friend Uri wrote a blog post about different types of supports:
Iām going to call any supportive device that degrades your ability to perform the activity un-aided an enfeebling support.
Of course, this also implies a definition for a better kind of support: a strengthening support not only helps you do something while supported, but also makes you better at doing the thing āby yourselfā once the support is taken away.
Before I get into what I disagree with, first I want to say I wholeheartedly endorse the distinction he draws between types of supports. It's really helpful to ask "is this a tool that's helping me build capability or is this a crutch that I'm becoming reliant upon?"
However, one of his examples of enfeebling supports caught my eye:
Tech legend and friend of the blog T.D. worries that chatGPT is degrading peopleās coding skills, making them slightly better at coding right now but worse (and more dependent on prosthetics) in the long run.
For me at least, ChatGPT is absolutely a strengthening support. I went years doing only minimal coding because I couldn't build anything I was excited about, sort of the equivalent of being able to do zero pull ups. With ChatGPT, I'm now able to finish things I care about, which is infectious, and I'm learning a bunch as I go. So I humbly disagree with tech legend TD!
## The culture that is Swiss brandingToblerone is moving it's chocolate production out of Switzerland and losing it's logo as a result.
The Matterhorn, that majestic mountain peak that graced Toblerone's packaging since 1970, holds a special place in the hearts of the Swiss people. In Switzerland, it's not just a mountain; it's an emblem of national pride. But hereās why Toblerone has to let go of itāthe Swissness Act of 2017, dictates that products should be intrinsically tied to the Swiss geography to bear the "Swiss-made" label and carry Switzerland-related imagery.
From the Trademark Factory's newsletter, which sadly doesn't have a link.
## Obesity trendsIn the past 5 decades, the obesity rate in the United States has never gone down two years in a row, paraphrased from the Odd Lots Episode linked above.
Totally astounding. I'd be on that changing over the next two years.
## Contingencies in history, 1848 editionThe Sicilian Revolution, the first one of the many revolutions of 1848, was kicked off by a single person putting up flyers in the name of a revolutionary committee that didn't exist.
You could argue that if something this small could kick off a revolution, then it was going to happen anyway... and yet, what if no one else had thought to do it?
Via The Rest is History.
## Maybe the placebo effect isn't real?Incredibly, the placebo effect is (mostly) not real.
— Jonatan Pallesen (@jonatanpallesen) August 21, 2023
It is a result of statistical confusion. Whenever you have a group with extreme values, they tend to exhibit regression to the mean. Eg. on average, sick people tend to become more healthy over time.
Thus if you give oneā¦ https://t.co/1s0qddGgk4 pic.twitter.com/8AOqGbyGtl
A more compelling explanation than any other I've heard to date.
## Forks
My favorite episode of television of the year so far, maybe in two years. It really taps into how being useful to others impacts a person's psyche.
"I wear suits now." Just perfect.
## Urban decline and renewal
Modern Europeās total urban population only surpassed the total urban population of the Roman Empire around the time of Isaac Newton, a whole 16 centuries after the ancient peak (which implies that the rate of urbanization was lower, as 17th century Europe had over 100 million inhabitants, a substantially larger population than estimated for the Early Roman Empire, whose population has estimates ranging from 50 to 75 million).
From Rafael Guthmann.
Weirdly, this reminded me of reading Dr. Zhivago. One of the things I'll always remember about that book is the way society deteriorated around Dr. Zhivago.
## The inner life of insectsA couple of excerpts:
Bees actively seek out drugsĀ such as nicotine and caffeine when given the choice and even self-medicate with nicotine when sick. Male fruit flies stressed by being deprived of mating opportunities prefer food containing alcohol (naturally present in fermenting fruit), and bees even show withdrawal symptoms when weaned off an alcohol-rich diet.
Recently weĀ confirmed this hunchĀ experimentally. We connected a bumblebee colony to an arena equipped with mobile balls on one side, immobile balls on the other, and an unobstructed path through the middle that led to a feeding station containing freely available sugar solution and pollen. Bees went out of their way to return again and again to a āplay areaā where they rolled the mobile balls in all directions and often for extended periods without a sugar reward, even though plenty of food was provided nearby.
A delightful read, via The Browser
## The adjacent possibleI really enjoyed this episode of Odd Lots with Dr. Ricardo Hausmann discussing the Atlas of Economic Complexity.
I had heard the term before, but hearing Dr. Hausmann describe how countries navigate the product space really brought it to life for me. It's a concept with a ton of richness not just for countries but also companies and even personal growth (running a 10k is in my adjacent possible, running a marathon is not).
One moment that stood out to me was right at the end when Joe Weisenthal asks how the US is doing and Dr. Hausmann actually says our economic complexity is falling and the guests just sort of blow right past it. Talk about a missed opportunity!
My hypothesis would be that the US is increasingly a market for consumers where the production work happens elsewhere. From an adjacent possible perspective, this is a bit dangerous because by not doing these activities we lose access to the adjacent possible.
## The lost generation of young foundersIn the most recent conversations with Tyler, Paul Graham and Tyler Cowen discuss the question: "Why are there so few great founders in their 20s today?"
While Graham challenges whether or not this is actually true, I want to offer an alternate explanation: they went into crypto and that turned out to be a dead end (for now at least).
Here are some major crypto busts and the ages of their founders:
So not everyone on that list fits the young founder profile, but had things turned out differently for the industry but here we have 3 founders that wouldāve been in their early twenties when their large and successful companies were founded. SBF was well on his way to creating young founder aura about himself when FTX collapsed. Without the failure of these companies / projects, we would have an additional cohort of young founders.
It strikes me that many of these founders were the same age as the people I spent time with while living in SF from 2014-2018, a period when companies like Uber, AirBnB, and Instagram had reached escape velocity. People were searching for the next big opportunity, and many found it in crypto.
I donāt think it was obvious at the time the main way that consumers would interact with crypto would be speculation. The crypto boom / bust cycle has gotten pretty toxic so people are going to see this as a value judgement, but I donāt mean it as one. I had lots of really enjoyable bar conversations with people that went into the industry about where it was going to go and none of them were saying āyeah, this is only good for fake internet speculation.ā There was real optimism that killer consumer or industrial use cases would be discovered that leveraged the decentralized nature of crypto currencies and this would lead to new economic models and usher in a new era of peace and prosperity. I never got the crypto use case, but the thing that made me think I might be wrong was the passion of the people leading it to create something new and useful.
This doesnāt excuse any wrong doing by anyone in the industry, but I donāt think that most of the people that got into the industry intended for it to be as scam ridden as it has become. I think it turned out that the speculative nature of crypto projects has outweighed the usefulness of decentralization for end users, leading to Ponzi-esque business models.
Had this turned out differently ā if crypto had found real, value producing consumer traction or if the dead end had been more obvious ā I think we would continue to have the same pipeline of young founders.
Because of this, I expect the pipeline to return to form in the age of LLMs and climate tech start ups. Weāll see!
## Charles II: the world's largest landownervia the Madison Trust. The full list is at the link, but the top 3 are the English Crown, the Catholic Church, and the Inuit People of Nunavut via an agreement with the Canadian government. Lots of Australian individuals represented as well.
## What types of innovation create human flourishing?Iāve been pondering this question a bit lately as a prompt for where itās worth focusing time and attention. I do mean flourishing here in the broadest, most sincere sense, although noting that it is tough to truly flourish if you donāt have access to shelter or clean water.
A comment a good friend of mine made a long time ago was about the impact of energy production on human productivity and thriving. Through a certain lens, more efficient and cleaner sources of energy are the only things that matter. This is why people get so excited about fusion.
I think a similar exercise can be done for public health interventions. Access to clean water, for instance, seems like basically a sure bet for adding healthy, productive years to peopleās lives.
So here is my hypothesis list, written not with the assumption that is it is all correct, but so that I can refine it over time.
Iām almost definitely missing something and will refine this over time.
## A model of the seaweed industryI wrote an overview of the seaweed industry for Macro Oceans:
This is a Twitter thread by George Mack:
What is ignored or neglected by the media -- but will be studied by historians?
— George Mack (@george__mack) July 13, 2023
Here's the full list of 25 examples: pic.twitter.com/yzj9yGjDz9
The ones that stood out to me most:
I think the frame of "what will seem consequential to us in hindsight" is an interesting one!
## Invention of the telephoneIf Alexander Graham Bell had not secured the patent for inventing the telephone, Elisha Gray would have gotten it because they both applied for the telephone patent on the same day (Feb 14, 1876).
That is via Noahpinion in his interview with Kevin Kelley. Kelley's point is that invention is an emergent property of society (what he calls the technium), rather than an individual act.
## HIV in AustraliaHIV transmission in Sydney has plunged more than 88% from the 2008-2012 average to just 11 cases per year.
It seems it will be possible to greatly reduce or even eliminate HIV transmission.
Via Tyler Cowen.
## Square watermelons
From Wikipedia:
Square or cube watermelons are watermelons grown into the shape of a cube. Cube watermelons are commonly sold in Japan, where they are essentially ornamental and are often very expensive, with prices as high as US$200.
They are grown in boxes, which form them to their distinctive shape. I discovered these via the Kroger app, which has distinctly wonderful food facts as it's loading screen.
## Practicing scalesI think about this Tyler Cowen question a lot:
What is it you do to train that is comparable to a pianist practicing scales?
I'm a product manager by trade and, until recently, never really been satisfied with the answers I've come up with. However, I've found one I really like.
I try to come up with two tweet length product ideas per day. I don't judge the ideas. They can be big or small. They just have to be plausible product pitches. I've been doing this for a couple of weeks now and it has been a small enough exercise that I can commit to doing it daily but big enough that I can feel myself stretching. Most importantly, I've noticed more, better ideas coming to me at other parts of the day.
Somewhat related: I really enjoyed this Range Widely post about "pouring out your lesser ideas to get to the great ones."
## Where trust comes fromI found this video to be a helpful distillation of concepts I'd heard before with a couple things that were new to me.
I'd known that trust is a combination of credibility, reliability, intimacy, and self interest, but I hadn't heard the sub components before:
Credibility
Reliability
Intimacy
Self interest (destroys trust)
The numbers are really small, both in comparison to US boarder crossings and (especially) the population of China. Still the increase is an interesting in what it suggests about the situation in China.
## The longevity of infrastructureThe Whiting hydroelectric power plant was built in 1891 and still operates today, and the 2018 Camp Fire in California was caused by a PG&E transmission line built in 1921.
From The Grid part IV by Brian Potter
## Hottest days everPost by @hausfathView on Threads
As I said before, two things are true at the same time right now at once: 1. We are not reducing emissions quickly enough 2. The speed with electrification is happening is underrated
## Books: The Golden Spruce
I enjoyed this book less because of the story and more because of the sense of place. Like taking a trip to the remote areas of Alaska and British Columbia ā and in some ways better.
## Disappearing navigation elements (powered by ChatGPT)The navigation elements on my blog now fade out as you scroll down my blog so as not to clutter your view. This has been bothering me for a while so I decided to clean it up.
You can see a demo below or by scrolling on your screen :)
You can see how I did it here
## Childless citiesThis piece from the Financial Times really resonated with me.
It hits on something I've experienced: the difficulty of finding a community in an American city that is affordable enough to live in, safe enough to let children play in, and has access to good education. It's so hard to find. In most parts of the US, if you want to have these things, you're pushed to the suburbs.
It seems to me that urban areas have an amazing lack of urgency around this problem.
While reading, I couldn't help thinking of this picture I took during my first week living in ZĆ¼rich:
In what US city could you imagine this scene?
A group of grade school girls on their way to school all by themselves, without a parent in sight. One of my favorite things about living in Switzerland was that it was not just possible to live in an urban area, but easy.
## Change log: image responsivenessOne of my proudest accomplishments from the past year is coding this website with the help of GPT.
When I first built the site, I wasn't sure I was going to finish it and GPT didn't have sharing features, so then I was done and I couldn't share with others what it was like, but it was really impactful for me. This skill that had always been outside of my grasp was now something I could do... and what did that mean for other skills?
Since finishing the site's MVP, I've periodically made changes, but haven't put all the effort into sharing them. I'm going to try and do more of that though, in part because ChatGPT's sharing features make it so easy to do.
So with that prologue, here's my first change log post: images are now responsive on the site. This was annoying me for a while and I'm glad I took the 15 minutes to fix it.
Here's how things looked before:
And here's how things look after:
You can see how I did it here
## Marathons and the increase in mortalityJust to give you a sense, the mortality rate for something like cardiac arrest or a heart attack goes up by about 15 to 20% on the day of a marathon. Now, most people are not having cardiac arrest or a heart attack, so the aggregate impact on a city might be limited. But I think if I were to talk to people about the Boston Marathon bombings, most people would say that was a horrific event. But more people die because of marathon-associated road closures every year in a given city with a large marathon than died in the Boston Marathon bombings. But the bombings, what they did are so salient to us. Deaths in these other channels, we donāt even think about that.
That is from David Epstein's newsletter Range Widely. The cause is that the closure of roads along the marathon route makes it difficult to get to hospitals quickly.
An idle thought: I wonder if the same holds true for cardiac arrests during rush hour traffic?
## Why are there so many Thai restaurants in the US?One of my favorite stories of the year. From Vice:
Comparatively, according to a representative from the Royal Thai Embassy in DC, there are just 300,000 Thai-Americansāless than 1 percent the size of the the Mexican-American population. Yet there are an estimated 5,342 Thai restaurants in the United States, compared to around 54,000 Mexican restaurants; thatās ten times the population-to-restaurant ratio. So, why are there so many Thai restaurants in the US?
The Thai government has created a company, the Global Thai Restaurant Company, to make it easier to start and run Thai restaurants. If I understand correctly, they aren't quite franchised (as in sharing a brand name), but effectively provide a pre-planned out restaurant. More from the article:
The Ministry of Commerceās Department of Export Promotion, most likely run by bureaucrats rather than restaurateurs, drew up prototypes for three different āmaster restaurants,ā which investors could choose as a sort of prefabricated restaurant plan, from aesthetic to menu offerings. Elephant Jump would be the fast casual option, at $5 to $15 per person; Cool Basil would be the mid-priced option at $15 to $25 a head; and the Golden Leaf prototype would cost diners $25 to $30, with dĆ©cor featuring āauthentic Thai fabrics and objets dāart.ā
Why go to the trouble? To increase exports and travel to Thailand through gastrodiplomacy.
## The value of speed to marketPost by @buccocapitalView on Threads
I interpret this in a couple of ways:
I also thought it was interesting that the comparative benefit (top quartile vs. bottom quartile) is greater in manufacturing than it is in software.
## Update on Meta(verse)Truly incredible statistic from the Nation on Meta's Metaverse, Horizon Worlds:
Metaverse platform, had only 38 active daily users. The Guardian reported that the monetized content ecosystem in Metaās flagship product Horizon Worlds produced no more than $470 in revenue globally. Thirty-eight active users. Four hundred and seventy dollars.
I don't consider this to be a definitive statement on the concept of the metaverse, but wow are those numbers small. Found via The Browser.
## Shipping profits during the pandemic"In 3 yrs from 2020 to 2022, the shipping industry generated as much profit as in previous six decades combined. Maerskās annual pre-tax income soared from $967mn in 2019 to $30.2bn in 2022 ā more than investment bank Goldman Sachs or Facebook-owner Meta."https://t.co/Ri0t1sdflo
— Adam Tooze (@adam_tooze) June 23, 2023
"In 3 yrs from 2020 to 2022, the shipping industry generated as much profit as in previous six decades combined." Just incredible the way that the pandemic created ripples through different industries.
## Wind and Solar generation are outpacing coalFrom Politico:
Wind and solar generated more electricity than coal through May, an E&E News review of federal data shows, marking the first time renewables have outpaced the former king of American power over a five-month period.
I believe two things are true at once: 1) we are not reducing emissions quickly enough; 2) the speed with which we are moving away from greenhouse gasses and towards electrification is underrated. ## Smoking and obesity correlations
From CafƩ Anne:
## The EV transitionI've long suspected an inverse correlation between smoking rates and obesity rates. As society, it seems, weāve merely swapped one distraction strategy for another in an effort to prevent the VOLCANO OF INNER PAIN from erupting into our everyday lives, haha!
The inverse correlation is nearly perfect. The percentage of people who smoke dropped from 42% in 1965 to about 12% today. Over the same period, the obesity rate climbed from 14% to 43%!
My conclusion? Symmetry is cool! Enjoy your sandwich!
An analysis from Boston Consulting Group found that between 25 and 80 percent of gas stations nationwide could be unprofitable in 12 years ā and that analysis was conducted in 2019, before a slate of new policies, including federal tax credits, were passed to promote electric vehicles.
From Grist.
It's not as simple as gas stations becoming charging centers because most people will charge at home (no one fills up their car with gas at home each night) and charging on the road takes longer and so is more likely to be coupled with things besides a convenience store.
Apparently repurposing the sites is easier said than done because of the contamination from the underground gas tanks.
## Robert Oppenheimer learned Dutch in 6 weeksThe Rest is History is doing Robert Oppenheimer, the father of the atomic bomb, this week.
Here's a fact that blew my mind: he learned Dutch in just 6 weeks in order to give a technical lecture at a university in the Netherlands.
This makes me feel way less proud of the three years I've spent getting to A2 in German...
## Nigerian beer fact of the dayNigeria drinks more Guinness than Ireland and is the second leading market in the world, behind the UK.
via CNN
Somewhat related:
The Guinness Book of World Records was started by a Guinness executive in 1951 to settle a debate: what is the fastest game bird in Europe.
## Growing mastery and agencyWe should be thinking much harder about ensuring children can make meaningful contributions, and we should be teaching them in ways that are sensitive to the context of the real world. We are not looking for a job but opportunities for mastery: learning and practice beyond the depth one would find along the common path, which demands no such thing.
That is from Simon Sarris's article in Palladium about how a schooling isn't enough for young people.
At this point, probably my most unconventional belief is that we should be giving teenagers more apprenticeship opportunities at companies. I feel weird saying this because it sounds like I want to return to a world where children worked in coal mines, which is very much not what I want. Instead, I think there is a type of learning that happens best hands on with real stakes and that by keeping children away from it we are doing them a disservice.
One of the things I notice with my children, who are very young (under 1 and 3), is that they are happiest when working on something that is outside their comfort zone, but within their capability, especially when it matters to the rest of the family. My three year old has had a toy cleaning set that she hardly ever played with but now uses daily to sweep up after dinner. I don't want to draw conclusions that are too sweeping from what I see observing my kids, but I do suspect there is something there.
## Diabetes in IndiaI found this to be astounding:
A study by the Madras Diabetes Research Foundation and Indian Council of Medical Research has found that over a quarter of India's population is either diabetic or in the pre-diabetic stage, with 101 million people living with diabetes and another 136 million in the pre-diabetes stage. The study also found that 35.5% of the population has hypertension, 28.6% is obese according to BMI measures, and 39.5% has abdominal obesity. The study is one of the largest ever attempted, with over 113,000 participants across 31 states and union territories in the country, and is highly representative of the Indian population. The findings can be used by states to develop health policies to target non-communicable diseases.
That is from News Minimalist, easily my favorite new media of 2023.
## Plant familiesSquash, zucchini, cucumbers, and watermelons are all from the same plant family: cucurbits.
Tomatoes, potatoes, eggplants, and peppers are all from the same plant family: solanaceae.
via my friend Matthew
## Light years are relativeI was today years old when I learned that "light years" is actually relative to time on Earth š¤Æ
— Steven Tey (@steventey) June 11, 2023
E.g. if you're traveling at 99.99% the speed of light to a star that's 4 light years away, it would take you just 3 weeks.
But to your loved ones on Earth, those 3 weeks = 4 years. pic.twitter.com/aMxNqdLswK
See also this article about time in Harper's Magazine. I feel like the more I learn about time the less I know.
Aside: I feel like there is an awesome pop culture book to be written about relativity and quantum mechanics that makes it accessible for people like me; or maybe it exists and I haven't found it yet?
## Horse lineagesEvery thoroughbred alive today is an offspring of just three stallions: Darley Arabian, Godolphin Arabian, and Byerley Turk, and a select number of mares dating back to 17th and 18th century England. Darley Arabian and Godolphin Arabian were both said to have come from Yemen. The origins of Byerly Turk are less known although he was captured in the Battle of Buda in Hungary.
From 3 Quarks Daily via The Browser
## Radiation fact of the dayThe natural rate of background radiation at the US capitol is above the acceptable threshold for a nuclear power plant.
Via Jack Devanney
## Leadership and performance in the NBAOn the relationship between leadership styles and performance, using NBA coaches and NBA players as the data set:
Specifically, experiencing abusive leadership at any point across the 6 years of the study shifted the trajectory of player performance downward. By moving beyond static relationships and demonstrating career-long performance effects, this finding extends inferences about abusive leadership from prior research. Giving added importance to this finding, the task performance measure was derived from objective performance data (rather than supervisor ratings that typify prior research), and the measure of task performance (i.e., player efficiency) is routinely used within the industry for major personnel decisions (e.g., hiring, salary, renewal).
Here is the paper.
This feels like a giant warning sign to me: look at how much impact positive management can make. Look at how much abusive management costs. If coaching style matters this much for NBA players, how much more important is parenting style?
## Engagements in America"Instead of 2.8M engagements this year, we're expecting 2.4M."
— Squawk on the Street (@SquawkStreet) June 8, 2023
Signet Jewelers CEO Gina Drosos tells @SaraEisen, @carlquintanilla and @davidfaber that $SIG expects the lull in engagements to pass by the end of 2023. pic.twitter.com/LPhpEcUy6v
Things I learned from this:
Thinking back, I began dating my wife 3 years and 1 month before we got engaged.
## The cost of a new ladder fire truckI'm not sure why this caught my eye, but it did. Our local city just agreed to purchase a new ladder fire truck, a steal at just $1,449,853.
I'm not sure if I would've predicted more or less.
## The Stack FallacyBy Anush Sharma. There's a lot of good stuff in here that I'll just quote directly:
## Switzerland has a navy?Stack fallacy is the mistaken belief that it is trivial to build the layer above yours.
Database companies believe that SaaS apps are ājust a database appā ā this gives them false confidence that they can easily build, compete and win in this new market.
In a surprising way, it is far easier to innovate down the stack than up the stack. The reason for this is that you are yourself a natural customer of the lower layers. Apple knew what it wanted from an ideal future microprocessor. It did not have the skills necessary to build it, but the customer needs were well understood. Technical skills can be bought/acquired, whereas it is very hard to buy a deep understanding of market needs.
Product management is the art of knowing what to build. The stack fallacy provides insights into why companies keep failing at the obvious things ā āthings so close to their reach that they can surely build. The answer may be that the what is 100 times more important than the how.
Apparently so, although it used to be bigger.
the Swiss flag currently flies on only 14 ocean vessels, well down from the 50 ships in 2017. In that year, an embarrassing shipping fraud cost the taxpayer CHF215 million ($237 million), forcing a rethink of maritime strategy.
via SwissInfo.ch
## What Stripe Gets RightI recently made the decision to leave Stripe and join Macro Oceans full time. This was a difficult decision and was more about not wanting to miss the opportunity to build Macro Oceans rather than quitting Stripe.
While I was working at YouTube, I began doing something that really helped me: team and company cultures the way I would about a friend. They have a lot of good qualities and some faults. [0]
But when youāre working somewhere everyday, even little faults can be really grating, the way that you might have a friend you really enjoy who you canāt stand after two weeks traveling together.
Part of my breakthrough was to spend more time focusing on the virtues rather than the faults. I also began to accept that the faults developed over time and would take time to change. This made it easier for me to navigate through the things that drove me a little crazy: it was part of what made the place what it was.
With this context, I want to reflect a little bit about what Stripe gets right, if only so I can remember it. These arenāt the only things about Stripe or a balanced picture of what itās like to work there, but the things I want to remember and to emulate. [1]
0: At every company where Iāve worked as an adult, with one possible exception, the good qualities outweighed the faults pretty significantly.
1: I enjoyed my time at Stripe and on some level wish it had been longer. If youāre considering a job at Stripe and want more detail about my experience, feel free to reach out: jdillaxyz@gmail.com.
2: The possible exception here is Bain. The difference that at Stripe the talent level is high and distributed across functions whereas Bain everyone basically did the same thing ā corporate strategy.
## US marriage rates
Amazing statistics from Pew. In 1980 63% of 25 year olds were married and 39% had children at home!
I'm agnostic to whether or not this is a good or a bad change, but I'm amazed at how different it is.
## Crucifixion factsThe only accounts we have of a crucifixion by the Romans are the 4 gospel accounts.
That is via Tom Holland on The Rest Is History.
## Notes on You and Your Research by Richard HammingYou and Your Research is a talk given by Richard Hamming about how to navigate onceās career. Itās packed with insight and is on the short list of essays I feel like I should return to every couple of years.
Some of the perspectives that resonated with me:
The cost of fighting the system: āBy realizing you have to use the system and studying how to get the system to do your work, you learn how to adapt the system to your desires. Or you can fight it steadily, as a small undeclared war, for the whole of your life.ā Talk about something I needed to internalize when I was 20.
The role of courage in doing great work. A recurring theme is that you have to believe you are capable of solving important problems and work with the intention of solving them. This is definitely something Iāve felt at companies where Iāve worked; some places there is an unsaid feeling that certain things are beyond us as a company where the more successful places Iāve worked (Stripe, YouTube) felt like anything was possible and invited people to believe that.
The humility required to do great work repeatedly. As you become more successful, it becomes harder to work on small problems because youāre now a big person who does big things; but major breakthroughs usually start in small places. This is another theme Iāve felt places where Iāve worked. Thereās something interesting, but itās not yet at scale yet, so it gets ignored, the problem being that when it does scale, youāre too late to make an impact.
The role of framing ones work. You should actively be reframing your work towards what matters; even mundane tasks can be elevated if you think about what makes them mundane or where they fit in the greater context of the project or the company. Similarly, you can also use your weaknesses to uncover opportunities to do great work as other people likely share those weaknesses, so if you find a way to make this task easy for them, then youāve done something substantial.
The compounding impact of effort. āKnowledge and productivity are like compound interest. Given two people of approximately the same ability, one person who works ten percent more than the other, the latter will more than twice outproduce the former.ā Even more important when you consider the importance of picking the problems you work on (vs. just adding hours to the day).
Mapping the field. To do truly great work, you need to have a mental map of your field, what the great unsolved problems are, which ones on the brink of being solvable. These are the āimportant problemsā; you also want to have ideas on likely paths for solving them. This helps you notice when there is an opportunity to step in. As an example, time travel isnāt an important problem because there isnāt a āline of attackā, but AGI is because we do.
Find the flaws. Almost a corollary of the previous note, you need to have a sense for where current accepted wisdom is flawed or incomplete (in Hammings words, what are the faults in the current theory). This requires tolerance of ambiguity ā you accept the current wisdom but also poke at it
Time to think. You should be reserving time to think about big problems and big solutions to those problems. Hamming reserved Friday afternoons for ābig thoughtsā.
Be opportunistic. When an opportunity to solve an important problem opens up, drop everything to go after it.
Make room for serendipity Keep your door open to others.
Presentation and selling matters; you have to be able to convince other people of your ideas, which requires meeting those people where they are.
The impact of WW2 on ingenuity. He notes, speaking about WW2: āThe war forced them into an uncomfortable situation and forced them to have an open mind, and then they drafted off of that. But the current generation doesnāt have that.ā This is a different post, but Iāve seen this theme in a couple of different things Iāve read where the generation that lives through the effort to win WW2 learns certain habits about how to work that they struggle to verbalize. There seems to be a practicality, especially around the little details that make the difference between a successful project and a failure, a desire to see up close and do all of the steps oneself, an simplicity of communication and expectation that others will contribute to the group.
1 million more horses served in World War 2 than World War 1 (2.5 million to 1.5 million).
That is from the The Rest Is History episode on the Year 1940.
## Pair programming with ChatGPTI built this blog from scratch using ChatGPT. I want to use that experience to explore what it taught me about how to use LLMs in my work and forecast the potential impact of LLMs.
First, some context around my skill level and the project. By trade, Iām a product manager without a deep technical background. I spend a lot of time around engineers, but I donāt code. About 10 years ago, I took a ālearn to codeā class which culminated in building a personal blog as the capstone assignment. I finished the project, but barely. From that point on, I settled in at the level of being comfortable scripting in Python, but uncomfortable doing anything beyond that.
About 3 weeks ago, I decided to use a Friday morning to see how far I could get in replacing my blog hosted on Squarespace with one I wrote myself using ChatGPT.
The project is simple Python Flask app with minimal styling. It incorporates a blog, a few static pages, a CMS, and an authentication system. The backend is powered by a PostgreSQL database, which Iāve been told is overkill, but it was important to me to prove to myself I could build an app with a database.
If you had asked me how long this would take me to do on my own, I wouldāve estimated about 2-3 weeks of full time, focused work. Doing it on nights and weekends, at least 2 months, maybe more. Using ChatGPT4, I had a working prototype by the end of the first hour and about ~25 total working hours later (spread across 3 weeks of morning work, so 30-45 minutes at a time), it was complete and deployed in production.
Iām not going to pretend that this is the most complicated piece of software ever written, but it is something that is well outside of my skill level without LLM assistance. In a way, focusing on the time spent to accomplish the task is misses the point; it was fun. I was hesitant at first, but as the project took I grew confident. I began to expect that I would be able to solve problems in a domain that had been out of my reach.
My workflow has evolved through the project, but at this point it looks lie this: Iāve got a main tab open with a GPT4 chat focused on the project (the same chat every time); the next tab over is a GPT3.5 chat (usually a new chat each time).
The GPT4 chat is where the main action happens and the GPT3.5 chat is where I ask the easy stuff that doesnāt require project context(e.g., remind me how to check what branch Iām in in the terminal using git). This helps me conserve GPT4 queries and is much faster overall. I tried to do more serious coding with GPT3.5, but the quality was so much lower that it wasnāt worth it.
At the beginning of the project I would regularly remind the GPT4 chat of the overall project and my goals at the beginning of each session but Iāve done less of this as time has gone on and not seen a decline in results.
In the main GPT4 thread I run through this basic loop:
Product scoping: I suggest the next feature for us to work on and outline what I think the requirements, both what is in scope and what can safely be out of scope. I play an important role here in making the scope size manageable. As an example, āLetās start by organizing the directory structure; how should I do that?ā is much better than āI want to build a blog, can you do that for me?ā Iāve also had the best results when I ask ChatGPT what else thinks we should consider in scope and then explicitly include or exclude it.
Implementation path: I then ask ChatGPT to suggest a plan for implementing this feature. So far, Iāve had the most success when I donāt suggest an implementation path at the outset. My biggest mistakes have been when I have had preconceived notions about how to implement something. ChatGPT and I usually discuss the pros and cons of different paths, and then we select one. Once weāve chosen an implementation path, I copy and paste the most relevant files or a skeleton outline of the directory so that GPT has context on the coding.
Coding: ChatGPT does almost all of the coding for me. We move step-by-step through the implementation plan. Occasionally Iāll catch ChatGPT drifitng on a variable name and fix it, but itās not uncommon for me to copy and paste entire files in and ask ChatGPT to work the code into the file for me. The biggest benefit from using ChatGPT here is itās knowledge of Python packages that I can leverage and its knowledge of how to use those packages.
Testing: Once the code is implemented, I test the app on my local machine, feeding any bugs I back to ChatGPT. I copy and paste the error message and ask how we should fix it. This then kicks off its own mini-loop of implementation path discussion -> coding -> more testing. Once the app runs locally, I then repeat the same loop in production. When everything works, I move on to the next feature.
Typically a small feature (e.g., moving to a secrets file) can be built in ~30-45 minutes. Something more complex like an image uploader or a tags feature might take 1-3 hours.[0] Iāve included a sample feature at the end of this post for a real life example of the above.
As mentioned above, the largest problems I had were when I assumed a particular path of implementation. One example here is choosing a hosting provider. Because Iāve worked with Firebase/Google Cloud before on a project with a friend I suggested we use that as a backend. (That sound you hear in the background is experienced programmers laughing at me!)
This added a lot of complexity my deployment. I got it working, but ultimately ended up switching to Render, which was simpler and cheaper. If I had gone through the loop above, I wouldāve avoided this delay. My lesson from this is that ChatGPT isnāt going to push back on a choice if I donāt explicitly ask for push back. Itās going to try and make it work.
The second biggest pitfall is managing ChatGPTās context window. Even skimming the code, I noticed directories or files would get misnamed (e.g., image/ would turn into img/).[1] This usually isnāt that big of a deal, but a big reason why I end up copying and pasting in my files into ChatGPT is to catch exactly this type of mistake.[2] Worst case, I catch it running the app locally.
Finally, I had one issue in deployment where GPT became pretty distracted by the possibility that there was a bug in Renderās deployment engine. I was pretty sure this wasnāt the case as the app had worked in the previous deploy, but every answer kept coming back to contacting Render customer support until I explicitly asked it to rule this possibility out. Once I did this, we found and fixed the problem.
Itās worth reflecting for a moment on just how human these pitfalls are. Anyone who has ever been a manager has had a situation where the team doesnāt push back on a direction because they werenāt explicitly asked to, small changes in naming are akin to typos and the who among us has not been fixated on the idea that there could be a bug in someone elseās code that is causing our problems?
The potential economic impacts of LLMs are a hot topic of conversation. I think itās interesting to look at this project through that lens.
Previously, I was using Squarespace for hosting + website design, costing ~$168 per year. Now Iām using Render for hosting, costing ~$84 per year and Iāve got a monthly chat ChatGPT membership of $20 per month. So the LLM has reduced the time cost of customization and shifted spending towards hosting services and, of course, LLM access. Including the full cost of the ChatGPT membership for the month I spent building puts me at a savings of $64.
At least at first glance, this fits for me:
Reducing the cost of something means you get more of it. It just became a lot less expensive to create software tools. I would predict this means that we get a lot more of them in a lot more niche areas. Business problems that couldnāt support a custom solution before will absolutely get one.
Somewhere over the course of the last month I saw a paper that studied the impact of LLMs in a call center. That study, which of course I canāt find, found that the LLMs had the greatest impact on the lowest productivity workers: new workers and under performing workers.
This definitely matches my experience. For about 6 months now, Iāve been using ChatGPT on things directly related to my expertise: product management. Iāll regularly feed in product requirement docs, strategy docs, important emails. The feedback I get is useful, but marginal. It makes it 5-10% better, not 50% better.
But programming is well outside my area of expertise. Iām not even sure I can characterize the impact of ChatGPT in percentage terms. Itās an order of magnitude difference.
The world I know well is software engineering teams. Coming out of this experience, it became clear to me that a lot of change is coming to this world.
At minimum, committing LLM generated code is going to become a part of the Product Manager and Designer work flow. I suspect that within 2-3 years, a PM of my level will be expected to come to the meeting with a strategy and a prototype product (maybe 2-3) running in production.[3]
Somewhat more speculatively, Iād assume that Product, Engineering, and Design are going to become more similar as disciplines, with the differences becoming one of emphasis, sort of like a Product Manager vs. a Technical Product Manager, with TPMs focused on more heavily technical codebases. So instead of having teams with ~1 PM, ~5 Engineers, ~1 designer, youāll have a mix of Product, Design, and Engineering focused creators.
Finally, I want to endorse this tweet from Amjad Masai:
People constantly ask me āgiven how good AI is at coding is it still worth it to learn to code?āā
— Amjad Masad ā (@amasad) April 16, 2023
Are you crazy? The ROI on learning to code has 10xād with AI.
Learn the basics and with some tenacity you can build your dream MVP.
Links to get started in thread:
At the time I first saw the tweet, I sort of dismissed it, but after this experience, I think heās right. The little bit of technical skill I had (e.g., understanding how to open the terminal, knowing how to write a little python) has opened up a whole new world of potential for me.
Where does this ultimately take us? I suspect that software development and a lot of other things is going to look a lot more like the Creator economy: low barriers to entry, high returns for the biggest stars, and a good living for those who can find their niche.
I suspect that this extends beyond software as well. There has never been a better time to be curious and self taught.
I wouldnāt be a PM if I didnāt come out of this experience without a list of feature requests.
[0]: Itās worth keeping in mind here when looking at these time estimates that these are features that I would be very unlikely to be able to build on my own.
[1]: GPT3.5 has way more problems with this than GPT4.
[2]: I find it also helps with managing the context window.
[3]: I see some echos here to analytical tools. Some PMs Iāve known a generation ahead of me have told me that when they got started knowing SQL wasnāt expected and that over the course of their career it became a part of the PM job.
Potentially not the best example since not much code was written, but a good example of a problem I didnāt know how to solve myself. Total time including redeploys and testing: about 1 hour
Me: hello! I would like to work on how images are stored and used on my blog.
right now, I have a /images route where I upload images. this ends up in a app/static/img directory which I reference in my markdown to get the images to show up. this works really well.
however, Iāve noticed a problem. every time I redeploy the site, all of my images get deleted. the reason for this is that Render rebuilds my app from github and none of these images are on github, theyāre only in the production environment.
What are my options to solve this problem? Ideally it will stay pretty simple and low cost.
ChatGPT: You have several options to tackle this issue. Here are a few:
Cloud Storage: One of the most common approaches is to store your images on a cloud storage service, such as Amazon S3, Google Cloud Storage, or Azure Blob Storage. These services are designed for storing and retrieving any amount of data at any time, and they provide high durability and availability. When you upload an image, instead of saving it to the local filesystem, you would upload it to the cloud storage. In your posts, you would then reference the URL of the image on the cloud storage. This approach is scalable and doesn't depend on your application's deployment.
Database Storage: Another approach is to store the images in your database. However, this can quickly use up your database storage space and may not be as performant as other solutions, especially for large images.
Use a Content Delivery Network (CDN): Services like Cloudinary or Imgix can store your images and also offer additional features like on-the-fly image resizing and optimization. They provide SDKs to upload images directly from your server and give back a URL that can be saved in your database and used in your application.
Volume Storage (Render Specific): If you're using a platform like Render, you could consider using their SSD-based persistent disk option. This allows you to have a separate disk that isn't wiped out when your app is redeployed. You would have to configure your app to store images on this disk.
Choosing the right option depends on your specific use case, the volume of images you expect to handle, your budget, and the complexity you're willing to introduce into your application. The cloud storage option is often a good compromise between cost, complexity, and performance.
Me: Which of these options will be the lowest cost?
ChatGPT: The cost of each option can vary significantly based on the amount of storage you need, the amount of data transferred, and any additional features you use. Here's a general comparison:
Cloud Storage (Amazon S3, Google Cloud Storage, Azure Blob Storage): These services typically charge based on the amount of data stored per month, the amount of data transferred out of their network, and the number of requests made (e.g., upload, download). For a small to moderate amount of usage, these costs can be quite low (a few dollars per month).
Database Storage: If you're already using a database, it might not cost you any more to store a small number of images in it. However, for larger amounts of data, this could require upgrading to a more expensive database plan. Also, databases are typically not optimized for storing large binary files, which can lead to performance issues.
Content Delivery Network (CDN) services (Cloudinary, Imgix): These services often have a free tier for a limited amount of storage and data transfer, but costs can increase quickly with heavier usage. They also offer additional features, like image resizing and optimization, which can justify the cost.
Volume Storage (Render SSD-based persistent disk): Render charges $0.10 per GB per month for SSD-backed persistent disks. If you're storing a large amount of images, this could be a more cost-effective option compared to others.
In summary, for a small blog with moderate image usage, a cloud storage service like Amazon S3 or Google Cloud Storage is likely to be the most cost-effective option. However, if you expect your blog to grow and image usage to increase significantly over time, you might want to consider a CDN service or Render's SSD-based persistent disk, which offers better scalability and performance.
Me: Letās rule out database storage. What would the simplest option be, given that Iām already using Render?
ChatGPT: If you're already using Render, the simplest option would likely be to use Render's SSD-based persistent disk storage. It's directly integrated with the rest of Render's services, so you wouldn't need to set up any additional accounts or learn to use a new platform. Here's a basic outline of the steps you'd need to take:
Create a new disk in your Render dashboard. You can specify the size of the disk (in GB) when you create it.
Attach the disk to your service. You'll need to specify a mount path for the disk, which is the directory in your service where the contents of the disk will be accessible. For example, you might choose to mount the disk at /app/static/img to match your current image directory.
Update your application to save uploaded images to the correct directory. If you're already saving images to app/static/img, and you chose to mount the disk at /app/static/img, you might not need to make any changes.
Update your image serving logic to serve images from the correct directory. Again, if you're already serving images from app/static/img and you mounted the disk at /app/static/img, you might not need to make any changes.
With this setup, uploaded images would be saved to the persistent disk, and they would survive redeployments of your service. Note that the disk is tied to the specific service it's attached to; if you delete the service, the disk and all of its contents will be deleted as well.
Me: okay, Iāve created a new disc on render mounted at app/static/img.
below is my image uploading logic from my routes.py file and my config.py files. do you see any changes I need to make?
image logic from routes.py:
def images():
form = UploadImageForm()
if request.method == āPOSTā:
if āimageā not in request.files:
flash(āNo file partā, ādangerā)
return redirect(request.url)
file = request.files[āimageā]
if file.filename == āā:
flash(āNo selected fileā, ādangerā)
return redirect(request.url)
if file and allowed_file(file.filename):
filename = secure_filename(file.filename)
img = Image.open(file)
img.save(os.path.join(app.config[āUPLOADED_IMAGES_DESTā], filename))
flash(āImage uploaded successfullyā, āsuccessā)
return redirect(url_for(āimagesā))
# Get a list of files in the directory
image_path = app.config[āUPLOADED_IMAGES_DESTā]
all_files = os.listdir(image_path)
# Filter out any directories and create image info dictionaries
image_files = [f for f in all_files if os.path.isfile(os.path.join(image_path, f)) and f != ā.gitkeepā]
images = [{āfilenameā: f, āsizeā: os.stat(os.path.join(image_path, f)).st_size} for f in image_files]
return render_template(āimages.htmlā, images=images, form=form)
here is the config.py file:
import os
from dotenv import load_dotenv
load_dotenv()
BASE_DIR = os.path.abspath(os.path.dirname(file))
class Config:
SECRET_KEY = os.environ.get(āSECRET_KEYā)
SQLALCHEMY_TRACK_MODIFICATIONS = False
LOCAL_DB_PASS = os.getenv(āLOCAL_DB_PASSā)
PROD_DB_PASS = os.getenv(āPROD_DB_PASSā)
class LocalConfig(Config):
DB_NAME = ādbā
DB_USER = āuserā
DB_PASS = Config.LOCAL_DB_PASS
DB_PORT = 5432
DB_HOST = ālocalhostā
SQLALCHEMY_DATABASE_URI = fāpostgresql+psycopg2://{DB_USER}:{DB_PASS}@{DB_HOST}:{DB_PORT}/jdillaxyz_localā
UPLOADED_IMAGES_DEST = os.path.join(os.path.dirname(os.path.abspath(file)), āstaticā, āimgā)
ALLOWED_EXTENSIONS = {āpngā, ājpgā, ājpegā, āgifā}
class ProductionConfig(Config):
DB_NAME = ādbā
DB_USER = āuserā
DB_PASS = Config.PROD_DB_PASS
DB_PORT = 5432
DB_HOST = āsample-url.render.comā
SQLALCHEMY_DATABASE_URI = fāpostgresql+psycopg2://{DB_USER}:{DB_PASS}@{DB_HOST}:{DB_PORT}/{DB_NAME}ā
UPLOADED_IMAGES_DEST = os.path.join(os.path.dirname(os.path.abspath(file)), āstaticā, āimgā)
ALLOWED_EXTENSIONS = {āpngā, ājpgā, ājpegā, āgifā}
ChatGPT: The image uploading logic and your config file look correct assuming the mount path of the disk you created on Render is at app/static/img. This path should be the same as UPLOADED_IMAGES_DEST in your ProductionConfig class.
Your application is saving the uploaded images in the app/static/img directory and also serving images from the same directory. So, if the persistent disk on Render is correctly mounted at app/static/img, there should be no issues and your images should persist across deployments.
Just remember to commit and push any code changes to your GitHub repository so that the latest version of your application is deployed on Render.
If you still face any issues, then the problem might be related to some other part of your code or it could be a deployment issue on Render. In that case, you might want to check the Render logs for more details on the error.
Me: hmm, okay when I test in production images arenāt saving. Iām wondering if this is related to my file structure.
hereās what my directory structure looks like:
jdillaxyz/
-app/
|- static
|- img
in render, Iāve made the mount path /app/static/img
is that correct?
ChatGPT: If the root directory of your project (where your application starts running from) is jdillaxyz, then your mount path in Render should be app/static/img, without the leading slash.
The leading slash /app/static/img would point to a directory at the root of your file system, which is not where your application is.
Change your mount path in Render to app/static/img, and then check if your images are being saved correctly. If they still arenāt, it may be worth checking the application logs to see if any errors are being thrown when you attempt to save an image.
Me: When I try to use app/static/img I get an error message from Render: āMust be an absolute, non-root path.ā
What should I do instead?
ChatGPT: I apologize for the confusion, youāre correct. The mount path in Render should be an absolute path, so it should start with a slash.
The path you mentioned (/app/static/img) should be correct if the app directory is at the root level of your project.
However, this will mount the disk to the absolute path /app/static/img within the container that Render creates. If your application starts in a directory other than the root directory of the container, you will need to adjust the path accordingly.
If your application starts in the /jdillaxyz directory within the container, the mount path would need to be /jdillaxyz/app/static/img. The exact mount path will depend on the structure of the Docker container that Render is creating.
One way to verify this is by checking the logs of the running service in Render to see what the current working directory is when your application starts. You could add a log line in your application to print out the current working directory. In Python, you can do this with:
/import os print(os.getcwd())
This should help you determine the correct absolute path for the mounted disk.
Me: I pretty certain it starts from the root directory because this is where my run.py file is. Iāll try /jdillaxyz/app/static/img and report back.
ChatGPT: Sounds like a good plan! Sometimes, the exact structure can vary depending on how the Docker container is configured, even if the run.py file is in the root directory of your project. If /jdillaxyz/app/static/img doesnāt work, you might want to try adding that log line to print out the current working directory, just to be sure.
Feel free to update here if you encounter any issues or if you have further questions. Iām here to help!
Me: hmm, okay, that didnāt work. where should I put the print statement? in my__init__.py file?
ChatGPT: Yes, you can put the print statement in your __init__.py file. Specifically, you can add it right at the beginning of the file, before the imports.
Hereās an example:
import os print(āCurrent working directory: ā, os.getcwd()) # your imports start here
from flask import Flask
from .config import LocalConfig, ProductionConfig ā¦
This will print out the current working directory when your application starts. This should help you understand the root directory from which your application is running in the production environment, and thus help you set up the correct path for your disk mount.
Remember to check the application logs in Render to see the printed output.
After youāve confirmed the working directory, donāt forget to remove the print statement.
Me: wouldnāt this always be the directory the __init__.py file is in?
ChatGPT: The os.getcwd() function returns the current working directory, which is the directory from which your Python script is executed, not necessarily the directory where the __init__.py file is located.
In a local development environment, this will often be the directory where you run your python or flask run command. However, in a production environment like Render, it may be set differently according to the platformās configuration.
By printing out the current working directory in the production environment, you can determine the correct absolute path to use for your disk mount.
Me: okay, the print file gave me: /opt/render/project/src
ChatGPT: That means the root directory for your application in Render's environment is /opt/render/project/src. Now, considering your project structure, where app/static/img is a directory within your project, you should be able to use /opt/render/project/src/app/static/img as the mount path for your disk in Render's settings.
After you make this change, adjust your UPLOADED_IMAGES_DEST in ProductionConfig accordingly:
python
UPLOADED_IMAGES_DEST = os.path.join('/opt/render/project/src/app', 'static', 'img')
Once these changes are made, redeploy your application and try uploading an image again. It should now be stored persistently on the disk you attached.
From there everything worked as expected. Feature done!
## 100% of WW2 civilian deaths on the US mainland were from balloon attacksSix civilians died on the US mainland during World War II, all from a balloon bomb launched in Japan that landed in Eastern Oregon. These were the only civilian deaths of the war on the US mainland
Via ChinaTalk: Balloon Talk Emergency Edition.
## India fact of the day## Swiss cheeseSince 2012, @timesofindia could have run a daily headline:
— Hannah Ritchie (@_HannahRitchie) February 5, 2023
"130,000 Indians gained access to electricity today".
Every *day*. https://t.co/kOlkl9zlVZ
American cheese was invented in Switzerland by two food chemists, Walter Gerber and Fritz Stettler. Apparently it's made from emmental with sodium citrate added to extend the shelf life.
Discovered via the Kroger App (not joking! they have crazy interesting food facts on their loading screen) and confirmed via Thrillist.
## The other Winston ChurchillThis is the clearest evidence yet that weāre living in a simulation. I learned recently that there was an American Winston Churchill, who was also a politician.
The American Winston Churchill won election to and served in the New Hampshire House of Representatives in 1903 and 1905 and lost the Republican primary for governor in 1906.
Even more amazing, the two Winstons met in 1900, where the British Winston said to the American Winston: āI mean to be Prime Minister of England, it would be a great lark if you were President of the United States at the same time.ā
For this discovery, I thank Robert Cottrell, founding editor of The Browser, who included it in one of his recent emails to subscribers.
## About timeA beautiful essay about the nature of time by Tom Vanderbilt. My favorite excerpt:
"For now, such clocks can at least confirm the old physics. āIf you were to lift this clock up a centimeter of elevation,ā Hume told me, āyou would be able to discern a difference in the ticking rate.ā The reason is Einsteinās theory of relativity: Time differs depending on where you are experiencing it. When standing upright, your head exists at a slightly different timescale than your feet. But with this accuracy comes a sobering reality. āThere are not just two times,ā notes the physicist Carlo Rovelli in The Order of Time. āTimes are legion: a different one for every point in space.ā What we think of the present, he writes, ādoes not extend throughout the universe.ā Rather, āit is like a bubble around us.ā The āwell-defined now,ā as he calls it, āis an illusion.ā"
Found via The Browser.
## Military dolphinsIn addition to dogs, the US armed forces includes dolphins and seals employed by the US Navy.
Discovered via Joseph Noel Walkerās conversation with Palmer Lucky
Update: More on military dolphins and their role protecting the nuclear stockpile via Marginal Revolution.
Things I learned today, via Our World in Data: sex ratios at birth are slightly biased towards boy babies for biological reasons. The ānaturalā sex ratio is about 105 boys born for every 100 girls.
Why is this? Miscarriage rates are slightly higher for female babies than for male babies, although this varies by phase of the pregnancy.
So the next time youāre guessing a babyās sex, the odds are in your favor if you guess itās a boy!
Trends for the next decade For a lot of reasons, as I survey the landscape of technology, I wonder about the trends and opportunities that will shape the next decade. I remember clearly seeing the iPhone in 2007 and being aware that mobile devices were going to be a big deal for a long time.
So what trends rise to this level now? I see four that I would bet on (although Iāve stopped short of crafting falsifiable bets for themā¦ maybe later).
Electrification of everything. This trend has two sides: supply of electricity and consumption of electricity. Global warming, of course, is the macro driver of this trend, but electric cars are the most important thing driving it right now. On the supply side, they create demand that pulls in new technology for supplying electricity and scale. On the consumption side, they are driving electrification technologies (batteries, but also others) that will be reused for other use cases. We are on a path towards carbon emitting fuels being antiquated / confined to niche use cases and electric being the default power source for everything. This is the trend Iām most intellectually confident of.
Generative AI / Large Language Models. As someone who has now become a daily user of ChatGPT, Bing, and GPT Playground, Iām feel most strongly that this is going to be impactful. I see in my own work places where Iām able to get certain tasks done in a fraction of the amount of time that it wouldāve taken previously. And Iām willing to bet that every consumer app from this point forward is going to be built with Generative AI / LLM capabilities in mind, in the same way that apps built after 2007 began to assume mobile phone capabilities. This is the trend I believe in the most (emotional confidence?)
Distributed work. Iām choosing my words carefully here and specifically not saying remote work because it seems clear to me that people and companies donāt really want to be entirely remote. The end of the pandemic makes it seem like this is over, but in reality it has just begun because the equilibrium has shifted away from the pre-pandemic norm of 5 days in the office, but a new equilibrium hasnāt fully formed yet [0]. The more time goes on, the weirder this is going to get, not just for companies, who have to figure out how to offer jobs / benefits / compliance ~globally or nationally, but cities and regions that are going to see patterns of living change substantially.
Decarbonization / carbon removal technologies. This is the trend Iām least confident of, however it is obvious that the only way weāre going to meet net zero carbon on the time horizon that we need to is by getting good at pulling carbon out of the atmosphere (and avoiding using it in the first place). Iām distinguishing this from the electrification like this: switching from gas heat to a heat pump is electrification; switching from a coal power plant to a nuclear power plant or coming up with clever ways to remove carbon from the atmosphere is decarbonization. Itās possible that the energy in this space dissipates, but I think that as we move forward in time it will become ever clearer that having a capacity for pulling carbon out of the area is a necessity and the cost of doing this will put pressure on companies to stop putting it there in the first place.
I see potential in each of these but couldn't quite bring myself to commit to them.
[0]: It seems like the emerging equilibrium is something like this: 1 day a week for teams that are more or less local to an area, 3-4 times a year for teams that are distributed across regions; but how much longer will there be a critical mass of people within regions where it is feasible to do a day a week on the office?
As a share of the occupation, there twice as many female fighter pilots as there are male kindergarten teachers.
This is from Richard Reeves on the Ezra Klein Show
Not new but new to me; interesting never less. From Smithsonian Mag:
Archaeologists dug the beads up in the late 20th and early 21st centuries. Now, a new study published in the journal American Antiquity asserts that the glass objects are among the oldest European-made items ever discovered in North America.
Per the paper, Michael Kunz of the University of Alaska Museum of the North and Robin Mills of the Bureau of Land Management studied ten glass beads found at three sites along Alaskaās Brooks Range. The researchers used mass spectrometry carbon-dating to analyze trace amounts of twine discovered alongside three of the beads and date the artifactsā creation to between roughly 1397 and 1488.
The beads provide evidence for overland trade networks reaching Alaska from the East (Siberia).
A book about volition, equal significance, and excellence. Easily my favorite book Iāve read recently, and a shoe-in for my top books list.
The concept of the book is that the goddess Athene attempts, as an experiment, to create Platoās Just City according to Platoās republic. She and Apollo choose to participate in the city. The book is about what happens next.
The internet describes it as a science fiction book, which isnāt wrong, but also isnāt how I would introduce it because I think it over emphasizes specific aspects of the book at the expense of others. Sort of like saying Hamlet is a play about the Danish Monarchy. Itās not wrong but itās also not exactly right.
I wonāt spoil the plot any further, but I will say that I have fallen deeply in love with several of the characters in this book (agape, not eros).
This book gave me the sensation all my favorite books have given me: a relentless need to finish them and a panging sense of loss, knowing that I will never be able to read them for the first time again. For me, if reading is about anything, itās about finding a book that makes me feel this way again.
Iāll leave you with a quote from the book that reminds me of what I want to remember most from it:
On my temple in Delphi there are two words written: Know Thyself. Itās good advice. Know yourself. You are worth knowing. Examine your life. The unexamined life is not worth living. Be aware that other people have equal significance. Give them space to make their own choices, and let their choices count as you want them to let your choices count. Remember that excellence has no stopping point and keep pursuing it. Make art that can last and that says something nobody else can say. Live the best life you can and become the best self you can. You cannot know which of your actions is the lever that will move worlds. Not even Necessity knows all ends. Know yourself.
Special thanks to Uri for introducing this book to me. I loved this book!
My column for State Affairs about why Georgia should implement open primary, ranked choice voting.
The current system, where any election goes to a runoff when a candidate doesnāt clear 50% of the vote, is a burden on voters, the State, and the candidates for office.
The state legislature should act now to fix this by implementing open primary, ranked choice voting statewide for a fairer, more responsive, and more efficient electoral system.
This might sound complicated at first, but really itās quite simple. An open primary means all candidates, regardless of party, compete on equal footing to be included in the general election. Ranked choice means that in the general election, voters rank their candidates in order of preference. Instant runoff means that if a winner doesnāt emerge in the first count, then a runoff is held instantly to determine the winner, rather than holding an additional election.
Structural changes like this are an underway to improve the quality of our democracy. I do not mean this in the sense of āput the finger on the scales for a particular party I like right nowā but in the sense of āmake sure that the parties we have 10 years from now are the best version they can be.ā
Iām late to link to this, but here are my Book Thoughts on Liftoff by Eric Berger, which is about the early years of SpaceX:
I picked up this book to better understand how big leaps forward in hard tech innovation happen, but as I got to the end, including the annecdote above, all I could think about is whether or not history is contingent.
I didnāt put this in the Book Thoughts post because it somehow felt out of place, but here are the factors that seemed most essential to SpaceXās success (not necessarily in order):
Have a big mission. This helps to pull in talent in a way that smaller projects donāt
Be as iterative as possible in taking on technical challenges. Your speed of failure is inversely correlated to your speed of learning. In a way, the culture of tight deadlines at SpaceX helped with this because it forced the organization to learn from reality rather than debate.
Either buy things off the shelf or make them yourself. If you try and find a middle ground between off the shelf and fully customized, you risk getting neither of the benefits.
Early customers are in many ways investors. They need to be treated as such, since theyāre likely to not get the full benefit of your company existing.
## The loneliness epidemicThis is an amazing table of the cost overruns of mega projects
— Philip Oldfield (@SustainableTall) January 28, 2023
Nuclear storage has a mean cost overrun of 238%, while with nuclear power its 120%
Solar power has a mean cost overrun of 1%https://t.co/1zsdkTWfgZ pic.twitter.com/ViOrj0dpnB
One third of Gen Z and Millenials consider themselves lonely or extremely lonely. For men, this figure rises to 41%.
This comes via Dan Frommer at the New Consumer. I think this is an underrated problem that our society doesnāt really prepare people for. We talk about having good eating or financial habits, but almost never about having good social habits.
Somewhat related: my thoughts from reading Bowling Alone.
Iāve never been into truffle infused flavor and I suddenly feel very justified in this.
There are no white and black winter truffles out of season. They cannot be frozen, cooked, sterilized, packaged, or stored. If you had a dish with an intense truffle flavor out of season - it was most definitely a scam as it is impossible to serve those truffles out of season.
This is from Matt Babich at Taste Atlas. The ingredient used to make most truffle flavoring is a peturoleum derivative. See also cinnamon.
Since 1890, there have been 27 documented humans killed by mountain lions in North America. Cows, by comparison, kill about 20 people each year.
From Craig Childs at The Last Word on Nothing via The Browser.
Coding - went sideways, invented a library and funcitons within that library that seemed plausible.. but werenāt real
Writing an update email to Stripe Apps developers, which I then proof read. It was fine! Saved me probably 10 minutes.
Taking a complicated sql query a data scientist had written last year and rewriting it so I could get the list of one particular set users that made up one of the sums; it was able to this for me while I was in a meeting. Saved me ~20 minutes (more likely, I never wouldāve made the time to do this task).
Asking it for feedback on PRDs - didnāt really get much here
Having it make a company wide email more pithy (accepted some of the suggestions)
Making 1 email to individuals in a group into 7 unique emails with the same message
Learning German
A recipe for Thai Curry in a crockpot
Or at least make sure there is really good ventilation.
According to a study in the International Journal of Environmental Research and Public Health, āalmost 13% of current childhood asthma in the US is attributable to gas stove use.ā
One of the ideas that I think is underrated is the impact of clean air and clean water. It seems like every study I see points in the same direction.
75% of people aged 17-25 suffer from eco-anxiety, according to Mary Sherman. Found via Sean Arians.
The cinnamon we eat in the US today is a cheap substitute of the one that was popular in the US prior to the Vietnam war. From USA Today:
In the 1950s, most of the cinnamon Americans consumed was the Saigon variety from Vietnam. Saigon cinnamon ā the peeled and ground inner bark of an evergreen tree native to mainland Southeast Asia ā has a rich and slightly spicy flavor thanks to high levels of essential oils and a flavonoid called cinnemaldehyde. When the U.S. government imposed a trade embargo on Vietnam beginning in 1964, Saigon cinnamon became almost impossible to import, and spice sellers were forced to find another way to fill American cupboards.
They chose and Indonesian variety, Korintje, that is more bitter and has less depth. Even though the trade embargo has ended, the store-brand cinnamon we use has not recovered.
Commercially available ground cinnamon ā almost always the Indonesian Korintje variety ā is often mixed with fillers. Scientists at the Indian Institute of Spices Research used a process called DNA barcoding to test market samples. They found that 70% contained powdered beechnut husk, ground hazelnut or almond shell dust, dyed and aromatized using cinnamaldehyde and marketed as cinnamon.
Found via Hayden Higgins.
New to me, although maybe it shouldnāt be. Old Christmas is a relic of the Julian calendar, celebrated in rural parts of West Virginia until relatively recently on January 6th. From West Virginia Public Broadcasting:
In the late 1500s, Pope Gregory XIII changed the calendar to match the solar cycle more closely. To do so, the Julian Calendar had to be reduced from 376 to 365 days, eliminating 11 full days. Some countries, though, resisted the change and kept the old Julian Calendar. It took nearly 200 years for England and Scotland to come around. Both countries adopted the Gregorian Calendar in 1752.
About this time, many of these English and Scots were emigrating to the Americas and settling in Appalachia. Some didnāt know about the change or refused to adopt the new Gregorian Calendar and kept the extra 11 days in their calendars. This meant that for them, Christmas fell on January 6 rather than December 25.
https://twitter.com/ag_guy04/status/1611023399509266432
From Why Not Mars by Maciej CegÅowski
"At this point, it is hard to not find life on Earth. Microbes have been discovered living in cloud tops[28], inside nuclear reactor cores[29], and in aerosols high in the stratosphere[30]. Bacteria not only stay viable for years on the space station hull, but sometimes do better out there[31] than inside the spacecraft. Environments long thought to be sterile, like anoxic brines at the bottom of the Mediterranean sea[32], are in fact as rich in microbial life as a gas station hot dog. Even microbes trapped for millions of years in salt crystals[33] or Antarctic ice[34] have shown they can wake up and get back to metabolizing[35] without so much as a cup of coffee."
As Ceglowski points out, this should make us pretty confident that microbial life already exists on Mars, if only from a stray asteroid.
As of 2020, the average undergraduate GPA at Harvard is 2020. From Aden Barton in The Crimson via Orin Kerr.
āIf grade inflation continues at its 2021-2022 pace, for example, the average GPA of the Class of 2028 would be over 4.0.ā
When I saw that David Epstein (whoās writing I love) was writing about Frances Hesselbein again, I sort of rolled my eyes. She is at the center of his book Range (which I love ā probably the book that impacted me most in the past 5 years).
But his remembrance of her is too good not to share. My favorite nuggets:
Her life philosophy as: ādoing whatās needed at the time.ā
She repeatedly declines offers to move up through the ranks of the Girl Scouts, ultimately becoming the CEO and turning the organization around; reading between the lines, it seems like in some ways declining advancement gives her more space to maneuver as she isnāt invested in protecting her reputation.
The ending quote: āLeadership is a matter of how to be, not how to do.ā
Long but so worth the read. The story of how a band of volunteers who became the go to data source for vaccine availability in the US, even being relied upon by Google Maps as their go-to data source, saving lives in the process.
Some things I thought and felt while reading it:
A deep sense of admiration and inspiration. This is American ingenuity at its finest. Solving the problem in front of you however you can, never being satisfied. A lot of the best aspects of the Stripe culture are also on display.
Sadness or a sense of missing out. Why didnāt I know this existed? Why didnāt I help out? I told my wife, in what was perhaps an over reaction, that I felt like I had missed my chance to do something that matters by not participating in this.
I think this does a good job of making the case that in practice simplicity and equity go hand-in-hand even if the additional complexity is meant to increase equity. Another way to say this is that if your goal is equity, you need to have a high bar for additional complexity.
Such a startlingly good example of ādo the simple thing first.ā Before you build a real time system for national vaccine tracking, call pharmacies and make a list.
I have some skepticism towards the claim that pharmacy websites were ābroken by design.ā I wouldāve liked to have seen more there.
I agree with this wholeheartedly: āWe as a society accepted so much mediocrity during the pandemic, and we do to this day.ā
Sesame will soon be required to be labeled as an allergen on food, which is leading to more sesame being used in food.
Food industry experts said the requirements are so stringent that many manufacturers, especially bakers, find it simpler and less expensive to add sesame to a product ā and to label it ā than to try to keep it away from other foods or equipment with sesame.
As a result, several companies ā including national restaurant chains like Olive Garden, Wendyās and Chick-fil-A and bread makers that stock grocery shelves and serve schools ā are adding sesame to products that didnāt have it before. While the practice is legal, consumers and advocates say it violates the spirit of the law aimed at making foods safer for people with allergies.
āIt was really exciting as a policy advocate and a mom to get these labels,ā said Naomi Seiler, a consultant with the Asthma and Allergy Foundation of America whose 9-year-old daughter, Zoe, is allergic to sesame. āInstead, companies are intentionally adding the allergen to food.ā
Via Alec Stapp
is ātimeā according to Dr. Dean Buonomano on the Ezra Klein podcast.
Relatedly, Wikipediaās list of sundial mottos will leave you with something to think about.
Das Lichterschwimmen via 8008-mb.ch
A wonderful ZĆ¼rich tradition that I just learned about this year: the Lichterschwimmen where students float lanterns down the Limmat through the center of the city. Some how we missed this while we lived there! Iāll have to go back and see it.
## Parenting culture in the United StatesJust one of the 270 jobs in the 1950 census has been eliminated by automation... elevator operator.
— Ethan Mollick (@emollick) December 20, 2022
Other jobs that were expected to be automated by tech, like bank tellers by ATMs, just shifted the nature of the job. Hopefully, AI follows this pattern. https://t.co/F7dGbApCyH pic.twitter.com/Ck5b0ae3eO
The average college educated mother in the United States spends ~2 hours a day with her children. Amazingly, this is a 100% increase from 1965.
From the Economist via Sarah Constantin
## Before peer reviewOnly one of Einsteinās papers was ever peer-reviewed, by the way, and he was so surprised and upset that he published his paper in a different journal instead.)
Via Adam Mastroianni, who makes a compelling case against peer review.
It does seem that writing publicly and emphasizing auditability (sharing of data and code) would be more important than anything else.
We reached an important milestone for Stripe Apps last week: the marketplace is now open for all developers!
Expanding the ability to publish apps into the Marketplace has been a top focus for the team since the summer. Iām particularly passionate about this launch because I think the promise of Stripe Apps is best realized in the niches ā developers making products for merchants with specific needs that will only be possible if anyone in the world is allowed to bring their creativity to bear on the problem.
Iām excited to see where this one goes!
## Congratulations, Argentina!Congratulations to Argentina and Lionel Messi on winning the World Cup. Thrilling game.
I saw Messi play in person in 2019 and it remains one of the three most impressive in person sporting performances Iāve ever seen. What Iāll remember most about watching the game, besides the atmosphere was how different Messi was in person.
He sat the first half of the game and then came in for the second half. Everyone else on the field was a football player; he was a great white shark. He scored the gameās only goal and was so business like about it. It felt like he couldāve scored it whenever he wanted and chose to do it then and not to do it again so as not to show up his opponents. Iām glad that guy has won a World Cup.
Macro Oceans has opened up a pilot manufacturing facility in West Sacramento. Congratulations to the team on the milestone. Coverage by the Sacramento Biz Journal is here.
India is going surpass China as the most populous country in the world next year, according to Pew Research. Chinaās population is expected to decline by 46% by 2100 according to the UNās middle of the road projection.
The population cliff ahead of China is astounding. And a fact that has been true all my life (China is the most populous country in the world) isnāt true any longer.
In the United States, 39% of adults say they believe āwe are living in the end times,ā while 58% say they do not believe we are living in the end times, according to a recent Pew Research Center survey.
Christians are divided on this question, with 47% saying we are living in the end times, including majorities in the historically Black (76%) and evangelical (63%) Protestant traditions.
From Pew Research. I wonder what the trend line is like here.
In the past five years, according to the Federal Trade Commission (FTC), āpeople have reported losing a staggering $1.3 billion to romance scams, more than any other FTC fraud category.ā In 2021 alone, that number was $547 million, which was a nearly 80 percent increase compared to the previous year.
That is from RF Jurjevics at Vice writing about romance scams targeting women and the men whoās social media photos are used by the scammers. Found via The Browser.
From Ezra Kleinās show with Maryanne Wolf about the childrenās show Cocomelon
[They] have set up a room, the place that makes Cocomelon, where they will have a kid watching the show. And set up next to it is another screen that shows an adult just doing normal household tasks, just sort of wandering around doing whatever you do in the house. And if the child becomes distracted from Cocomelon by what the adult is doing, they go back to the edit and they amp up the interestingness, the cuts, the whatever makes a Cocomelon episode interesting.
As a product manager, I admire the understanding of the use case and the commitment to execution, but as a parentā¦
Triple Falls, Dupont State Park, North Carolina, part of the Appalachian mountains and the most beautiful place I went in 2022.
I borrowed this concept from Tom Whitwell as a way of cultivating a habit of curiosity. You can read his 2022 version here.
My 2022 in a nutshell: I had a son, moved back to Georgia, launched Stripe Apps, and helped the team at Macro Oceans build the kelp economy of the future.
Here are 52 things I learned along the way, with some quotes and pictures that inspired me mixed in:
“Just because nobody complains doesnāt mean all parachutes are perfect.”
The Upper Nepean by WC Piguenit
“A tiger doesnāt proclaim his tigritude, he pounces.”
Poppies by Robert William Vonnoh
This year, I began tracking for the first time the people and websites that pointed me to the place where I found the things I learned. The two most common where The Browser and Marginal Revolution. If these aren't already a part of your regular reading, you should consider them. The third, surprisingly, was Twitter. I hope it continues to survive in 2023!
My list for 2021 is here.
If you think weād have an interesting conversation, please reach out; Iād love to chat. You can send me a note at jdilla.xyz@gmail.com.
1 in 3 owned homes in the US belongs to someone 65 or older. Pointed out by Tony Fratto from a recent episode of Odd Lots.
This week Greeceās electric grid ran entirely on renewable energy. And it looks like per person CO2 emissions have probably peaked.
However at the same time, the Alaska snow crab population has collapsed (although fishing methods may also have played a role).
My mental model for the climate change story is that for the foreseeable future, thereās going to be significant progress on renewable energy, decarbonization, and the like. Itās actually going to be shockingly fast. Eventually carbon removal is going to be figured out and costs will fall pretty quickly. A critical mass of people have accepted that decarbonization is a problem that has to be solved and it is technically feasible to solve it (with some innovation required along the way).
Alongside this though, a fair amount of environmental cost going to be paid and people and communities are going to suffer because of it. To a large extent, these costs are no longer preventable - the time for that was 10 years ago.
Iām not advocating for this path, but think itās the likeliest outcome and if you suffer from climate anxiety, it will be crucial to remember that both those things things are happening at the same time.
Bowling Alone, by Robert Putnam, is a classic book about trust and society. As a part of my obsession with trust, I felt like I had to read it. The book, which was published in 2000, is about the decline of social trust in the United States. I didnāt find it to be dated ā if anything, I think the additional distance from the publish date helped the book. From the vantage point of 2022, I felt like the book previewed some of the challenges that lay ahead of American society when the book was written and was detailed enough to help me hypothesize about how trends since 2000 may or may not have continued since then. Itās a classic for a reason!
The main idea of the book is that social trust ā the invisible quality that makes us feel a bond with our fellow human ā has been decaying in the United States. This much I (and I suspect many others) knew about the book even without reading it because it gets cited a lot.
This is a big deal because social trust makes it possible for society to function efficiently. Business and government work better in high trust societies. Additionally, according to Putnam, trust and participation seems to be closely linked to personal health and happiness.1
So what leads people to have trust in their community and society? The book isnāt as explicit here as I wish it would be, but far as I can tell, though, the critical factors are:
Having relationships with people in the community; spending informal time together, having both close friends and friendly faces in the crowd; in everyday life, this looks like things like game nights, card games, social dinners, church attendance, and yes, bowling leagues
Feeling agency over how the community is organized; in particular, joining in a group activity with other people to solve a community problem seems to be a particularly powerful contributor to trust; in everyday life, this looks like things like club membership, running for office, serving in a volunteer group
Trust in the United States has been falling pretty steadily since the late 1960s.
Data from Bowling Alone, thanks to Engaging Citizens and Building Social Capital: The Exceptional Civic Story of Portland Oregon and the Role of Information Technology. Steve Johnson, Ph.D for the visualization.
Putnam makes a convincing case that this is not just people changing the way they answer a survey question, but a general decline in social capital. Across almost every dimension he studies ā church membership, bowling leagues, running for office, formal involvement in volunteer groups (like the Lions club), sports leagues, volunteering at local charities, dinner parties, you name it ā Americans are less involved and less trusting by the end of the 1990s than they were in 1960s.
Moreover, this pattern holds across different groups. People with more education are more likely to be involved and trusting than people with less education, but both cohorts are less involved and less trusting than they were 30 years ago. Same with race, gender, and income. It really is striking how many different cuts of data tell basically the same story ā a decline in trust and participation starting in the late 1960s.
I want to pause on this for a minute, because I think itās an underrated point. America in 1998 was not producing social capital at the same rate it was in the 1960. While America in 1960 had many flaws, black people and women of the 1960 reported higher levels of social trust than did their counterparts of the 1998. This isnāt to say that we should seek to go back to the way society was in 1960, but it is worth understanding ā after all, wouldnāt we expect a more equal society to have higher levels of social trust? Putnam suggests that it was the social cohesion of the 1960s that allowed America to make the steps it made towards racial and gender equality ā the implication being that lower levels of trust are causing us to miss out on further progress.2
The one exception to this trend of declining participation are social activities that can be done alone. So as an example, people donāt join a local volunteer group, they write a check to an issue based organization; they donāt join a bible study, they are spiritual at home. Individual activities, however, donāt create trust.
So where is all this trust going? The book makes the case that there are four factors worth considering:
Generational replacement. The generation that lived through World War II seems to have had particularly high social trust that they learned as a part of their formative years. This social trust habit hasnāt been passed down (or hasnāt been activated?). This is by far the largest contributing factor.
Time spent watching TV.3 Or said differently, as our entertainment options at home have gotten better, weāre less likely to venture out into the real world and do the sorts of things that lead to trust. For individuals, the time spent watching TV is the āsingle most consistent predictorā the author discovered.
Commuting. At least during the period when the book was being written, time spent commuting was going up. Almost by default, commuting in a car happens alone and is time that canāt be spent on other things. Additionally, itās not just commuting workers who pay this penalty; in communities with long commute times, even retirees are less involved.4
Financial pressures are causing us to spend more time working and leave us with less time to spend on leisure. Before reading the book, I wouldāve guessed that this would be the dominant story, but it doesnāt appear to be. There does seem to be some pull away from community engagement due to work, but it very much appears to be on the margin.
So where do we go from here?
First, from the vantage point of 2022, I have a hard time not seeing a lot of the trends outlined by Putnam getting worse. Based on what Putnam said about the impact of TV on participation, itās hard to imagine that social media and online gaming have made the situation better. Add to this the pandemic, which broke the habits of engagement for many people, and itās easy to see how people might start to feel like the world is spiraling out of control for them. I would predict this to continue in the near term!
The one silver lining I see is that I feel like the pandemic and shift to remote work has made it significantly easier to create social capital online. In an additional chapter from the 2020 version of the book, Putnam discusses the potential impact of the internet on social capital and posits that it will be good for organizing, but bad for creating relationships that lead to meaningful change. Based on my experience, I think that this was true before 2020, but sometime during the pandemic, it shifted. Since about mid-2020, Iāve seen a significant uptick in the number of professional relationships I have with people Iāve never met in person (even outside of my current company); a handful of these people have become legitimate friends. This feels like something Iāll have to learn how to cultivate through the rest of my career.
Second, reading this book has made me feel more strongly that we should be nudging young adults towards civic service and potentially even have mandatory/highly encouraged civic service programs. Based on the generational replacement chapter, it seems likely to me that oneās habits towards trust and civic participation are set somewhere between the ages of 18-28, so nudging people towards service during this time in their life should pay dividends for years to come.
Third, it became clear to me while reading this book that I needed to have a personal social capital plan in the same way that I have an exercise routine and other personal health habits. Finding both informal ways and formal ways to be a part of the community matter for individual happiness as well as community outcomes and I should be intentional in how I invest in it. Ideally civic leaders would latch onto this message and start to reinforce it within the communities they lead.
Finallyā¦ at some point while reading this book and looking at all of the charts where things go sideways starting between 1968 and 1974, I started to think of this chart of productivity in the United States.
Could there be a relationship between declining trust and declining productivity? At least at an intellectual level, this makes sense: if more trust allows two people or four people to be more productive together, why wouldnāt more trust allow a society to be more productive?
Interestingly, whilte Putnam does discuss the importance of trust to economic productivity and he shows a lot of charts, he never shows this one. Even more interesting, while Iāve seen a lot of musings about the causes of the Great Stagnation, Iāve never seen anyone put forward declining social trust as an explanation; it doesnāt mean itās not out there, but Iām surprised that I havenāt run into it. Iād like to do more reading here!
1: An alternate version of this book is the self help version ā Bowling Alone: Why the key to health and happiness is cultivating friendships and community involvement
2: wonder if this means that some of the nostalgia for an earlier American age is in fact driven by a longing to return to a higher trust society? Another interesting question is whether more social conformity / segregation is needed to create higher levels of social capital; Iām not sure I agree but would love to see data.
3: This is where I hear a voice in the back of my head saying āafter the defeat of Carthage, the Romans became complacent and decadentā¦ā
4: Itās interesting to imagine how more flexible work arrangements could change this.
A great reminder that price is a function of demand and supply:
Another thing is that I donāt know what my inventories are worth. I know that ten years ago I bought floppy disks for eight to 12 cents apiece. If I was buying a container of a million disks, I could probably get them for eight cents, but what are they worth today? In the last ten years theyāve gone from ten cents to one dollar apiece, and now you can sell a 720KB double density disks for two dollars. I just donāt know what the market will do. Itās very hard to run a business when you donāt know what your product is worth.
That is from Eye on Designās article about Tom Persky, the last man selling floppy disks.
The Alaskan Seaweed crop has expanded ~30x since 2017. That is from Macro Oceanās grower survey in partnership with the McKinley Research Group.
I was struck by this excerpt from the most recent edition of Range Widely by David Epstein. Here he is interviewing Brad Stulberg, author of the book The Practice of Groundedness:
DE: Ah. Fair enough. Seems like good advice. And on the advice note, youāve actually reminded me of an unrelated piece of advice you espoused that I took to heart: to get involved with some real live Homo sapiens in my community. I acted on that one by joining the board of a phenomenal early childhood education center focused on poor families in my area. Iāve definitely found it challenging; it has led me to do some event logistics ā not my strong suit. But Iāve also found it uniquely rewarding, often even more so than volunteering Iāve done with much more prominent national nonprofits. Please explain.
BS: Iām so glad you said this! Hereās the deal: at the risk of sounding woo-woo (though decades of psychology research and clinical practice support this) we are looking for love in all the wrong places. When we are intimately involved with other human beings in the real world, working on meaningful projects, having meaningful conversations, and striving toward meaningful goals, we donāt feel the need to go on the internet to look for status, validation, and love there.
Iāve been reading Bowling Alone which in many ways anticipates these themes. As a society, we have retreated from in person, coordinated action, towards activities that can be done individually and in aggregate, weāve gotten less happy.
I wonder if we should be thinking about in person participation the same way we think about exercise ā something thatās required for us to have healthy and fulfilling lives?
Image by Cecil Beaton, Public Domain
Queen Elizabeth has reigned for an incredible 30% of US history. That is from Matt Glassman.
This is chart via Max Roser. You can argue that decarbonization isnāt happening fast enough, but itās definitely happening.
Via Ethan Mollick, who is an excellent twitter follow,
It is our tendency to fixate on the 1st solution we come up with, preventing us from finding better ones
Iāve definitely felt the pull that comes from my first idea on how to solve a problem and how difficult it is to set it aside. Sadly the abstract doesnāt give any strategies for neutralizing the effect.
is closer in time than you think it is. From Scientific American:
In 2004 mathematical modeling and computer simulations by a group of statisticians led by Douglas Rohde, then at the Massachusetts Institute of Technology, indicated that our most recent common ancestor probably lived no earlier than 1400 B.C. and possibly as recently as A.D. 55. In the time of Egyptās Queen Nefertiti, someone from whom we are all descended was likely alive somewhere in the world.
The mechanism here, which is intuitive as soon as you understand it is that the number of branches in your family tree grows exponentially as it goes backwards. But thatās not all:
āBranches of your family tree donāt consistently diverge,ā Rutherford says. Instead āthey begin to loop back into each other.ā As a result, many of your ancestors occupy multiple slots in your family tree. For example, āyour great-great-great-great-great-grandmother might have also been your great-great-great-great-aunt,ā he explains.
Some other surprising estimates from this article:
It is estimated that everyone alive today in South America has at least some European ancestry
It is estimated that ānearly everyone of Jewish ancestry has ancestors who were expelled from Spain beginning in 1492ā
Via Max Roser
By Lionel Royer - MusĆ©e CROZATIER du Puy-en-Velay. ā Public Domain
Inspired by my recent favorite podcast The Rest is History, I finally read the copy of The Gallic Wars that I had lying around. It is Julius Caesarās account of conquering the various peoples of Gaul and bringing their territory firmly under Roman rule.
The book is surprisingly readable and relatable, even two thousand years later. The fact that it was Caesarās own account of his campaigns is really fantastic. Even though you canāt take it all at face value, how often, even in the modern era do we have a first hand account from someone of Caesarās stature?
I love the fact that Caesar wrote these accounts in order to raise his own status in Rome. The image that came to mind for me are the giant status updates that teams send inside of tech companies to keep everyone up-to-date on project status and build momentum. Julius Caesar was literally doing this!!! And he became dictator for life!
Caesar comes across as charismatic. His ability to project confidence and rally his teams despite the odds ahead of them is compelling. You can see why the men wanted to follow him. He both pushed them and related to them. Like a character from The Sopranos or The Wire, you root for him despite the human cost of his actions. He claims to have killed a million Gauls. Historians consider this to be an exaggerationā¦ but still, this is who you find yourself rooting for!
The high point of the book is the Siege of Alesia, where the Romans surround their foe Vercingetorix in Alesia and then put a wall around their siege in order to hold off the force that had been sent to relieve Alesia. Absolutely astounding and not the sort of thing that works out for a lesser general. If youāve never heard of this, take the time to read the wikipedia article. Itās worth it.
Vercingetorix is a tragic figure in his own right. His appeals to liberty and self rule resonate today. He is clearly a talented leader, going toe-to-toe with Caesar. And yet he comes on a little too late and is not quite able to unify the Gallic tribes enough to prevail. Iām surprised that there hasnāt been a novel / movie / Netflix show made from his perspective yet.
Iām not a military historian, but I was struck by how often Caesar created advantages for himself by showing up places where his enemy didnāt expect him to be or by understanding what his enemy was going to do via intelligence. He also does a great job of weighing the cost of inaction ā frequently the risks he takes are due to his assessment that delaying an engagement is an even greater risk.
Despite the readability, I wouldnāt recommend it unless youāre already interested in the topic. I am a total dork for Roman history and I still got lost in the names of tribes, commanders, and local rulers.
āThe new always looks so puny ā so unpromising ā next to the reality of the massive, on-going business.ā ā Peter Drucker in Innovation and Entrepreneurship
No matter how big the ambitions are for something, they always start small. But big companies have big expectations. This incongruence makes starting new things at companies particularly challenging.
Some types of these expectations:
Big companies expect big results. If youāre a start up, a $1B market opportunity looks life changing; if youāre Apple, not so much. This creates a tension where interesting ideas that look small today donāt get funded unless someone is willing to promise big results. The incentive here is to overpromise because if you overpromise, you might get lucky and more or less hit expectations. But if you underpromise, you wonāt get funded and you never even get started. So you overpromise because you have to, but overpromising creates pressure to generate results where the opportunities might not yet exist.
Your teammates have expectations about their jobs. Your ability to define their your role is extremely important within larger companies. If you donāt define what youāre responsible for and what youāre not responsible for, you wonāt get anything worthwhile done. But starting something new frequently requires small amounts of a specific types of skills. You donāt need someone to do full time content marketing or partner management, you just need someone to roll up their sleeves and work on a specific project. So on the internal start up, you run into a lot of people saying āthatās not my jobā or āIām sorry, I donāt have time for this.ā This is a particular challenge for less common functions where the model is to embed someone in the team (vs. project based or on demand staffing), because if you donāt have enough work to justify at least a quarter of a persons time for the foreseeable, you risk not getting anyone at all. This also shows up in small ways where people push back on owning specific projects almost reflexively ā after all, in their last role at the same company this wouldāve been someone elseās job and a major part of their success to date has been their ability to define their role.
Performance assessment. Even if the people youāre working with are willing to do whatever it takes to succeed personally, the institution is going to push back against this. As companies get bigger, they develop career ladders. These ladders define the skills necessary to progress at a given level within a given function and inevitably describe what success looks like in the core business better than whatever new thing youāre working on. This will be a problem both for people on your project, who risk paying a penalty for doing work that isnāt at their level and people considering joining your project and comparing it against the career trajectory offered by other options.
So how do you succeed in spite of this? I canāt claim special expertise, but hereās what Iāve observed to work:
Understand the incentives. If you understand it, you see it coming, and avoid the worst of it. Some of it you can mitigate with planning and some of it you just have to accept. This is the tax you pay in exchange for not having to build something new while chasing funding and setting up IT systems.
Reward effort. The people youāre working with are multifaceted. Most people, particularly those in high demand professions, arenāt optimizing purely for the next promotion or personal prestige. If youāre excited about them going outside of expectations to do whatās best for the team, theyāre more likely to keep doing it, even in the face of incentives to do otherwise.
Create a safe haven for experimentation. Progress on new things isnāt linear and people need the space to mistakes. Itās much easier to do this without needing to create the artifice of consistent results. So make that space where you can. If youāre more junior, this might look like a side project outside of your formal OKRs. As you get more senior, give your people the space to dabble around in areas that have the potential to be productive without demanding that they show results.
Even better, have a plan to create the artifice of consistent results. Teams and leaders that are really cooking are actively experimenting with one part of their portfolio while banking results with another part of their portfolio. By doing this, theyāre able to shield the experiments from the need to produce results right away and can use the lessons theyāve learned to drive the next set of results.
From Bloomberg:
A full 50% of office visits globally were just once a week in the second quarter, up from 44% in the first quarter, according to data from Basking.io, a workplace-occupancy analytics company. At the same time, fewer people made the commute four to five days a week, especially in large cities.
Iām surprised that the norm here is 1 day a week. Right now I assume that most people live close enough to an office to make this work and are basically coming into the office to have all their 1/1s in person.
I think the longer term equilibrium for jobs that can be remote is going to be ~1 week per quarter mandatory in person together with the rest distributed. I do think that some people will choose to go in more often, especially if there is a critical mass of people that live nearby.
Weāll see!
I appreciated this observation from Austin Vernonās post on the potential for Geothermal energy: When it comes to low information environments, just trying stuff is powerful. Only the simplest models are worth using.
He gives this example from fracking and low quality sand:
Models that predict frac (sic) job performance have consistently lost to some engineer saying forget it and upping how much sand and water gets pumped downhole. Besides failing to predict the success of slickwater, models didn't account for the success of low-quality sand. Until recently, the industry used only the most spherical sand at great expense. Theoretically, spherical sand should drastically improve permeability over wonky sand. Sourcing this sand from places like Wisconsin became a problem once oil companies started using trains worth of sand in a single well. Eventually, companies sourced local, low-quality sand at a much lower cost and rarely saw performance decreases. They bought more sand on the same budget and made better wells. Each formation eventually reaches a limit where further intensification doesn't help. The industry always finds that point empirically.
This reminds me of a theme from one of my favorite books, Range: when planning a career, thereās not a great way to know in advance what is going to fit for you, so experiment ā have hobbies, change jobs, take on projects.
Alpfahrt im Sommer, Albert Manser - 1980 via Artnet.com
Itās somewhat popular now to throw shade at Google1, particularly as a place to work: itās big and bureaucratic, itās not a good place to start your career and so on. I understand this impulse, but I disagree with it. Particularly for its size, I think Google is really effective company and rather than bagging on it, people should think about what has allowed it to stay as effective as it has despite being as big as it is.
I think people typically misunderstand Google in two key ways that then causes them to misjudge it:
Google isnāt a start up. It is one of the largest companies in the world, with more than 150,00 employees and about as many contractors. Its peer companies, in terms of number of people, are General Motors, Darden Restaurants, and Aeon, to name a few. Just to try and put this into context, my entire current company (Stripe) is smaller than my product area at Google (YouTube).
Because many people still remember Google as a younger, smaller company, they judge Googleās agility against much younger and smaller companies instead of against its peer set. Certainly itās less agile than it once was, but on a size adjusted basis itās one of the most agile companies in the world.
Google is much more decentralized than a typical company. I joke that the best way to think about Google is a university attached to a money printing machine. Like a university, Google has many different departments that really arenāt trying to coordinate with each other. This is by design. You may not agree that this is the right strategy, but I think you have to understand it to effectively evaluate the company.
So what does Google get right? Here are four things that come immediately to mind for me.
Product focus: an incredible amount of attention is paid at all levels of the company to the specifics of the product ā what it actually does for the user. As a typical YouTube product manager, I regularly had to review product details with head of product, head of engineering, and other very senior leaders. These leaders, several levels up from me, were more well versed in the specifics of my product area than my manager or my head of product was at much smaller companies. This product focus isnāt top down, but cultural, which makes it much more powerful. People at Google encourage other people at Google to use their products and have opinions about how they should work.
Distributed decision making: For me, this is where the misunderstandings about Google begin to really show up. Google is incredibly effective at decision making for a company of its size. An incredible amount of relatively high stakes decisions can be made at Google very quickly. As an example, if my team made a product change that positively impacted our key metrics, I could have that launched globally to billions of people in two 15 minute meetings.
Additionally, Amazon gets a lot of mileage out of the Bezos API mandate, but itās pretty common at Google as well to have some internal help text and an API be all that is needed for different orgs to collaborate. Additionally, as long as team incentives are aligned, itās pretty easy for individual teams in different orgs to collaborate without any formal sign off from their Product Area Leads.
Where things do get complicated at Google is major collaborations across product areas where there is a lot of ambiguity. So for instance, if I wanted to get a couple of teams from Google Maps and a couple of teams from YouTube to collaborate on a set of features with high but uncertain potential, I knew that this was going to be a difficult and challenging path. And yet, as difficult and challenging as this would be, on a per person basis, it was much easier to get collaborations like this to happen than at <1,000 employee companies that Iāve worked at.
Talent friendliness and talent development. The slides and free lunches are really easy to make fun of, but Google legitimately tries to be a good place to work. The default policies are sensible and friendly to employees. When they make a change, they try to make sure to not punish their people along the way. It isnāt perfect, but especially for its size itās really good.
Beyond its policies, Google is a great place to develop your career if youāre thoughtful about it. As long as youāre keeping up with your day-to-day work, you can get exposure to almost any career path as a twenty percenter (once again, think of a university rather than a traditional company). Here the breadth of Googleās product offerings is a real asset; you can get exposure to almost any discipline or industry without changing companies.
Caveats: I worked at Google for about two and a half years, entirely at YouTube and mostly in the ZĆ¼rich office. Google is a big place that varies a lot by team, so itās entirely possible that my experience is an outlier. There are also many legitimate criticisms of the company that Iām not the best person to articulate.
Notes: 1: Alphabet
The podcast Revolutions by Mike Duncan is probably my favorite one these days. It follows the history of different revolutions, starting with the English Revolution in the first season and continuing through to the Russian Revolution in the current and final season.
Episode 100 of the Russian Revolution season is titled History Never Ends and it tells the story of how Lenin and the Communist Leaders, in the aftermath of the October Revolution, continuously sought a ābreathing spellā, when things would calm down and they could implement Communist principles and begin improving the lives of the people. From the show:
This was the logic behind the treaty of Brest-Litovsk: make peace at enormous cost because we need a breathing spell. This was the great prize to be won during the civil war. If we defeat all our enemies, we shall be able to finally work in peace. But this breathing spell they yearned for was a mirage. And it's always a mirage. We all know from our own daily lives, that fabled next week or next month or next year, when we will finally be able to do all the things we have to put off today because we're too busy, too harried and dealing with too many other emergencies, big and small, sudden deadlines that force us to drop everything, unexpected events that just upend our lives, except when we get to that next week and next month and next year we find the same set of unexpected emergencies, often the same type in category that have stalked us throughout our lives. And we are forced back into our natural state of scrambling a reaction and improvising a response.
Whatās true for revolutionaries and history is also true for companies and products. There are only two modes: minor crisis and major crisis. Breathing space almost never materializes. Itās exceedingly rare for a product or company to have sufficiently cleared the field of all rivals such that there are no short term concerns1.
So what are the implications for leaders?
You have to learn to differentiate between the different types of crises. One potential frame here: do you have more momentum or problems? Another: if you look across all the products and companies you know of, how would this season rate? If itās not obviously a time of major crisis, itās probably a time of minor crisis.
You have to make the time and space to work on the long term. The time when it will be easy is never going to come and if you donāt do it today, it wonāt be easier tomorrow.
Expect a background level of chaos and messiness. Learn to perform at a high level in the midst of it.
Notes:
In my observation, companies tend to make some of their worst decisions in their moments of least crisis
An unrelated fact I learned from this episode: the reparations placed on Germany by France at the end of World War I was calculated to be the same amount plus interest that Bismark had placed on France after the Franco-Prussian War. How did Bismark get this number? By calculating an amount identical to what Napoleon imposed on Prussia in 1807.
Shoe Dog is the story of Phil Knight and the founding of Nike through the first 12 years of its existence. Itās a great entrepreneurial story, one that I would recommend to anyone starting a business, because of how honest it is about the journey.
As an outsider, one thing that seems to surprise first time entrepreneurs is how much time they spend solving problems that are both existential and completely unrelated to the companyās core mission. Knight spends the better part of a decade fighting through pushing through challenges like balancing his startup with his day job, finding a bank willing to fund his growth, and personnel issues before he finally gets to the point where he can say: āthe problems were never going to stop, I realized, but for the moment we had more momentum than problems.ā I think this is a good frame: do you have more momentum or problems?1
Another thing I appreciated about this book was getting to know Nike before it was dominant. All my life, Nike has been the premier sportswear brand. I had no idea that it began life as Blue Ribbon. Or that it started out importing another companyās shoes, basically taking advantage of outsourced production. Or that it only launched the Nike brand when its production partner tried to go around it to distribute in the United States.
One final story from the book: the name almost wasnāt Nike. Knightās most trusted lieutenant had observed āthat seemingly all iconic brandsāClorox, Kleenex, Xeroxāhave short names. Two syllables or less. And they always have a strong sound in the name, a letter like āKā or āX,ā that sticks in the mind.ā Despite this insight, he very nearly named the company Dimension Six.
1: A benefit of having an executive team of distance runners: they have a high tolerance for pushing through pain towards a goal.
Often when I find myself stuck, itās because I donāt have conviction about how to move forward. One option gives me part of what I think is important, another option gives me another part, but no option gives me everything I want. I can spend a lot of time churning with indecision, looking for an additional option that doesnāt exist.
Oddly, I find these situations harder to deal with in my personal life than in my professional life. It seems that at work I expect to be faced with choices where there is no ideal outcome, but running into this situation in my personal life is more distressing.
Recently Iāve been struck by the power of laying out the options that exist and āsimplyā choosing the least bad one. In practice, this looks more like:
Lay out the options
Make sure no other options exist
Identify the least bad one
Is there anything I can do to make this option even less bad?
Pick the least bad option
Decide to revisit the decision and/or how to move towards the ideal option in the future
A common pattern I see in the people whoās work I admire is consistently choosing the least bad option1. If you can do this regularly, you can improve a situation a lot over time. And remembering this seems to reduce the mental toll of not having an ideal option
So hereās to the least bad option: humbler and less satisfying than the best option, but powerful nonetheless.
1: I think of this Jon Gruber column about how Apple rolls, which I revisit about once a year.
Iām really excited to finally show the world Stripe Apps.
Stripe apps lets developers customize a userās Stripe experience. They can make workflows simpler, like the Intercom app, that makes it easier for a support agent to see who theyāre refunding a payment for and reply inline. They can make collaboration easier, like the Render app, that makes it easier for a company that uses Stripe to process payments to make that data accessible to PMs or Analysts on the team. Or they can make it easier for businesses to get financing based on their Stripe financial data, like FounderPath, that helps SaaS founder access capital more easily.
This has easily been the most ambitious launch that Iāve ever worked on. Recruiting 35 launch partners seemed impossible when we started and yet here we are with more than 50. Iāll probably write more about what I learned while doing this in the future, but for now Iāll just say that ecosystem product management reminds me of the way I felt when I first started being a platform PM.
Thereās a fair amount of terror, that comes from not having direct control over peopleās roadmaps and decisions. Thereās a sense of accomplishment that comes with understanding how to use a focus on the user and attention to detail to get people to move in the same direction. Thereās a thrill that comes with making something much higher impact than any one team or individual could make on their own.
You can see the entire list of amazing launch apps here or start building an app here. You can find coverage of Stripe Apps in the Stripe Newsroom, Hacker News, Tech Crunch, and Venture Beat.
Update: Stripe Apps covered by Ben Thompson in Stratechery.
The last major thing I worked on at YouTube has finally shipped: a way for Creators to find inspiration for videos using YouTube data on what people have been searching for.
This feature is particularly useful if youāre trying to find your niche on the platform. You can brainstorm topic areas and then use those to find search terms where there isnāt enough good content today, allowing you to get started.
Iām proud to have played a small part in bringing this into the world.
A public water fountain in my old neighborhood in ZĆ¼rich.
From the Conversations with Tyler podcast:
COWEN: If I visit every major country in Europe, what I observe is the highest living standard is arguably in SwitzerlandāāāNorway and Luxembourg aside. Switzerland has one of the smallest governments, and they attempt relatively little redistribution. What is your understanding of Switzerland? What if someone said, āWell, Europe should try to be more like Switzerland. Theyāre doing great.ā Why is that wrong?
PIKETTY: Oh, Switzerland. Itās a very small country, so itās about the size. . . . Actually, itās smaller than Ćle-de-France, which is a Paris region. Now, if you were to make a separate country out of Ćle-de-France, GDP per capita, I think, would actually be higher than Switzerland. Of course, you can take a wealthy region in your country and say, āOkay, I donāt want to share anything with the rest of the country. Iām going to keep my tax revenue for me. Iām going to be a tax haven based on bank secrecy.ā Thatās going to make you 10 percent or 20 percent richer. Iām not sayingāā
COWEN: Itās been a long time since Switzerland relied on bank secrecy, right? Following 9/11, that Swiss advantage largely went away.
PIKETTY: Oh, thatās wrong. Oh, youāre wrong on this.
COWEN: Itās the US thatās the secrecy haven.
PIKETTY: No, it still brings . . . No, no, I can tell you in terms of the banking sector and the status as a tax haven still brings an additional income of at least 10 percent or 20 percent to Switzerland. But I agree with you, Switzerland would still be rich even without this. But they would be a bit poorer, and they will certainly not be richer than if you compare to, say, the Paris region GDP per capita or the London region, if you take the wealthiest region. Itās important to compare countries of comparable size, regions of comparable sizeā¦.
COWEN: But Switzerland is a real country with a diversified economy.
PIKETTY: Yes, sure.
COWEN: Very little of it is poor.
PIKETTY: The Paris region is a real region.
COWEN: But thatās a clustering effect within France. France is much poorer than Switzerland. Could not France bring Swiss prosperity toāā
PIKETTY: This is not comparable in size. I donāt think it makes sense. Again, if you want to compare a region of about 5 million,10 million inhabitantsāāāwhich is the size of Switzerlandāāāyou find many other regions with comparable GDP per capita all across Europe.
There are many good things with Switzerland, by the way. I think the local democratic system has lots of good aspects to it. The education system has . . . I think thereās a lot to learn from each of these experiments.
I feel lame excerpting this part since itās the same part that Tyler excerpted, but Switzerland is one of my obsessions so here we are.
A couple of observationsā¦
At least as far as I can tell from the internet, Piketty is wrong about GDP per capita. The latest GDP per capita I could find quickly for Ćle-de-France was $69,423 in 2016 (from Wikipedia). Swiss GDP per capita was $83,073 in 2016 according to the World Bank. The Swiss figure is in 2020 US dollars. Iām not sure what year the Ćle-de-France figures are in, but if you assume itās 2016 dollars, then that would put the Ćle-de-France GDP per capita figure at $74,862 (based on this admittedly less than fully convincing website). Still less.
This doesnāt invalidate Pikettyās point larger point about scale . Before I lived in Switzerland, I had several people who had lived there mention how small of a country it is when comparing it to the US. At the time, I would roll my eyes.
But having lived there, Iāve changed my mind. It is a really small country and this seems to make some problems more tractable. US federalism and Swiss federalism get compared a lot and they are similar in many ways. But Swiss Cantons, which relate to the Federal government the same way that US States do, are closer to US counties in terms of population and land mass. Thereās a lot more consensus about how things should work at the US county level than there is at the US state level, which makes choosing a direction easier. (Aside: You could argue that the implication here is that even more US governance should happen locally, but I wonder if at some point that would erode the benefits of scale that the US enjoys.)
Even with the caveat of size, I still think Switzerland is super impressive. The way I describe it to Americans is that the entire country is like the wealthiest US suburbs. Everything just works. The infrastructure is well maintained. Itās clean. The bureaucracy is friendly and easy to navigate. Itās safe enough to leave your door unlocked. There arenāt that many places in the world where 8 million people live that you can say that about.
The last thing Iāll say is that I donāt think the Swiss secret is simply less government and less redistribution, although I do think these are assets for Switzerland. If I had narrow it down, I would say that the political culture of trust and compromise and the high investment in human capital are the driving factors. If anything, I think these are what allows the government to be both small and effectiveā¦ which I think places me closer to Pikettyās answer than Tylerās question.
From Daniel Cox at the Survey Center on American Life:
College graduates live increasingly different lives than those without a college degree. They are more socially connected, civically engaged, and active in their communities than those without a degree. I find that college graduates have more extensive systems of social support and a larger number of close friends. Consequently, they feel lonely and isolated less often.
The whole piece is interesting. The message is pretty clear that American Society āworksā for people who go to college in a way that it doesnāt for those who donāt. The differences in levels of social integration are striking and somewhat confusing. Itās intuitive to think that this is just financially driven; if you have more money and stability, itās easier to make friendships and get involved in the communityā¦ but itās also not clear that this is the root cause (at least from this work).
Some other things that stood out to meā¦
The relationship between litter / graffiti and trust in oneās neighbors:
Americans living in neighborhoods where trash and graffiti are common express far lower trust in their neighbors. Less than half (45 percent) of Americans who say garbage or litter are everywhere in their neighborhood say they have a great deal or a fair amount of trust in their neighbors. Eight in 10 (80 percent) Americans who live in places where there is no trash or graffiti nearby say they trust their neighbors at least a fair amount. This pattern holds across community types. Feelings of trust are higher in places without trash, litter, or graffiti marring the physical environment, whether thatās a dense urban neighborhood, a suburb, or a town.
A similar dynamic exists with tap water:
Americans who trust their tap water express a stronger connection to their community and the people who live there than those who do not trust their tap water. A majority (58 percent) of Americans who say they would be very comfortable drinking water from their tap say they feel closely connected to their neighborhoodāa feeling shared by only 44 percent of those who say they would be very uncomfortable drinking unfiltered tap water.
Both of these are pretty intuitive if you think about it ā if a place is dirty, you probably arenāt going to feel comfortable there. If you donāt think you can trust the tap water in a place, youāre unlikely to feel really at home there.
Being involved with a place of worship seems to help with social integration, which makes sense:
Regardless of educational experience, Americans who belong to a religious congregation are much more active in community life and report stronger social connections. Overall, Americans who are members of a place of worship are much more likely than those who are not to volunteer in the community at least a few times a year (47 percent vs. 23 percent), talk to someone in their community they do not know well (64 percent vs. 54 percent), and attend a community meeting or local event (60 percent vs. 41 percent). They are also more likely than others to feel connected to their neighborhood and the people who live there (58 percent vs. 46 percent).
But so does living near stuff to do:
Simply living near a public park, library, coffee shop, or bar is strongly associated with greater community engagement, higher feelings of social trust, and connection to the community.
So to summarize:
thereās clearly a relationship between going to college, participation within society, and trust ā itās not clear what the driver is here.
my best hypothesis is that the people that are naturally inclined to go out and do stuff are now being routed into college (thatās the default path and you have to work hard to get off of it) and that this accounts for most of what weāre seeing, but I canāt prove this.
it seems like having communities that are clean and that provide quality basic services is correlated with trust
being involved with a religious organization is correlated with participation
living near stuff (reducing the friction of participation) is correlated with social trust
Over the weekend, I read We Donāt Sell Saddles Here where Stewart Butterfield outlines how Slack plans to take over the world. Itās a great piece and worth reading if youāre planning on launching a new product. It really captures the dynamism that Slack had in the early days.
This part in particular stood out to me though:
The best way to imagine the reward is thinking about who we want our customers to become:
* We want them to become relaxed, productive workers who have the confidence that comes from knowing that any bit of information which might be valuable to them is only a search away.
* We want them to become masters of their own information and not slaves, overwhelmed by the neverending flow.
* We want them to feel less frustrated by a lack of visibility into what is going on with their team.
* We want them to become people who communicate purposively, knowing that each question they ask is actually building value for the whole team.
As someone who uses Slack every day, I had a visceral reaction to each of these propositions. I never feel relaxed when using the app and I almost always feel overwhelmed by the flow of information.
The app became what it set out to fix ā in many ways, the reduction in friction, which made it so addictive, made many of the problems it set out to solve worse.
Otto von Bismark, Chanellor of Germany, who said: āāPolitics is the art of the possible, the attainable ā the art of the next bestā. Image is from Wikipedia.
Sometimes you have the opportunity to give a user an insanely great product. The organizational support, project funding, and technology all line up to give exceed user expectations. These are rare opportunities ā enjoy them!
Most of the time youāre missing one of the key ingredients: project funding, technology, or organizational support.
Frequently the limiting constraint is a political one. You canāt launch the new product to all users because the sales team is afraid of how enterprise partners will react. You have to launch sooner than you want because a leader has drawn a line in the sand. A partner team wonāt change their roadmap to help you with a dependency.
These political constraints can be the most frustrating because they seem arbitrary. But that doesnāt make them any less real. The best product leaders I know play the long game. They make the case for the best theoretical path, but are willing to accept the best one available. Then they move on to the next iteration. After all, iconic products are built one well thought-out iteration at a time. This flexibility gives them credibility with others, which gives them more space to operate in the future.
Itās common for companies to talk about product managers as mini-CEOs, masters of their own feature set. In many situations Iāve seen, this is actively unhelpful because it doesnāt prepare the product manager or their stakeholders for the reality of what the pm is being asked to do: find the possible, the attainable, the next best.
Delightful treatment of the idea that efficient markets produce products that work by Dan Luu.
The whole thing is worth reading, but the most interesting section to me was the one on trust in between firms and within firms.
Coming back to when it makes sense to bring something in-house, even in cases where it superficially sounds like it shouldn't, because the expertise is 99% idle or a single person would have to be able to build software that a single firm would pay millions of dollars a year for, much of this comes down to whether or not you're in a culture where you can trust another firm's promise. If you operate in a society where it's expected that other firms will push you to the letter of the law with respect to whatever contract you've negotiated, it's frequently not worth the effort to negotiate a contract that would give you service even one half as good as you'd get from someone in house. If you look at how these contracts end up being worded, companies often try to sneak in terms that make the contract meaningless, and even when you managed to stamp out all of that, legally enforcing the contract is expensive and, in the cases I know of where companies regularly violated their agreement for their support SLA (just for example), the resolution was to terminate the contract rather than pursue legal action because the cost of legal action wouldn't be worth anything that could be gained.
If you can't trust other firms, you frequently don't have a choice with respect to bringing things in house if you want them to work.
Although this is really a topic for another post, I'll note that lack of trust that exists across companies can also hamstring companies when it exists internally. As we discussed previously, a lot of larger scale brokenness also comes out of the cultural expectations within organizations. A specific example of this that leads to pervasive organizational problems is lack of trust within the organization. For example, a while back, I was griping to a director that a VP broke a promise and that we were losing a lot of people for similar reasons. The director's response was "there's no way the VP made a promise". When I asked for clarification, the clarification was "unless you get it in a contract, it wasn't a promise", i.e., the rate at which VPs at the company lie is high enough that a verbal commitment from a VP is worthless; only a legally binding commitment that allows you to take them to court has any meaning.
Of course, that's absurd, in that no one could operate at a BigCo while going around and asking for contracts for all their promises since they'd immediately be considered some kind of hyperbureaucratic weirdo. But, let's take the spirit of the comment seriously, that only trust people close to you. That's good advice in the company I worked for but, unfortunately for the company, the implications are similar to the inter-firm example, where we noted that a norm where you need to litigate the letter of the law is expensive enough that firms often bring expertise in house to avoid having to deal with the details. In the intra-firm case and you'll often see teams and orgs "empire build" because they know they, at least the management level, they can't trust anyone outside their fiefdom.
While this intra-firm lack of trust tends to be less costly than the inter-firm lack of trust since there are better levers to get action on an organization that's the cause of a major blocker, it's still fairly costly. Virtually all of the VPs and BigCo tech execs I've talked to are so steeped in the culture they're embedded in that they can't conceive of an alternative, but there isn't an inherent reason that organizations have to work like that. I've worked at two companies where people actually trust leadership and leadership does generally follow through on commitments even when you can't take them to court, including my current employer, Wave. But, at the other companies, the shared expectation that leadership cannot and should not be trusted "causes" the people who end up in leadership roles to be untrustworthy, which results in the inefficiencies we've just discussed.
People often think that having a high degree of internal distrust is inevitable as a company scales, but people I've talked to who were in upper management or fairly close to the top of Intel and Google said that the companies had an extended time period where leadership enforced trustworthiness and that stamping out dishonesty and "bad politics" was a major reason the company was so successful, under Andy Grove and Eric Schmidt, respectively. When the person at the top changed and a new person who didn't enforce honesty came in, the standard cultural norms that you see at the upper levels of most big companies seeped in, but that wasn't inevitable.
Iām not sure at the moment how much I agree with his approach to build vs. buy decisions but I know I agree with his assessment that the essential ingredient for productivity is trust.
Seaweed from Charles Durantās collection ā more here.
Unlike the typical seaweed collector, Durant was not a British woman but an American man, an inventor and scientist. Born in 1805, he seemed to assume, like many educated men of his era, that no aspect of the mechanical, physical, or natural world was beyond his ken. His biography is a litany of claims-to-fame.16 With a hot-air balloon ascent from Battery Park, New York in 1830, he made his name as the first American aeronaut, staying aloft for two hours in a balloon heād sewn himself. His second ascent, in 1833, was attended by President Andrew Jackson and thousands of others, during which Durant dropped leaflets featuring his own ecstatic poetry. These poems, about the virtues of ballooning, seem to have been the worldās first instance of aerial propaganda. He was the first US manufacturer of silkworm gut, a filament used for fishing line, and his raw silk and cocoons took high awards. He turned his attention then to Mesmerism, a faddish belief in clairvoyant hypnosis that was sweeping the nation. After infiltrating Mesmerist circles by pretending to find the practice credible, he wrote one of the first anti-Mesmerism screeds to be published in America, in which he debunked the supposed science with great relish.17 After that book, Durant became interested in hydraulics. He maintained a year of technical correspondence with Ellis S. Chesbrough, chief engineer of Bostonās waterworks, and soon-to-be engineer of Chicagoās sewage system. Their letters were published as Hydraulics: On the Physical Laws that Govern Running Water (1849).
Charles Durant went on to write the first American book on seaweed, and according to this article, at least, the best one. The whole article is worth reading, especially if like me, youāre interested in seaweed.
The geography of the industrial revolution in Britain was shaped in part by the corrupt, rentier economy of London in the eighteenth century. The various by-laws, regulations and tax regimes in and around London largely precluded the establishment of new industrial processes. This drove industrial investment to the midlands, and especially to the steep and narrow valleys of Yorkshire, Lancashire and South Wales, where there was abundant water-power to drive the new industrial machinery. The need to efficiently pump water provided the impetus for the development of steam, driving industry forward.
From Does Britain Exist by Tim Watkins. You often hear about regulations constricting growth and leading innovation to happen elsewhere, but I think this is the first time I can remember someone citing a specific example.
Thanks to the Browser for the recommendation
I love this post by Uri Bram on export subsidies and how they could be applied to personal skill development.
Hereās Uri on export subsidies:
South Korea is one of the greatest development stories rarely told. In just 60 years, they've taken 50 million people from largely rural poverty to a higher GDP per capita than Japan's - it's a bonkers achievement. How did they do it?
Through export subsidies. For the full story, read How Asia Works, but the short version is that Korea strategically brought its manufacturing sector up to scratch by subsidising firms conditional on them successfully exporting products; that is, on selling things abroad on the open market to people who had plenty of other options and no particular reason to buy Korean.
His really interesting idea is how you might apply this to skill development:
So paying your kid's costs while they do a writing degree is not an export subsidy, it's just a regular-subsidy. But promising to match any money they make from submitting freelance articles to magazines? This is what we'll call a Personal Export Subsidy. By supplementing the money they make from successfully placing freelance articles, you're letting an impartial external arbiter (the various magazines) decide whether your kid's work is actually worth something, while acknowledging that in the early days you'll need to increase that "something" for your kid to survive while climbing the ladder.
Iāll add to this that you think of things like the YouTube Shorts Fund as a sort of export subsidy for short form video creation. Another example of this is venture capital, particularly at early stages where funding is pretty directly tied to showing some form of commercial traction and the bar for company traction gets raised as the company matures.
Product management is a generalistās field. You do a little bit of user research, a little bit of strategy, a little bit of design and a little bit of engineering. Everyone you work with is better at what they do than you are, yet you still have to find a way to help them.
Within a field of generalists, I am a generalist PM. Iāve worked across agriculture, ad tech, consumer social, and most recently fintech. This makes me doubly a generalist.
Perhaps because of this, I am especially appreciative of the art of being a useful generalist, which Ross Simoini captures so well in this article. Hereās an excerpt:
In this way, the generalist must have a high tolerance for complexity, confusion, and uncertainty. Generalism does not offer the clearly tiered progression offered to the specialist. Working across various fields means you will likely spend long periods being unskilled at them. The generalist can acquire new talents, but they are also a perpetual amateur in a cycle of discovery and failure. There are benefits to this process: slowness encourages a certain quality of attention; novelty encourages a sharp perspective; and an outsiderās position keeps you immune from the insiderās tunnel vision.
Specialism, on the other hand, offers an easily measured form of success. In fact, specialization usually defines a spectrum of success and failure.āBestā and āworstā can exist only when the boundaries of success have been narrowed to a single parameter: the best RBI hitter in baseball; the most dividends earned in a single day.
I left this article thinking there should be a Society for Generalists the way there is are professional groups for specialities. Worth reading the entire thing, especially if youāre a generalist.
A brilliantly insightful speech on how to foster creativity . Here's a quick summary:
Now here's the negative thing: Creativity is not a talent. It is not a talent, it is a way of operatingā¦
You see when I say āa way of operatingā what I mean is this: creativity is not an ability that you either have or do not have.
It is, for example, (and this may surprise you) absolutely unrelated to IQ (provided that you are intelligent above a certain minimal level that is) but MacKinnon showed in investigating scientists, architects, engineers, and writers that those regarded by their peers as āmost creativeā were in no way whatsoever different in IQ from their less creative colleagues.
So in what way were they different?
MacKinnon showed that the most creative had simply acquired a facility for getting themselves into a particular mood ā āa way of operatingā ā which allowed their natural creativity to function.
Cleese describes an āopen modeā where sort of meander or play with a problem. In open mode, there's no right or wrong. Crucially, it's extremely difficult to be open with time pressure. He tells relays this story about how Alfred Hitchcock would help move his writers into āopen modeā:
āWhen we came up against a block and our discussions became very heated and intense, Hitchcock would suddenly stop and tell a story that had nothing to do with the work at hand. At first, I was almost outraged, and then I discovered that he did this intentionally. He mistrusted working under pressure. He would say āWe're pressing, we're pressing, we're working too hard. Relax, it will come.ā And, says the writer, of course it finally always did.
Then there is a āclosed modeā, where we implement our solution and are rigorous about speed, efficiency, details, and outcomes.
The most creative people as ones who can move most quickly between these two modes.
Cleese then gives 5 tricks for getting yourself into open mode:
Space where you will be undisturbed at least until a specific time.
Time to get into open mode, usually at least 30 minutes, where your mind wants to go back towards execution. Then after about 60 more minutes, usually the most creative time is past and you need a break.
Time to play with the problem, to stick with the discomfort of not having a solution. The most creative people spend the most time in this space of not knowing or considering alternatives before picking a path.
Confidence in yourself, to handle the discomfort of not knowing the answer, and to be wrong as you try things out.
A 22 inch waist (humor) , Cleese's way of reminding us that nothing moves us into open mode faster than humor.
I have bookmarked this to read again in 6 months. Sent to me via the Flux Collective .
That is the subject of my latest Browser Bets episode with Andrew Flowers of AppCast.
Hereās a snippet:
Andrew: The quits rate in the US had an all-time high of three percent in November of last year, that's the latest data we have. So what does that mean? It means in November three percent of all workers quit their job and in the 20 years that the bureau of labor statistics has been tracking this that's the highest point. My bet in quantitative terms is that the quits rate in the US by the end of 2022 is going down below two and a half percent which is where it it peaked before the pandemic.
I respect Andrew a lot, but I actually think heās wrong here. I mention this in the show, but Iāll say it more succinctly here. I think three things are going to keep this number high for the rest of the year:
A structural shift towards remote work. Iām on the record as thinking that back-to-the-office is never going to happen for a certain categories of workers. But adjusting to this new normal is going to take time and some employers are going to try to force employees backā¦ and these employees are going to quit to take jobs that allow for more flexbility.
The retirement of the Baby Boomers. This is simply a demographic tailwind, perhaps pulled forward by employers trying to enforce return to work (if youāre going to retire in 6 months anyway, do you really want to commute again?)
Remote jobs will have a higher natural quit rate. If youāre a remote worker, thereās way less friction in quitting a job. Thereās no new commute, thereās minimal disruption in your schedule. You get shipped a new computer and log in to different video calls. For remote first workers, itās going to be easier to hop between jobs and this will show up in the numbers.
Only time will tell who is right. You can join in on the bet here.
I loved this podcast between Ezra Klein and Alex Tabarrok for multiple reasons, not the least of which is my continuing obsession with the importance of trust in flourishing societies:
ALEX TABARROK: I mean, very similar to this, I think undermining trust in government. Andriei Shleifer has some work showing that trust, as you know, is down in the United States. And the kind of weird thing is, is that trust in government is down, but actually, this doesnāt lead you to kind of a libertarian paradise where people say, I donāt trust the government, letās use the market.
Actually, what happens is trust goes down in all kinds of institutions. And if anything, people become more in favor of government. Not that they actually think itās going to work, but they just think everything is unfair. And they think that nothing is going to work, and they become removed from the political process. But decline of trust doesnāt lead you to something which I want. And itās unfortunate, in a way, because you would hope that people would sort of ā well, if not the government, then the market. But thatās not the way it works.
Oddly enough, right, the societies in which people have the most trust, not only do they trust the government more, but they actually also trust markets more. These things work together. So if you have a lot of trust in government, then youāre actually willing to have free trade because you figure, well, the bottom half of people are still protected. Youāre willing to have vaccines, which actually gets the economy going. So it turns out that trust in markets and trust in governments correlates actually pretty highly.
And later:
EZRA KLEIN: But that strikes me as somewhat to the side of the issue Iām bringing up here, which is that if you want to align the incentives of more groups to push for more overall growth, as opposed to pushing for just their slice of a stagnant pie, to mix my metaphors here, then you actually want lower levels of inequality. And I even see that in the housing example you turn to there. Iāve attended these meetings in S.F., and I try to watch what happens with them.
And something you cannot miss in a city like this, which has insane levels of internal inequality, just insane, like nothing Iāve ever seen, is that thereās no trust.
This certainly matches my intuition that trust in society is self reinforcing. But how to start moving towards trust to begin with? That I wish I had a better answer to.
Notre Dame Cathedral in the spring of 2019
Iāve researched and consulted on megaprojects for more than 30 years, and Iāve found that two factors play a critical role in determining whether an organization will meet with success or failure: replicable modularity in design and speed in iteration. If a project can be delivered fast and in a modular manner, enabling experimentation and learning along the way, it is likely to succeed. If it is undertaken on a massive scale with one-off, highly integrated components, it is likely to be troubled or fail.
That is from Bent Flyvbjerg in HBR.
Two things this article pointed out that I hadnāt considered before:
Long timelines in a project create risk because it leaves more time for things to go wrong. Iāve rushed to ship a product before for fear of the competition, but the concern is more general than that. The longer it takes to get going, the more likely that the environment changes ā a project sponsor leaves the company or an interested beta customer changes their plans. Even more so for public works.
Projects that are useful in stages are more resilient. Iāve prioritized shipping a project in iterative chunks as a way of testing a hypothesis about my product before, but never considered that the faster you get a project producing, even partially, the more time its benefits compound and the faster it pays back itās costs.
The entire article is worth a read, especially if youāre in the midst of planning or executing a major project. The example of the Madrid subway is both obvious and eye opening.
One more thing it made me consider: modularity creates resiliency. One of the ways cathedrals like Notre Dame were able to be built over hundreds of years was by using patterns that were easily understood and able to be extended. Calamities could come and derail projects, leaders could die, but new ones could take their place and take up the work again because of the modularity of the system.
Thanks to The Browser for bringing this one to my attention.
Iāve always thought of myself as an 80 percenter. I like to throw myself passionately into a sport or activity until I reach about 80 percent proficiency level. To go beyond that requires an obsession and degree of specialization that doesnāt appeal to me. Once I reach that 80 percent level, I like to go off and do something totally different; that probably explains the diversity of the Patagonia product line ā and why our versatile, multifaceted clothes are the most successful.
From Let My People Go Surfing by Yvon Chouinard, the founder and owner of Patagonia. Reminds me of David Epsteinās book Range and the compounding value of exposure to different problem types for innovation.
Probably a good habit to ask āWhat new skill have I reached 80% proficiency at recently?ā
I really enjoyed this interview between Casey Newton of the Platformer and Molly White of Web3 is going just great.
In it, Molly made a point Iāve thought about before but not been able to fully articulate, which is that in an effort to align incentives financially, many web3 projects undermine their own effectiveness:
There also aren't really intrinsic monetary incentives for people to contribute to Wikipedia, which I think is a very good thing. Where people are paid to edit Wikipedia by outside parties, it warps the incentive to contribute into one that's very different from (and sometimes at odds with) the incentives for most community members, and is often a very negative thing.
She references Everipedia, a blockchain-based wikipedia alternative:
If you look at their recent blog posts, it's all about how many tokens their editors have supposedly earned, and it even brags about the fact that "Over 70% of stakers have locked their IQ up for over 3.5 years to earn max APR". This is the same token that people are supposed to be spending to edit and vote on the quality of edits, but they're excited that people are locking them up on staking platforms? The goal is not to create a reference work, it's to make money off the token.
Speaking more broadly, monetizing things just shifts the dynamics in enormous ways. We've seen this same thing happen with play-to-earn gaming, where people start doing things really differently when monetary incentives are added.
I think this is a really smart point and poses problems for Web3 projects.
Financial incentives arenāt always the best way to get people to do what you want (see the daycare that started charing parents for being late and then saw an increase in tardiness), except that itās worse than that for Web3 projects because it makes it harder for them to tell if they have product market fit. Because there isnāt a great way to distinguish the speculators from the true believers, it feels like the project has true momentum, when all it has are people trying to make a quick buck.
Another episode of Browser Bets, this time with Tom Chivers. We made bets about Boris Johnsonās future as Prime Minister of England and the likelihood of AI wiping out humanity.
Bets aside, my favorite part was processing the experience of going through COVID together. Tom and I had never met before this conversation and live thousands of miles away from each other, yet we instantly had this shared experience to talk about. This is pretty unique. Watch the whole thing.
I did an interview with Helen Toner for The Browser where we made bets on Chinaās economic development, the possibility of an AI delivered milkshake, and quantum computing.
One of the things that living in Switzerland caused me to appreciate is the impact of trust in society. Switzerland measures as an exceptionally high trust society.
When you live there, itās something you can actually feel. The way I describe it to people is that in Switzerland, everywhere feels like high end American suburbs (holding aside for the moment that not everyone feels welcome in American suburbs). Things just work. You can leave your door unlocked.
I have a hypothesis that this trust is self reinforcing. Because people trust each other, additional things are possible. Because of these things, people trust the system. My example here is the Swiss Recycling system, although Iām sure someone could come up with something better.
Because of this experience, societal trust has become something I really want to better understand. Where does it come from? How can we make more of it? What destroys it?
With that background, I loved this interview between Phares Kariuki and Uri Bram. The whole thing is worth reading, but Phares offers two hypothesis about what creates trust and one about what destroys trust.
Trust creator #1: Violence
How you move from one equilibrium to another, from observation seems to be violence. It is cruel to think about but Europe went through countless wars in order to integrate.
Trust creator #2: Contract enforcement
The primary thing that can be done to increase trust in society is to have a level of justice for crime / breach of contract. This enforces good behaviour and dissuades bad behaviour; places with high trust have the highest rates of contract enforcement but also contracts aren't needed -- folks can shake on it.
Trust destroyer: Foreign interference
Phares Kariuki: Additionally, I've seen high trust societies get decimated by foreign interference (Korea, Somalia, Germany).
Uri Bram: Iād love to hear more about the Somalia example -- I think some people reading this might be surprised to hear it had a previous high-trust phase.
Phares Kariuki: The Somali were one people, largely Sunni. Their territory covered part of Eastern Ethiopia, North Eastern Kenya. They were split into multiple countries during the colonial era, with Kenya famously oppressing them during the Shifta wars of the 70s. They wanted to secede. The interference in their leadership due to the Cold War led to oppression and clan based mistrust; the fallout stands until today.
St. Moritz in the winter
How St. Mortiz, a town in the Engadin Valley became a tourist destination:
Johannes Badrutt, the business-savvy owner of the Engadiner Kulm hotel in St. Moritz, is celebrated as the father of winter tourism. According to legend, he spent a rainy evening in September 1864 by the hotelās fireplace, talking to visitors from London. Badrutt claimed that on especially sunny winter days it was possible to walk around the area without a coat because of the warmth of the sun. He encouraged the English tourists to return in winter and see for themselves, offering to pay their travel costs if his account proved untrue. Travelers could not resist this intriguing bet and returned in mid-Decemberāarriving under a bright sky and covered in sweat. They stayed until March.
From Laphamās Quarterly. The Swiss donāt have a reputation for branding experts, but they should. It is one of the best and most subtly branded countries in the world. See also this article on Little Switzerlandās.
Both articles were brought to me via The Browser.
I contributed to a piece for the The Morning News about what 2021 will be known for and what we thought it would be known for, but didn't pan out.
My prediction for what we'd remember the year for is the breakout of electric vehicles. From the piece:
2021 was the year that the electric car won . Electric-car makers have sky-high valuations and traditional car makers are plunging billions of dollars into electric-car manufacturing and telling markets that 40 percent or more of their sales will come from electric vehicles by the end of the decade.
I almost had commercial fusion energy in this spot, but decided the evidence for it actually happening isn't strong enough yet.
For what was supposed to happen, but didn't pan out, I had return-to-office. I go on to predict that return to office will never really happen for most workers.
Covid is going to continue to cause havoc over the course of 2022 and 2023, whether it's additional variants or concern about breakthrough cases. While this will diminish over time, by the time it does, any job that doesn't explicitly require the worker to be in person will have moved to remote-first by default and any company that tries to buck this trend will be at a disadvantage in the marketplace.
Omicron, of course, is an example of how return-to-office will continue to be delayed. Eventually it will get too hard for companies to put the genie back in the bottle.
Luzern, Switzerland, my favorite place I visited in 2021
Last year, I was reading Tom Whitehall's 52 things I learned in 2020 and thought āwhat a clever way to build the habit of curiosity,ā so I decided to copy it (here is his 2021 version , if you're interested). With that background, here are 52 things that were new to me or caused me to think differently in 2021, loosely organized by category.
1) Greenland sharks commonly live 200+ years. Some are likely still swimming from before Shakespeare was born. They don't reach sexual maturity until 150 years of age and their normal pregnancy is 12 years long. Source .
2) You can walk from Norway to Canada. Or at least you can if you're an arctic fox. Source .
3) Human made stuff weighs more than all life on Earth , by one measurement at least. Source .
4) The importance of clean air: $ 700 air purifiers in Los Angeles improved test scores by almost as much as if almost as much as it would if class sizes were reduced by a third, according to David Wallace Wells , who estimates that 10M lives a year are lost due to air pollution.
5) Many countries in the world are transitioning towards reforesting . Scotland is the best example, after bottoming out around 4% of land area with forests in 1759, it's at 18%, close to where it is estimated to have been in the year 1000 AD, when it was 20%. From Our World in Data .
6) You can use sound to put out fires. Via Ted Goia .
7) 18 of the 20 horses that raced in this year's Kentucky Derby were descended from Secretariat . Juliette Kayyem
8) The Swiss constitution regulates the number of second homes in various communities, capping it at 20%. This is just one example of the way that the Swiss constitution is different than the American one - it can be (and is) frequently amended by referendums that get incredibly specific. From Why Switzerland.
9) From 1872 through 2003, no sitting member of the Bundesrat (Swiss Federal Council) was not re-elected. From Why Switzerland.
10) LSD was first synthesized in Basel, Switzerland. Via Crooked Timber.
11) The concept of concordance, the Swiss political model of seeking mutually acceptable compromises between competing political factions or between management and labor. From Why Switzerland and my friend Lovro.
California12) The bear used as the model for the California state flag was the pet bear of William Randolph Hearst . The bear's name was Monarch. Via the Voice of San Diego podcast .
13) Palm trees are not native to California. They were imported to make it look more like the holy land. Via Justin EH Smith ; special shoutout to Scott Lewis who also regularly points this out.
14) In California, any item can use the recycling symbol , regardless of whether or not it can actually be recycled. California is seeking to change this. I can't believe this needs to be legislated. Via the NYT .
15) The city of San Diego uses 30% less water than it did in 1990. That is an overall reduction, not per capita. This happened despite the city growing by about one third during that time. Via the Voice of San Diego podcast .
16) Joshua Tree National Park is larger than the state of Rhode Island . I hope to visit at least a corner of it in 2022. Via Wikipedia .
17) The original name of Bank of America was Bank of Italy. It was founded in San Francisco. Via California: A History .
18) The Los Angeles Clippers have never retired a number. Via Bill Simmons .
19) The most decorated American unit in WW2 were Japanese Americans serving in Italy. From California: A History : āFour months later, some 110,000 Japanese aliens and Japanese Americans were behind barbed wire, where they would remain for the next three years and more, except for those young nisei who volunteered for the draft in 1943 and, assigned to the 100th Infantry Battalion and the 442nd Regimental Combat Team, fought their way up the Italian peninsula in 1944 and early 1945, emerging as the most decorated combat units of the war. "
20) The US Military is the one US institution that has gained trust since the 1970s . Improvements in diversity of hiring, reducing the gap between stated and actual values, the volunteer army, and the Gulf War success are credited with improving their standing. Via Amanda Ripley . I love this framing of the origins of trust, from the same article:
Typically, trust gets traced back to three central ingredients: ability, benevolence, and integrity. Ability captures the obvious, rational reason to trust something or someone: they seem to know what they are doing. Benevolence reflects the sense that an organization has our best interests at heart, that they are motivated by the forces of good. And integrity means that the institution has strong, admirable values to which it adheres, even under pressure to do otherwise.
21) When median rent starts to exceed one third of median income, homelessness starts to rise rapidly . You could argue that I should have put this in the California section. Via Chris Glynn and Alexander Casey .
22) Zoning reform by itself doesn't lead to development (affordable or otherwise). Minneapolis removed zoning restrictions in 2018 on duplexes and triplexes and a grand total of three got built by 2020 because of a lack of changes to the building code. From Strong Towns .
23) Scaling back prosecution of small, non-violent crimes reduces violent crime by keeping people out of the criminal justice system . It seems that going to prison increases the likelihood of committing future crimes. From reasons to be cheerful.
24) Famously bike friendly Amsterdam was as car choked as many other world cities as recently as the 1970s. From The One-handed Economist .
25) The role of deregulation in the shale boom . The Energy Policy Act of 2005 allowed for drilling of oil on public land to skip environmental review as long as the project met predetermined limits. This led to more certainty for investors, more rapid turnaround in projects, and rapid technological innovation. We should do the same thing for carbon neutral technologies (eg, geothermal). From Eli Dourado.
26) The average time to complete a National Environmental Protection act environmental review now takes 4.5 years to complete and is more than 600 pages long. From James Pethokoukis .
27) It took just 16 days to plan the Central Park Zoo. The Power Broker .
28) Lack of resilience seems to be a primary factor in civilizational collapse (at least this is how I understand Spencer Greenberg's conversation with Samo Burja ). I'm editorializing a bit here, but I think we have to find ways to improve environmental (or societal) outcomes without putting a straight jacket on our society,
29) Non-violent protest is statistically more likely to create lasting social change than violent resistance . I heard this first while listening to the amazing City of Refuge podcast and didn't quite believe it, but was pointed to Erica Chenoweth's book Why Civil Resistance Works by the show's producer (this podcast is a good introduction). The way this works is that stable social change ultimately requires buy-in from people, especially leaders of institutions. While violence can be an effective short run deterrent, overtime it repels people.
30) Americans are rapidly becoming less likely to think God exists. Especially Gen Z. Via Ryan Burge .
31) Unmarried young adult Americans are having sex less often . The driving factor seems to be later marriages and, increasingly, religious observance. Via Lyman Stone .
32) The great downsizing is coming. The leading edge of the baby boomers are 75 today and the share of 80 year olds in the population is set to increase rapidly over the next 10 years , which is important because 80 is when people tend to downsize and move into nursing homes. Via Calculated Risk .
33) When in a group of people who speak different languages, the chosen language for conversation tends to be the one known best by the participant that knows it least well, not the one that most people speak best, or the one with the highest average proficiency. Via Eurozine .
34) People who have been connected tend to double down out of embarrassment rather than change their minds. Via Brooke Harrington .
35) The phenomenon of terminal lucidity , where patients with dementia become themselves again shortly before they pass away. From The Guardian .
36) Obesity is almost definitely not caused by overeating, will power, or self control. In 1975, no country in the world had an obesity rate greater than 15% ; today this is common . From Slime Mold Time Mold .
37) AIDs has existed for almost 100 years , not just since ~ 1980. It is thought to have been in the US as early as 1940. Via Sarah Schulman on the Ezra Klein show .
38) The placebo effect is getting stronger in the United States. āThe implications of this are pretty serious - the placebo effect in the United States has actually become quite a lot stronger over time, meaning that drugs that once would have been approved may not be now - because their performance relative to that of placebo is less convincing. "From All That Is Solid .
39) Orphans were used to transport the smallpox vaccine across the Atlantic from Spain to Venezuela in 1803 . They were intentionally given cowpox, which prevents smallpox, two at a time across the ocean until they made it to Caracas (this is where the root word for vaccine comes from, cow is vaca in Spanish). It's unlikely that the orphans were asked if they wanted to do this, but they likely saved thousands of lives. Via The Atlantic .
40) Be careful about your heuristics. Over focusing on what is easily measurable can cause you to miss what's truly important. This is called the McNamara fallacy. It's best summed up by this quote by the Great Bill Russell: Let's talk about statistics. The important statistics in basketball are supposed to be points scored, rebounds and assists. But nobody keeps statistics on other important things - the good fake you make that helps your teammate score; the bad pass you force the other team to make; the good long pass you make that sets up another pass that sets up another pass that leads to a score; the way you recognize when one of your teammates has a hot hand that night and you give up your own shot so he can take it. All of those things. Those were some of the things we excelled in that you won't find in the statistics. Via Aeon .
41) Be more adventurous with your experiments. At Bing, 2% of experiments led to 74.8% of gains. This suggests that we're probably being way too conserative with what we're testing. Via University of Chicago Press .
42) Habits are very sensitive to environments. If you are having trouble breaking a habit, try changing your environment; if your environment is changing, be intentional about your habits. Via David Epstein .
43) Don't forget to look for what you can remove to solve a problem . We systematically overlook subtractive changes that could be beneficial (removing something to improve it). From Scientific American.
44) Forecasts made after a vacation are more accurate. Source .
45) You can't self talk and scan your peripheral vision at the same time . From Allan Parker via Oscar Trimboli .
46) Busyness is a form of laziness , which I heard for the first time from my friend Uri Bram in his interview with Oliver Burkeman . I think about this at least once a week.
47) Fundamental error attribution , where we explain our faults as due to our situation and the faults of others as due to their characters so applies to friends vs. enemies. So if we have a friend who makes a mistake, we're more likely to attribute it to their situation, and if it's an enemy, we're more likely to ascribe it to their character. The key is to try and figure out what the other person thinks went wrong from their perspective. From non zero .
48) The most precious resource is agency : āThis is not worship of employment, but a simpler observation: It seems that the more you ask of people, and the more you have them do, the more they are able to do later on their own. It is important to note that while we shouldn't allow children to be bobbin boys, no one would describe Steve Job's summer job at 13 as his exploitation. We should be thinking much harder about making sure children can make meaningful contributions to the world." From Simon Sarris and a theme of the parenting books I read this year, Montessori Toddler .
49) France has more successful jailbreaks by helicopter than any other nation . GQ
50) The world's first unicorn was United States Steel, founded in Pittsburgh PA . Via Google Arts and Culture .
51) The original definition of the word weird: destiny-changing power. From Grow by Ginko .
52) Mariah Carey made 1.9M from All I want for Christmas is you in 2020 ( NBC Philadelphia via Conor Sen )
## House of Broken AngelsHouse of Broken Angels Cover
I picked up House of Broken Angels as another stop on my literary tour of California, but after reading it, I wouldnāt call it a California book. Itās a San Diego book, through and through. Iām not sure how it reads to someone who hasnāt lived here, but even as a relatively new resident, I recognized places and neighborhoods.
House of Broken Angels is the story of two half brothers, both named Angel, of a man from Tijuana as the elder Angel (Big Angel) nears death from cancer. The two Angels wrestle (figuratively and literally) with each otherās and their fatherās sins, as do the rest of their families. Itās a book about how short life is and about how frail humans are.
The other thing that places this book in San Diego is the connection the characters have to Tijuana, where the main character immigrated from. The two cities, just 20 miles a part, share an economic and social relationship. They mayors regularly meet with each other. Now that I live here, this seems obvious, but it surprised me at first.
The book brings the relationship between the two cities to life. The Angelsā father is from Tijuana, but immigrates to San Diego when he leaves his first wife (Big Angelās mom) for his second wife (Little Angelās mom). When I lived in Europe, one of my favorite things were the border regions where cultures bled into each other. The Italian part of Switzerland, which feels like both Italy and Switzerland, the northern part of Spain, the feels both Spanish and French. San Diego has elements of that. This book the way that people, relationships, and culture move back and forth across the border.
To me, the best part about this book was the way that it approached the end of life, the yearning for one more Christmas morning, and regret over mistakes. The other thing Iāll take from it is some understanding of the Mexican-American San Diego experience.
The brilliant Dr. Maggie Lieu aka Space Mog was kind enough to join me on Browser Bets. I was surprised to learn that she volunteered to go to Mars and was very nearly selected.
Our bets:
Bet 1: Humans will land on Mars by 2035 James says before, Maggie says later. Maggie thinks it will be China, James hopes it will be the US.
Bet 2: Weāll have generalizable robots in our houses by 2035 Maggie says no way, James says yes out of an unfounded sense of optimism.
Bet 3: The standard model of cosmology will be overturned by 2035 by something that aligns quantum physics and relativity. Maggie says yes, James says no.
Sebastian is an e-sports executive, investor, and entrepreneur. We made bets about the future of NFTs, the Creator Economy, and UGC gaming. In between, I learned about how working for Darryl Morey influenced how Sebastian thinks about predictions.
Here are the bets we made:
The future of NFTs: No more than 2 of the top 50 NFT projects on OpenSea.io will still be there in October 2023 (Last 30 day trading volume)
The future of the Creator Economy: If you look at the top 10 web3 companies in 2023, less than 30% of the value will come platform companies.
The future of UGC gaming: By 2030, we will see the same fragmentation in gaming that we do in video. More hours of playtime for top UGC game platform more than the top studio game.
A transcript is available here.
The next book in my literary survey of California has been Grapes of Wrath, the iconic novel that I missed in high school.
The book tells the story of the Joan family as they leave Oklahoma and head west to California, driven by the Great Depression, mechanized farming, and poverty.
[Iām not going to review the book since I donāt think thatās useful for a book of this stature. Iām just going to reflect on it. Iām also not going to worry about spoilers, so if that bothers you be warned].
The book is extremely well paced. Steinbeck slowly turns up the pressure, showing how poverty forces the Joads from one no win situation to another. After enough least-bad options, the family eventually breaks apart. The slow creep of ruin really affected me, as did the images of starving children and pregnant women.
I was surprised at how deep I was into the book before I realized how bleak the ending would be. There wouldnāt even be the satisfaction of a shootout.
The ending! I canāt believe no one I knew let on even a little bit about how weird it is. I canāt claim to understand it.
Itās impossible to read this book now and not think of the homelessness crisis in California. Iām sure some of the members of Californiaās current tent camps would agree with Steinbeck that the police cause more injustice than the people.
For better or worse, the claiming of land is a part of the California psyche. Wether it is water rights, land claims, zoning, or the right to keep your view in a California beach town, there is a preoccupation with protecting oneās claim from others in the golden state.
I think the biggest thing that this book added to my understanding of California is an appreciation for how turbulent the period mass inward migration was, even though it is a crisis that has now passed. I think I understand better why it was such a big deal.
On the whole, I enjoyed the book. Although I liked East of Eden more, I thought the characters in Grapes of Wrath, particularly the female characters were more fully imagined.
I'm more proud of this than I have any right to be. This morning I made it to 365 straight days of practicing German on Duolingo. Unsurprisingly, after using it for this long, it's become one of my favorite apps (1).
When I first started using Duolingo, it was as a supplement to tutoring sessions with a teacher, but when we moved from Switzerland, I no longer had the bandwidth for classes. Since then, my German practice has been Duolingo, the NZZ podcast Akzent, and reading the occasional children's book to my daughter. I have to admit, I'm surprised at how effective Duolingo is. I can tell from my other German language activities that I'm better today than I was 3 or 6 months ago.
Here's what I love about the app:
After 365 days, here's what I wish they'd do differently:
1: I'm obligated at this point to mention that Duolingo is a Pittsburgh company
## Where I was fromEarlier this year, I moved back to California. I am a Californian by marriage and a somewhat reluctant one at that. The muted seasons and even the beautiful beaches, which I sometimes enjoy, arenāt really my thing. And yet this is where life has taken me again.
In an effort to make the best of it, I resolved that if Iām going to live in this state, Iām going to appreciate it, and I kicked off a California reading tour.
The latest book in my survey is Where I was from, a memoir by Joan Didion. Iām still at the beginning of my California reading journey, but so far this would be the first book I suggest anyone read if they want to understand the state.
Where I was from is about the authorās relationship with California, how her understanding of California changed as she grew up, how the California of her youth faded away, and her relationship with her parents as they grow old and ultimately pass away. Itās exceptionally well written. It weaves the authorās family history into notable events from California history and both of these into the human experience of leaving behind a version of a place, of yourself, and of those you love you as you age.
Where I was from isnāt primarily a history book. Perhaps because of this, it has unlocked California for me in a new way, like how meeting a friendās parents for the first time helps explain who they are.
Itās common to talk about Californiaās boom/bust cycles because of the gold rush, but itās also apt. After all, this is a state thatās population has increased by more than 50% in a decade 5 times since 18501. Didionās book spends time on less famous booms: the aerospace industry after the Second World War and the development boom of the 1960-80s when the great ranches of California were broken up and developed into communities.
The aerospace boom is told through the perspective of Lakewood, California, a suburb that grew up next to the McDonnell-Douglas plant, and then struggles with its identity as that industry moves away in the waning days of the Cold War.
The development boom is told through the stories of the heirs of the Irvine and Hollister estates, great California ranches passed down intact from Spanish and Mexican land grants. These heirs proceed to break up these great ranches to make suburbs and shopping malls, along with a tidy fortune.
For Didion, the settlers that rushed West during the gold rush, the families of Lakewood and the heirs of the great ranches all have a common Californian experience: Each generation rushes headlong to make California into their version of paradise. Some make it and prosper and others are left behind. In in doing so, they change California. In the end, those who profited most from the changes look back and wonder what theyāve lost in the process2.
Didionās Californians are not fearless and self sufficient pioneers, but real people terrified of getting caught in the mountains before the weather turns. They benefit greatly from the investments of the Federal Government, be those investments railroads, aqueducts, or jet planes. Their experience of striving to change the world and then having to live with the consequences is a human one. If there is something uniquely Californian, it is the speed and the scale with which those changes take shape.
The best example I have of how this book has changed how I view California is Cannery Row in Monterey. Today, Cannery Row is a tourist trap. A couple of the old canning operation buildings are preserved but no actual fishing or canning continues. When I had visited it in the past, all I saw where the tschotskes and doodads. But now I see it differently: a monument to a past generation of Californians who strived to build a new world and succeeded, only to have that way of life slowly become irrelevant and fade away3.
Notes
I didn't have anywhere to put this, but I especially appreciated the Didion's mother's observation that California had become "all San Jose."
1: Californiaās growth is really stunning. The first decade where it didnāt grow more than 20% was 1980.
2: The addendum to this book set south of Market street in San Francisco about the tech industry almost writes itself. I also appreciated the author's observation that it is especially Californian to feel that anyone who shows up after you is altering the state beyond repair.
3: In a typically Californian way, just up the street from Cannery Row, an industry that literally fished itself out of existence, is the Monterey acquarium, with emphasizes the importance of ocean conservation.
# About About | jdilla [Homeš”](/) [Aboutā¹ļø](/about) [Blogš](/blog) [Booksš](/bookshelf) [Searchš](/search) [Subscribeš¬](/subscribe) # About me I live in Roswell, Georgia, just north of Atlanta. I'm currently taking a break and exploring what I want to build next. Most recently, I co-founded and scaled [Macro Oceans](https://www.macro-oceans.com/) , a company that makes high-performance biomaterials from kelp. I led the launch of our first products and signed up our first customers. With my extra time, Iām tinkering on [Recap](https://heyrecap.com/) , a side project supported by an [Emergent Ventures](https://marginalrevolution.com/marginalrevolution/2024/08/emergent-ventures-36th-cohort.html) grant that uses AI to summarize local government meetings. One of the things that makes me unique is that I've lived a lot of lives on the way to this one. I've worked in non-profits, Big 3 consulting, and product management at YouTube and Stripe. I've also lived in Pittsburgh, North Carolina, San Francisco, and Switzerland. Perhaps because of this, I'm a [passionate believer in the ability of generalists to solve problems in ambiguous environments](https://www.amazon.com/Range-Generalists-Triumph-Specialized-World/dp/0735214484) . **What Animates Me** * **Building things that matter.** "[The world is a museum of passion projects](https://x.com/collision/status/1529452415346302976?lang=en)." We have incredible power to shape our future. Courage and tenacity are required for the job. * **Trust.** An invisible power that allows people, teams, and societies to flourish. * **Ecosystems and communities.** My friend [Alex](https://komoroske.com/) has created (or revealed?) in me an obsession with how ecosystems catalyze, grow, and sustain themselves. **Current Interests** * **Novel materials** and the ecosystems required to bring them to market. I love a great "how [material x] became ubiquitous" story. Send those my way! * **How to use LLMs / AI to do great work**. For the first time since the internet, the tools available for doing great work are changing, opening up new possibilities. Let's harness them for good! If you have thoughts about these or just think we'd have a good conversation, put some time on my [calendar](https://calendly.com/macro-oceans-james/1-1-meeting) or send me an email: hello@jdilla.xyz. One of the main reasons I keep this site is to inspire some serendipity. ### Past Products and Projects * **[Macro Oceans:](https://www.macro-oceans.com/)** Co-founded the company and led launch and early commercialization of seaweed-based biomaterials. * **[Stripe Apps Ecosystem:](https://marketplace.stripe.com/)** Launching and scaling the app ecosystem for Stripe users worldwide. * **YouTube Search and Creator Studio:** Lots of different features for Creators of all types and sizes. By far the most fun product I've ever worked on. My product [portfolio](https://www.youtube.com/playlist?list=PLJrFwgrLH5AjcDWy5gIX5YtyDn4-2vzD9) has specifics. * **[Space Fox:](https://spacefox.app/)** A simple app for YouTube creators who want to collaborate with an editor securely. A side project with a dear friend. * **NextRoll (nĆ©e AdRoll):** Product lead for the social integrations and video. * **Climate Corporation:** Product manager for Weather and Scouting Features for Climate Fieldview * **GoodBrackets:** A failed attempt to create a March Madness for Charity tournament * **[Nourish:](http://www.nourish.org/)** A student movement for social entrepreneurship. I scaled the organization from a regional one to a national one. Sadly, after 19 years, the organization is no longer active. My domain is a homage to the producer J Dilla, whose work I love and whose moniker sounds like my name. _"Do it now. The conditions are always impossible." ā Doris Lessig_ # Bookshelf jdilla | books i love [Homeš”](/) [Aboutā¹ļø](/about) [Blogš](/blog) [Booksš](/bookshelf) [Searchš](/search) [Subscribeš¬](/subscribe) # Bookshelf I've organized my bookshelf by the importance of the book to me. ## My pile What I'm currently reading. * The Timeless Way of Building by Christopher Alexander * Stalin Volume 1 by Stephen Kotkin * Confessions by Augustine of Hippo * William Tecumseh Sherman: In the Service of My Country: A Life by James Lee McDonough ## So good it hurts Books that gave me that feeling of regret that I'll never be able to read them for the first time again. I'd suggest these even if you have no interest in the subject. * 1984 by George Orwell * Hamlet by William Shakespeare * The Just City by Jo Walton * Breaks of the Game by David Halberstam * City of Thieves by David Benioff ## I love this book So good it hurts for me, but I'm not sure it will for you. * East of Eden by John Steinbeck * Catch-22 by Joseph Heller * Team of Rivals by Doris Kearns Goodwin * The Great Gatsby by F. Scott Fitzgerald * Death and Life of Great American Cities by Jane Jacobs * What It Takes by Richard Ben Cramer * Middlemarch by George Eliot * The Power Broker by Robert Caro ## This book has stuck with me Books that have had a major impact on how I think about the world. The quality of writing in this category is less important than the impact of the ideas on my life. * Range by David Epstein * Defying Hitler: The Germans Who Resisted Nazi Rule by Gordan Thomas and Greg Lewis * Manās Search for Meaning by Viktor Frankl * Innovation and Entrepreneurship by Peter Drucker * Where Good Ideas Come From by Steven Johnson ## Really freaking good. Great writing, just not quite _so good it hurts_ . * Pereira Maintains by Antonio Tabucchi * The Road by Cormac McCarthy * A Perfect Spy by John le CarrĆ© * Just Mercy by Bryan Stevenson * The Sun Also Rises by Ernest Hemingway * The Glory and the Dream by William Manchester ## Honorable mentions Almost but not quite. * Decisive by Chip Heath and Dan Heath * Made to Stick by Chip Heath and Dan Heath * Switch by Chip Heath and Dan Heath * Huckleberry Finn by Mark Twain * Hild by Nicola Griffith * The Inner Game of Tennis by W. Timothy Galloway _"Chips on shoulders put chips in pockets" ā Josh Wolfe_ # Great Reads jdilla | articles worth your time [Homeš”](/) [Aboutā¹ļø](/about) [Blogš](/blog) [Booksš](/bookshelf) [Searchš](/search) [Subscribeš¬](/subscribe) # Great Reads A collection of writing I find myself returning to periodically for inspiration. Not in any particular order. * [Creativity in Management](https://jamesclear.com/great-speeches/creativity-in-management-by-john-cleese) by John Cleese * [How Will You Measure Your Life?](https://hbr.org/2010/07/how-will-you-measure-your-life) by Clayton Christensen * [How to Do Great Work](http://www.paulgraham.com/greatwork.html) by Paul Graham * [Having Kids](http://www.paulgraham.com/kids.html) by Paul Graham * [You and Your Research](https://www.cs.virginia.edu/~robins/YouAndYourResearch.html) by Richard Hamming * [YCombinator's essential start up advice](https://www.ycombinator.com/library/4D-yc-s-essential-startup-advice)by Geoff Ralston, Michael Seibel * [This is How Apple Rolls](https://www.macworld.com/article/205387/apple-rolls.html) by John Gruber * [A relatively small amount of force applied at just the right place](https://www.tonykulesa.com/p/a-relatively-small-amount-of-force) by Tony Kulesa * [The Mystery of Trust](https://comment.org/the-mystery-of-trust/) by Amanda Ripley * [Three Types of Meetings](https://camdaigle.com/posts/three-types-of-meetings/) by Cam Daigle * [Welcome to the Age of Overparenting](https://www.bostonmagazine.com/2011/11/29/the-age-of-overparenting/) by Katherine Ozment * [How to do things if you're not that smart and don't have any talent](https://adaobi.substack.com/p/how-to-do-things-if-youre-not-that) by Adaobi Adibe * [You can't tell people anything](http://habitatchronicles.com/2004/04/you-cant-tell-people-anything/) by Chip Morningstar * [How to be successful](https://blog.samaltman.com/how-to-be-successful) by Sam Altman * [Lies we tell ourselves about focus](https://also.roybahat.com/lies-we-tell-ourselves-about-focus-75e59e88072d) by Roh Bahat * [The Universal Advice](https://www.youtube.com/watch?v=ojjzXyQCzso&t=330s) by 3Blue1Brown * [What our kids see](https://also.roybahat.com/what-our-kids-see-7dc650be8a90)by Roy Bahat * [Startups and shame](https://also.roybahat.com/startups-and-shame-a9ea52b5976b)by Roy Bahat * [How to Start Google](https://paulgraham.com/google.html) by Paul Graham * [How to get Startup Ideas](https://paulgraham.com/startupideas.html) by Paul Graham * [A thread on consumer social start ups](https://x.com/nikitabier/status/1481118406749220868)by Nikita Bier * [The Zombocom Problem](https://newsletter.squishy.computer/p/the-zombocom-problem) by Gordon Brander * [Divine Discontent and the Eternal Pursuit of Unhappiness](https://snowymonkey.blogspot.com/2013/06/ogilvy-divine-discontent-and-eternal.html) by David Ogilvy * [What I Wish Someone Had Told Me](https://blog.samaltman.com/what-i-wish-someone-had-told-me) by Sam Altman * [Productivity](https://zhengdongwang.com/productivity.html) by Zheng Dong Wang * [Principles](https://nabeelqu.substack.com/p/principles) by Nabeel Qureshi * [Selling](https://nabeelqu.co/post-selling) by Nabeel Qureshi * [Do not end the week with nothing](https://training.kalzumeus.com/newsletters/archive/do-not-end-the-week-with-nothing) by Patrick McKenzie * [Designing the ideal bootstrapped business](https://mtlynch.io/notes/designing-the-ideal-bootstrapped-business/) by Michael Lynch * [A practical guide to executive presence](https://blog.staysaasy.com/p/a-practical-guide-to-executive-presence) by Stay Sassy * [There is no speed limit](https://sive.rs/kimo) by Derek Sivers * [The Layers](https://www.poetryfoundation.org/poems/54897/the-layers) by Stanley Kunitz * [18 mistakes that kill startups](https://paulgraham.com/startupmistakes.html) by Paul Graham * [Sales Safari](https://www.youtube.com/watch?v=67JVkG4dpj4) by Amy Hoy * [Six lessons about agency I learned working at an art gallery](https://www.henrikkarlsson.xyz/p/art-gallery) by Henrik Karlsson * [How to increase your surface area for luck](https://usefulfictions.substack.com/p/how-to-increase-your-surface-area) by Cate Hall * [How to succeed at Mr. Beast Productions](https://drive.google.com/file/d/1iiKi4BSjnAkViTJfCs4V6Xa5t-0dE6eo/view) by Mr. Beast _"When in doubt, ignore your opponent" ā Chinese Proverb_
