Assorted links #41

1440. The Seinfeld Chronicles: Digital Edition. A data-driven exploration of Seinfeld (my favourite sitcom from the 90s). A lot of great interactive visualisations, including audio and video, so do not expect to scroll through this feature in a few minutes. I did not care much for the audio presentation of the episodes, but it is an impressive amount of personally collected (pre-LLM) data. A must-read if you are into data visualisations or Seinfeld or, preferably, both. You can find a blog post about the project here.

1441. World Wide Words. More than 3,000 articles on aspects of English by the etymologist Michael Quinion. When you browse the index and pick any letter, it is almost impossible not to find something interesting. The site has not been updated for the past few years, and it looks like Michael Quinion has not written any new entries for almost a decade, but it is a great resource.

1442. The Secret to Winning on Jeopardy. I recently watched some old episodes of Jeopardy (episodes older than me, which is, alas, quite old), and I was surprised by how timeless and general the answers were. This article covers what it takes to actually be good at Jeopardy. In brief, you have to specialise in being a superficial generalist: "To win on Jeopardy, you don’t need to learn everything. You just need to learn one thing about everything."

1443. The Sparkling Water Report. A blind test of sparkling water. In the blind test, there is a positive correlation between price and overall score (r = .24), but keep in mind the low N. Always happy to see San Pellegrino doing well. I guess I find this post interesting because I have not thought much, if at all, about the quality of sparkling water.

1444. The Andouillette Does Not Travel. This is very true: "A layered, folded, malodorous sausage that inspires devotion or revulsion - and nothing else - the andouillette is made from pork intestines and tripe (hopefully) thoroughly washed and scrubbed before being boiled in a court bouillon, spiked with champagne, pepper and nutmeg and encased in a final - disconcertingly thick - layer of intestinal passage. It can be heavenly. It can be horrifying. Again, nothing in between. There is no such thing as an ‘indifferent’ andouillette." I am not a fan.

1445. How the Words We Teach English Language Learners Changed. Death, taxes, and plugging pudding dot cool. A great article on "essential vocabulary" lists with a focus on the changes from 1953 to 2023. As always, it is best to scroll straight to the section on methods and notes, get a sense of what is going on, and then go through the content.

1446. fivethirtyeightindex. FiveThirtyEight is no more (RIP), but admirable efforts have been put into archiving the content for the future. The material includes 166 datasets related to various topics.

1447. Chili Peppers of the World. An informative overview of chili peppers with illustrations of 176 different peppers. See also Hot Sauces of the World from a couple of years ago.

1448. The Film Experience. A course on movie analysis and criticism with lectures available for free. If you want to study films and increase the likelihood of actually watching good ones, the best heuristic is to go with the Criterion Collection (do check out the great resource The Closet).

1449. Antidote to Slop: An Archive of Life-Changing Material. A very clickbaity title and, let's be honest, no material you will encounter today will be life-changing, but a reasonably good collection of links to various articles and videos.

1450. Journal of Geek Studies. A collection of articles going back more than ten years, covering topics such as the Infinite Fish Playing Pokémon Theorem (funny to read in a time when LLMs play Pokémon), the birds of James Bond, the plants of J.R.R. Tolkien’s legendarium, the influence of Studio Ghibli in zoological nomenclature, the personality and behaviour of Thompson and Thomson, the science of Monkey Island, etc.

1451. Early Web Links. If you are feeling nostalgic about the internet of the 90s, look no further. From the description on the page: "Before the internet went corporate, it was weird, wonderful, and deeply personal. People built websites just because they had something to share. This directory is full of sites that bring back that feeling."

1452. Calvin and Hobbes and the Price of Integrity. On how not to be merchandised. Bill Watterson authored Calvin and Hobbes over a period of ten years and stopped at the age of 37. There is something truly inspiring about people not caring about financial incentives beyond a certain point (a huge privilege, I know). I have given away hundreds of books over the years, but if there is one book (or three books) I have never even thought about getting rid of, it is The Complete Calvin and Hobbes. I wrote a post (in Danish) about my love for the comic strip back in 2011.

1453. Lifts in Film. A collection of more than 500 elevator scenes. Uplifting material. Think of any good (or bad) elevator scene you have watched and chances are that it will be included here.

1454. The absurdly optimized pancake. A pancake recipe. I would like to see more recipes structured like this. That is, if not a brief, straight-to-the-point description, then a deep dive into all aspects of what you are cooking (from the history to the chemistry). From the summary: "A systematic investigation of acid-base neutralization, CO₂ production kinetics, gluten inhibition, and the Maillard reaction as applied to a 125-gram flour batter, with an interactive stoichiometric calculator that adapts to whatever is in your refrigerator".

1455. What We Get Wrong About How 9/11 Began. A depiction of the 16 minutes and 31 seconds between the first and second planes hitting the World Trade Center. An interesting read on how to make sense of the event before it became the event, especially given that a lot of people do not remember the event (and in particular the immediate minutes before the second plane hit the tower). One interesting fact in the article that makes me feel(?) old is that around 40 percent of Americans are too young to remember the events of 9/11. It is also yet another reminder of how time is moving fast. Like really fast. It was 25 years ago, and it feels like just a year or two ago that I watched the brilliant documentary series, 9/11: One Day in America, but I guess it is now five years ago. The best documentary about the World Trade Center is still the episode from New York: A Documentary Film, The Center of the World. I also watched the new documentary 102 Minutes: Inside the Towers, which I cannot recommend (the main issue is that the documentary is too long - 102 minutes). Some of the best pictures from the day are from Bill Biggart.

1456. The Making of Indian Statistics. On the history of statistics in India, from the first census to today's data challenges. I like the framing of public statistics as the "infrastructure of national self-knowledge". The article is also presented in a beautiful layout and good use of notes for sources. See also Designing a Summer Ice Cream for India.

1457. 100 years /100 people. A campaign from 66°North with pictures of 100 people from Iceland (from age 1 to 100). I hope they will stay in touch with these people and maybe in, say, ten years, do a follow-up photo shoot.

1458. The library level: Archives in videogames. An article from the National Library of Scotland on the most intriguing libraries to feature in video games. I would have appreciated more of a library of libraries in video games, but there are good examples throughout this article.

1459. How to Visit Every European Microstate. Another suggestion for a road trip within Europe, namely to visit all European microstates (Vatican City, Monaco, San Marino, Liechtenstein, Malta, and Andorra). Notably, if you go on such a road trip, you will - unsurprisingly - see a lot of other places as well.

1460. Fairly Ranking the Most Brilliant Birds. Be prepared to look at pictures of beautiful birds with brilliant colours. A lot of good reflections on how to (fairly) rank birds. I always find it more interesting to discuss methods than results (e.g., rather than discussing favourite movies, it is more interesting to discuss how to go about identifying favourite movies), and I enjoyed the discussions of the various statistical properties that matter here. Also some interesting bird facts, such as "Birds are very good at being red and yellow, due to the carotenoids that some birds transfer to their feathers from their diets."

1461. Shortverse. A collection of short films. Not all films are available (or, I guess they are if you use a VPN), but there should be something for everyone here.

1462. 250 for 250 Booklist. To celebrate the United States' 250th anniversary, the Brooklyn Public Library has assembled a list of the 250 most influential books in United States history. There is a lot of variation on the list (including 16 graphic novels), and while there are many books I like on the list, you can easily question whether they are among the most influential books in the history of the United States.

1463. The Naturalist’s Library. From the description: "A complete digital edition of William Jardine’s 40-volume series on natural history published 1833–1843, including more than 1,300 restored hand-drawn illustrations of birds, mammals, insects, and fish." This is a comprehensive resource (see, e.g., Dogs: Part 1). For a detailed introduction to the project, check out this blog post.

1464. The quiet grief of adult friendship. A good essay on why it is difficult to make and maintain friendships as an adult.

1465. Understanding large values: It's our ethical duty. Some good advice on how to grasp the magnitude of large numbers: convert, visualise, personalise, and experience.

1466. ikea complexity index. Complexity ratings of more than 20k IKEA products. Easy to filter by category and sort by price, manual pages, etc. If you are curious about why the products have the names they have, check out How does IKEA come up with names for its products?. And for additional fun material on building instructions, check out LEGO® building instructions through time.

1467. David Attenborough’s life’s work, searchable. David Attenborough recently turned 100. Here is a great tool to explore more than 70 years of different episodes and shows (from Zoo Quest in 1954 to A Gorilla Story: Told by David Attenborough in 2026).

1468. The idea of a difficulty curve is all wrong. On how to think about difficulty curves in video games. Do also check out the posts on how to build levels at the perfect difficulty, scores in games, and level select screens.

1469. Tuxedo No. 2. A cocktail companion. More than 200 different cocktails published over a period of more than ten years. There is a good overview of all cocktails on the page as well as an ingredients page, not to mention that there are photos of all cocktails.

1470. The History of the Bloomberg Terminal. A deep dive into the primary interface used by traders, the Bloomberg Terminal. It is interesting to see how finance-specific hot keys were designed for traders, and how many innovations were introduced over the years (such as biometric authentication).

1471. The lamps in my house. If you are looking for new lamps for your place, this is a strong place to look for inspiration. As should be mandatory in these kinds of posts, great to see some Danish design as well.

1472. The 100 Best TV Shows of the 21st Century. A ranking of the best American and British TV shows of the 21st century (so far). I have watched 72 of the shows on the list, and my view is that the list seems pretty good and non-controversial (the top five are Breaking Bad, The Wire, Mad Men, Succession, and Fleabag, which seems fair). Of course, you can easily find people online who find the list bizarre (see, e.g., here and check out the quote shares), but that is always the case with lists like this one. In a similar vein, do also check out The 50 Most Underappreciated Shows of the 21st Century (where you will find good entries such as Colin from Accounts and It's a Sin).

1473. How I over-engineered my book. I like the idea of having a CI pipeline for writing a book. One issue with my R book is that I cannot be sure how much of the content will still work once there is a new version of R, but a CI pipeline could help with this. There are several good ideas in this post, and I don't think anybody should try to replicate the exact setup (unless they want to). However, if you are working on a technical book, it is definitely worth checking out for inspiration. See also How I Under-Engineered my Book.

1474. Temperature Zero for Culture: Why Everything Is Starting to Look the Same. I do not agree with the main argument presented here. "The more the machines learn about us as individuals, the more alike our worlds look." It is an interesting thesis (hence, I share it here), but I do not agree with it and I do not find any of the empirical evidence convincing. I find the argumentation speculative at best and it presents a very simplistic view of recommender systems (the assumption that modern recommender systems do not already have some idea of exploration v. exploitation is fun, though). There is a certain irony to the fact that this post stands out and very few people will read it, and - if anything - the existence of this post is in itself evidence against the thesis being presented. Interestingly, the post is about how everything is starting to look the same but the post looks at only a few examples and a single time-series graph (with one trend going against 'sameness' accompanied by a post hoc interpretation). You cannot say anything about everything starting to look the same by exploring variation within London (and not only within London, but the Greater London Authority), and it is easy to see how you are conditioning on a collider (if you have a street inside London with no Pret, chances are that the offerings on that street will be more unique). These issues are even more prominent when you look at culture (e.g., Billboard Hot 100 hits), and looking at movies does not say anything about changes over time but just that the popularity of movies follows a Pareto distribution (and the audience for movies can easily have changed over time due to streaming, TikTok, etc., in ways that do not align with the thesis). My theory is that this post resonates with some people because there is a feeling of things looking the same, but this is mostly due to selecting on the dependent variable within a category. More generally, while it is interesting to say that it is all about the ‘conditional mean’, there is clearly too much focus on the mean rather than the conditional (the more data there is, the more this mean will be conditional upon personal characteristics, and the less alike our worlds will look). Well, again, and as should be clear by now, I very much disagree with the post and I do not see how any of the data shows that "everything is starting to look the same", but I find the topic interesting and it is worth reading and engaging with the post.

1475. Removed Substack link. I found an interesting Substack post that I initially added here. I checked the post again after a few weeks, but by then it was behind a paywall. There was nothing in the post indicating that the content would be put behind a paywall at some point. Alas, it is a pattern I see more and more, and I truly dislike everything about Substack. I will keep this as a reminder not to share Substack links if possible. I don't mind paywalls (people gotta pay the bills and I can easily use archive.today, archive.org, etc.), but I am against the way people writing on Substack work with paywalls.

1476. 50 Places in London I Would Like to Live. Angus Colwell's list of 50 places in London where he would like to live (on Substack, but not behind a paywall). A funny reminder of how great London is (especially a must-read if you are into architecture). Yet, my personal list of places in London where I would like to live is closer to 0 than 50.

1477. Quantifying Colour. A great article on how to move beyond labels and work with numbers when dealing with colours, and not only as colour codes but as physical entities that can be used for calculations.

1478. The Design of Chess. I have seen several pages over the years going through the various designs of chess pieces over time, but I could not find any I have actually shared here. This is one good page with a timeline of the design of chess pieces (including a good list of references and sources).

1479. Thinking in Python. This is a good, technical Python book. Definitely a recommended resource if you want to improve your general Python skills. I was not sure whether to share it here due to this description: "I know some people don’t like AI. Without it, this book wouldn’t exist. The book is free, so if AI bothers you more than the book might benefit you, please ignore the book." As I have said before, the problem with using AI is the lack of transparency, and I appreciate the honesty. I guess I am less against AI writing in the context of technical writing where I mostly look at the code examples anyway. Most importantly, there is nothing in the writing here that indicates lazy AI writing (I could even see some mistakes that indicate human writing). Alas, a lot of writing on technical matters these days is AI slop (most recently Systems Thinking for Agentic AI, which was one page of AI slop after another).