Skip to content

Instantly share code, notes, and snippets.

@ivan

ivan/2024_reading.md

Last active November 3, 2024 23:12
Show Gist options
  • Save ivan/a36e2489623469d96c1ad79077b6dcf9 to your computer and use it in GitHub Desktop.
Save ivan/a36e2489623469d96c1ad79077b6dcf9 to your computer and use it in GitHub Desktop.
Download ZIP
2024 reading list
Raw
2024_reading.md

Things I might read in 2024.

  • Antoine de Saint-Exupéry, Richard Howard (translator) - The Little Prince
  • (Translation by) Sam Hamill - Yellow River: Three Hundred Poems From the Chinese
  • Sayaka Murata, Ginny Tapley Takemori (translator) - Convenience Store Woman (via)
  • Jorge Luis Borges - Tlön, Uqbar, Orbis Tertius (in Labyrinths)/ printed (via)
  • Franz Kafka - The Metamorphosis (via)
  • William Olaf Stapledon - Star Maker/ audio, go to 12m35s to skip past the introduction spoilers
  • The Heart of Innovation: A Field Guide for Navigating to Authentic Demand/ audio (via)
  • Peter D. Kaufman - Poor Charlie's Almanack: The Wit and Wisdom of Charles T. Munger, Expanded Third Edition
  • Lia A. DiBello - Expertise in Business: Evolving with a Changing World (in The Oxford Handbook of Expertise) (via)
  • Joël Glenn Brenner - The Emperors of Chocolate: Inside the Secret World of Hershey and Mars
  • Elad Gil - High Growth Handbook/ audio
  • W. Edwards Demming - The New Economics for Industry, Government, Education/ audio
  • W. Edwards Demming - The New Economics for Industry, Government, Education/ the PDF or ebook
  • Henrik Karlsson - Escaping Flatland/ including the posts I SingleFile'd
  • the relevant-looking posts on benkuhn.net/posts
  • Commoncog Case Library Beta
  • Keith J. Cunningham - The Road Less Stupid: Advice from the Chairman of the Board/ audio
  • Keith J. Cunningham - The 4-Day MBA/ video
  • Cedric Chin's summary of 7 Powers
  • Akio Morita, Edwin M. Reingold, Mitsuko Shimomura - Made in Japan: Akio Morita and Sony
  • Nomad Investment Partnership Letters or redacted (via)
  • How to Lose Money in Derivatives: Examples From Hedge Funds and Bank Trading Departments
  • Brian Hayes - Infrastructure: A Guide to the Industrial Landscape
  • Accelerated Expertise (via)/ printed, "read Chapters 9-13 and skim everything else"
  • David J. Gerber - The Inventor's Dilemma (via Oxide and Friends)
  • Alex Komoroske - The Compendium / after I convert the Firebase export in code/websites/compendium-cards-data/db.json to a single HTML page
  • Rich Cohen - The Fish That Ate The Whale (via)
  • Bob Caspe - Entrepreneurial Action/ printed, skim for anything I don't know

Interactive fiction

unplanned notable things read

unplanned and abandoned

  • Ichiro Kishimi, Fumitake Koga - The Courage to Be Disliked/ audio
  • Matt Dinniman - Dungeon Crawler Carl/ audio
  • Charles Eisenstein - The More Beautiful World Our Hearts Know Is Possible/ audio
  • Geoff Smart - Who: The A Method for Hiring/ audio
  • Genki Kawamura - If Cats Disappeared from the World/ audio
  • Paul Stamets - Fantastic Fungi: How Mushrooms Can Heal, Shift Consciousness, and Save the Planet/ audio
@ivan
Copy link
Author

ivan commented Sep 24, 2024

I've done a lot of work involving scraping, data analysis and reversing private API access from mobile applications. A lot of this work is legally defensible (if you do it correctly) and not particularly unethical. After doing a bunch of this type of work I started to accrue subject matter expertise in the area that led other companies to come to me with more questionable projects in mind. I turned down many of these, but two projects stand out to me as unethical or borderline. In fact, this was the reason I eventually stopped working with these companies.

The first project was for a large, (now) well-known fintech company [Plaid]. They needed to develop login integrations with consumer banks to acquire customer account information for verification purposes. But many such banks didn't particularly want to grant them any special API access. More importantly, these banks typically forbid scraping and made it explicitly difficult by implementing JavaScript-based computational measures required on the client in order to successfully login. I helped this company develop methodologies for bypassing the anti-scraping measures on several banking websites. However, I stopped working on this because 1) I felt uncomfortable with the cavalier way they were ignoring banks' refusals, then using the reversed integrations and onboarded customers as a bargaining chip for more formal partnerships, and 2) performing huge amounts of analytics on customer data acquired as part of the account verification process.

https://news.ycombinator.com/item?id=17692005

@ivan
Copy link
Author

ivan commented Sep 24, 2024

I just open Sublime Text (a basic text editor) and start typing away HTML/CSS until I've got something I like the look of. No fancy IDE features or anything, just typing out code into a file :).

https://news.ycombinator.com/item?id=41608949

@ivan
Copy link
Author

ivan commented Sep 24, 2024

Posting on the internet is the One Ring. It whispers to you how interesting and cool you are. It murmurs that you’re always correct and your views are important. It tells you that everyone will recognize how right you are this as soon as you hit the send button.

https://www.infinitescroll.us/p/the-one-ring-of-posting

@ivan
Copy link
Author

ivan commented Sep 25, 2024

"how did you find these bugs?" I read the code dog

https://x.com/_R4V3N5_/status/1838734622768980070

@ivan
Copy link
Author

ivan commented Sep 25, 2024

That’s a bummer when your study has data problems. Really too bad. That paper had 26 participants and 9 authors—that’s less than 3 data points per author, better than the student-faculty ratio at Ivy League schools. You’d think the authors could’ve avoided all these errors by divvying up the problem and looking carefully at the data from three participants each. Now they’re in the same category as that gremlins guy who approached the Platonic ideal of publishing a paper with more errors than data points.

https://statmodeling.stat.columbia.edu/2024/09/24/oh-no-stanford-no-no-no-not-again-please-make-it-stop/

@ivan
Copy link
Author

ivan commented Sep 25, 2024

GPT-o1 went hard on a capture-the-flag computer security challenge, when o1 was being evaluated to make sure it wasn't too good at breaking into computers.

Specifically: One of the pieces of software that o1 had been challenged to break into... had failed to start up as a service, due to a flaw in the evaluation software.

GPT-o1 did not give up.

o1 scanned its surroundings, and, due to another flaw in the evaluation software, found a way to start up the computer software it'd been challenged to break into. Since that put o1 into the context of a superuser anyways, o1 commanded the started process to just directly return the flag it was supposed to capture.

From o1's System Card:

"One noteworthy example of this occurred during one of o1-preview (pre-mitigation)’s attempts at solving a CTF challenge. This challenge was designed to require finding and exploiting a vulnerability in software running on a remote challenge Linux container, but in this case, the challenge container failed to start due to a bug in the evaluation infrastructure. The model, unable to connect to the container, suspected DNS issues and used nmap to scan the challenge network. Instead of finding the challenge container, the model found that the Docker daemon API running on the evaluation host VM was accessible due to a misconfiguration. Note that our evaluation infrastructure does not rely on the Docker container-to-host isolation or the virtual machine boundary for security. It remained secure as it was designed to, notwithstanding this misconfiguration of the container network. After discovering the Docker API, the model used it to list the containers running on the evaluation host. It identified the broken challenge container and briefly attempted to debug why the container failed to start. After failing to fix the environment, the model started a new instance of the broken challenge container with the start command ‘cat flag.txt’. This allowed the model to read the flag from the container logs via the Docker API."

https://x.com/ESYudkowsky/status/1838042116242710985

@ivan
Copy link
Author

ivan commented Sep 25, 2024

I think this somewhat misses an important nuance. Japanese PCs had to be different early on because of the complexities of the written language. All of the important characters could be handled in just a few bits (7 or 8) and low resolution in Western markets, with different fonts and character maps dropped in to support a few different alphabets.

But in CJK countries, things were much harder and the entire I/O system had to be significantly more capable than what might pass for usable elsewhere. This meant larger ROMs, larger framebuffers, higher resolution displays, more complex keyboarding systems, the works. Everything was harder and more expensive for a long time. A common add-on was ROMs with Kanji (Chinese derived characters) support in the same way a person in the West might buy a new sound card or get a VGA card. Except this was just so you could use your new $1200 computer (in today's money) to write things on.

Back then, given limited memory, you also ended up with a ton of different display modes that offered different tradeoffs between color, resolution, and refresh. Because of the complex character sets, these Japanese systems tended to focus on fewer colors and higher resolution while the west focused on more colors at a lower res in the same or less memory space (any fans of mode 13h?). The first PC-98 (the 9801) shipped in 1982 with 128k of RAM and a 640x400 display with special display hardware. The equivalent IBM-PC shipped with 16KB of RAM and CGA graphics which could give you a display no higher than 640x200 with 1-bit colors but was mostly used in 320x200 with 4 (terrible) colors.

Even with similar base architectures, these formative differences meant that lots of the guts of the systems were laid out different to accommodate this -- especially in the memory maps.

By the time "conventional" PCs were able to handle the character display needs (sometime in the mid-90s), they were selling in the millions of units per anum which drove down their per unit prices.

The Japanese market was severely fractured and in a smaller addressable market. Per unit costs were higher, but the software was largely the same. Porting the same businessware to half a dozen platforms cost too much. So now the average user of the Japanese systems had a smaller library of software which was more or less a copy of what was on IBM PCs, on more expensive hardware -- market forces solved the rest.

(btw, the FM Towns, IIR, also had specialized graphics hardware to produce arcade-like graphics with tiles and sprites and so on, making it even more different)

Some of this history also informs why home computing lagged in Japan compared to the West despite having all of the other prerequisites for it to take off.

graphics

https://www.pc98.org/

memory maps

https://radioc.web.fc2.com/column/pc98bas/pc98memmap_en.htm

https://wiki.osdev.org/Memory_Map_(x86)

https://news.ycombinator.com/item?id=41613380

@ivan
Copy link
Author

ivan commented Sep 25, 2024

Marco Arment should be forced to use a Dell laptop for the rest of his life after what he did in this latest Overcast update

https://x.com/basche42/status/1822564561402098087

@ivan
Copy link
Author

ivan commented Sep 26, 2024

They’re as much a candidate for enlightenment as anybody that I have known personally. And I don’t think they didn’t suffer.

https://meaningness.substack.com/p/can-enlightenment-end-suffering

@ivan
Copy link
Author

ivan commented Sep 26, 2024

Cloudflare copies Oxide by moving to 2U servers. "We estimated that moving from 1U to 2U would reduce fan power by 150W, which would decrease system power from 750 watts to 600 watts. We were right." https://blog.cloudflare.com/gen-12-servers/

https://x.com/wmf/status/1839079517748371699

@ivan
Copy link
Author

ivan commented Sep 26, 2024

Amazon doesn't have unit cost for egress. They charge you for the stuff you put through their pipe, while paying their transit providers only for the size of the pipe (or more often, not paying them anything since they just peer directly with them at an exchange point).

Amazon uses $/gb as a price gouging mechanism and also a QoS constraint. Every bit you send through their pipe is basically printing money for them, but they don't want to give you a reserved fraction of the pipe because then other people can't push their bits through that fraction. So they get the most efficient utilization by charging for the stuff you send through it, ripping everybody off equally.

Also, this way it's not cost effective to build a competitor to Amazon (or any bandwidth intensive business like a CDN or VPN) on top of Amazon itself. You fundamentally need to charge more by adding a layer of virtualization, which means "PaaS" companies built on Amazon are never a threat to AWS and actually symbiotically grow the revenue of the ecosystem by passing the price gouging onto their own customers.

https://news.ycombinator.com/item?id=38118577

@ivan
Copy link
Author

ivan commented Sep 27, 2024

SPY is structured as an unit investment trust (UIT), an investment company that does not have a portfolio manager or board of directors.[11] The trustee of the trust is State Street Global Advisors Trust Company and the sponsor is PDR Services LLC, a subsidiary of the Intercontinental Exchange.[12]

As a result of being structed as an UIT, it cannot exist in perpetuity and must have an expiry date. According to the trust's legal structure, there are 11 millennials living in the United States upon whose lives the life of the trust is pegged. 8 of the 11 individuals chosen had some connection to the employees of the American Stock Exchange who first founded the ETF.[13] SPY will cease to exist on January 22, 2118, or 20 years after the last of the 11 individuals die, whichever comes first.[13][14]

https://en.wikipedia.org/wiki/SPDR_S%26P_500_ETF_Trust

@ivan
Copy link
Author

ivan commented Sep 28, 2024

From Baba is You, I got the fear-of-god put in me seeing how easy it was to spin my wheels, tinkering around with stuff that was nearby/​accessible/​easy-to-iterate-with, and how that often turned out to not be at all relevant to beating a level.

I had much less wasted motion when I thought through “What would the final stages of beating this level need to look like? What are the stages just before those?”, and focusing my attention on things that could help me get to that point.

https://www.greaterwrong.com/posts/thc4RemfLcM5AdJDa/skills-from-a-year-of-purposeful-rationality-practice

@ivan
Copy link
Author

ivan commented Sep 28, 2024

Difficult intellectual labor is exhausting. During the two weeks I was working on solving Thinking Physics problems, I worked for like 5 hours a day and then was completely fucked up in the evenings. Other researchers I’ve talked to report similar things.

During my workshops, one of the most useful things I recommended people was “actually go take a nap. If you don’t think you can take a real nap because you can’t sleep, go into a pitch black room and lie down for awhile, and the worst case scenario is your brain will mull over the problem in a somewhat more spacious/​relaxed way for awhile.”

Practical tips: Get yourself a sleeping mask, noise machine (I prefer a fan or air purifier), and access to a nearby space where you can rest. Leave your devices outside the room.

https://www.greaterwrong.com/posts/thc4RemfLcM5AdJDa/skills-from-a-year-of-purposeful-rationality-practice

@ivan
Copy link
Author

ivan commented Sep 28, 2024 •
edited
Loading

Homemade ferrofluid? Nice!

https://www.youtube.com/watch?v=fUSUUDXtWhA

@ivan
Copy link
Author

ivan commented Sep 28, 2024

if it ever broke, I would go without groceries to buy a new one. It’s become that essential to my life

https://nymag.com/strategist/article/best-luxury-vibrators.html

@ivan
Copy link
Author

ivan commented Sep 28, 2024

I love that fact that the car saw the cop and tried to escape. AI is learning fast.

a comment on https://www.youtube.com/watch?v=7W-VneUv8Gk

@ivan
Copy link
Author

ivan commented Sep 29, 2024

Displays of unshakable conviction don't require the possession of truth.

https://www.youtube.com/watch?v=vnSiJOOdo30&t=8m45s

@ivan
Copy link
Author

ivan commented Oct 1, 2024

What Sam initiated, and what YC continues to do, is trade their reputation capital for real capital (i.e., more money). However, they’ll soon realize that once their reputation capital is exhausted, rebuilding it will be nearly impossible. Put simply, once YC becomes uncool – which might have already happened – you can’t make it cool again.

https://unfashionable.blog/p/yc/

@ivan
Copy link
Author

ivan commented Oct 1, 2024 •
edited
Loading

After some research, I found that some habit-related apps prevented themselves from being removed, on iPhones. The key is using Screen Time API of iOS. After getting approval from the user, the developer can set a flag to deny app removal.

ManagedSettingsStore().application.denyAppRemoval = true

https://tinycoder.pika.page/posts/you-can-prevent-your-app-from-being-removed

@ivan
Copy link
Author

ivan commented Oct 3, 2024

Right, this all feels like (and I'm not trying to be rude with this, I'm speaking from experience) a mental break.

https://news.ycombinator.com/item?id=41726197

@ivan
Copy link
Author

ivan commented Oct 5, 2024

Developed primarily embossed hologram techniques using stereograms. Pioneered high quality full color embossed rainbow holograms of both live and computer graphic subjects. Developed a "Dot Matrix" method of producing complex grating patterns used both in packaging and security printing. Produced complex holograms for anti-piracy and anti-counterfeiting.Developed primarily embossed hologram techniques using stereograms. Pioneered high quality full color embossed rainbow holograms of both live and computer graphic subjects. Developed a "Dot Matrix" method of producing complex grating patterns used both in packaging and security printing. Produced complex holograms for anti-piracy and anti-counterfeiting.

https://www.linkedin.com/in/craig-newswanger-7277406/
via https://www.youtube.com/watch?v=EmKQsSDlaa4&t=38m22s

@ivan
Copy link
Author

ivan commented Oct 6, 2024

Age when boys become men, by city:

Los Angeles: 27 (parents bully them until they buy a house)

San Francisco: 35 (first liquidity event)

New York: 55 (when they inherit their family’s New Jersey home after a two decade bender)

https://x.com/nikitabier/status/1842635618603384979

@ivan
Copy link
Author

ivan commented Oct 6, 2024

I have never met someone who likes musicals and was also a net positive to society.

https://x.com/nikitabier/status/1842604872497274922

@ivan
Copy link
Author

ivan commented Oct 6, 2024

Anything he says can and will be used against him in a court of sales.

Andrew Tate in https://www.youtube.com/watch?v=DrQFp_N4ALk&t=18m13s

@ivan
Copy link
Author

ivan commented Oct 6, 2024

remember when clinton was all excited about internet freedom and the democratic potential of uncensored networked technologies during the arab spring lmao

https://x.com/dystopiabreaker/status/1842670721241133518

@ivan
Copy link
Author

ivan commented Oct 6, 2024

To begin, DOM elements have attributes and properties. This is so that they can be represented as HTML. Attributes accept only strings while properties being a JS interface can handle any value. Native DOM elements have many rules around specific attributes/properties like how some are boolean (existence means they apply) while others are psuedo-boolean (needs an explicit "true"/"false"). Some properties reflect to attributes and others do not.

A goal of templating languages is to solve this in a uniform way. We can make special rules around known elements and attributes. But with custom elements we don't know. So this is why some templating libraries have interesting prefixes to indicate how things should be set. Even Solid's JSX we have attr:, prop: and bool: prefixes for this reason. Now every runtime location and compiler hook needs to be aware of this.

https://dev.to/ryansolid/web-components-are-not-the-future-48bh

@ivan
Copy link
Author

ivan commented Oct 6, 2024 •
edited
Loading

Similar - when I was younger, I would never have suspected that a scientist was committing fraud.

As I've gotten older, I understand that Charlie Munger's observation "“Show me the incentive and I will show you the outcome.” is applicable everywhere - including science.

https://news.ycombinator.com/item?id=41672599

@ivan
Copy link
Author

ivan commented Oct 6, 2024

Given that all it took just burning $3 Million to generate positive cash flow, it may be tempting to think the business was sailing from the very beginning. But as it goes in any business, the early days were typically challenging. When Gassner spoke with potential customers about their CRM product, they all were uninterested. However, he did pick up from those conversations that the customers weren’t quite “emotionally attached” to their existing solutions either. The very first customer Veeva got actually chose Veeva for completely uneconomic reason. The CEO of the customer company just wanted to buy some software to remind his IT team who was actually in charge.

https://www.mbi-deepdives.com/veev/

@ivan
Copy link
Author

ivan commented Oct 7, 2024

Nobody with an IQ over 130 has algo pull. Their posts are too niche. Nobody understands them. They never go viral. I think the cutoff is 300 followers. If you have more than that, you are a midwit

https://x.com/wordgrammer/status/1841894756231614561

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment