Interesting

Collections

The Niche Programmer
Why […‘👩‍❤️‍💋‍👨’] returns [‘👩’, ‘‍’, ‘❤’, ‘️’, ‘‍’, ‘💋’, ‘‍’, ‘👨’] in JavaScript?
Bugs in Hello World
Tailwind CSS Docs Workflow for Alfred
Fastest-ever logic gates could make computers a million times faster
- https://news.ycombinator.com/item?id=31356109
- Photonic computing, to accelerate both logic gates and data transfer, is an incredibly broad and exciting field. While a lot of the promise is still in the lab, real advances are currently being commercialized.
  - https://spie.org/news/photonics-focus/marapr-2022/harnessing...
  - https://www.nextplatform.com/2022/03/17/luminous-shines-a-li...
Quine Relay is a Ruby program that generates Rust program that generates Scala program that generates ...(through 128 languages in total)... REXX program that generates the original Ruby code again.
CSS Snake & Ladders: a multiplayer game developed in HTML+CSS with no JS
Map of the Internet
Ask HN: What is the most impactful thing you've built?
GPS vs GLONASS vs Galileo
An AI generated, never-ending discussion between Werner Herzog and Slavoj ŽIžek
- https://news.ycombinator.com/item?id=33437296
FFmpeg - The Ultimate Guide
Pixelblog - 41 - Isometric Pixel art
Mars Now | Explore – NASA Mars Exploration
The Page With No Code
- https://no-ht.ml/
⚙️ Math Breakdown: Anime Homing Missiles
Image Compression with Singular Value Decomposition
Email explained from first principles
https://pointerpointer.com/
a = 0-x is about 3-10x faster than a = -x
Firecracker internals: a deep dive inside the technology powering AWS Lambda
- https://github1s.com/firecracker-microvm/firecracker
https://github.com/nemutas/helix
Redditor creates working anime QR codes using Stable Diffusion
CIA 2010 covert communication websites
Hacker News Daily Top 10 posts - You can subscribe by watching this repo or via RSS.
MMO Architecture: Source of truth, Dataflows, I/O bottlenecks and how to solve them discusses architecture challenges for massively multiplayer online (MMO) games. Unlike enterprise systems, the source of truth for an MMO's game world is the in-memory state, not the database, as persisting frequent updates would cause bottlenecks. To address this, the article recommends using a data service with the full state cached in memory. It should handle data access through RPCs while persisting only critical updates. This allows for direct, low-latency access while still enabling recovery from failures. Additionally, using techniques like compare-and-swap for writes can help avoid race conditions in the distributed system. Managing what data needs persistence and when is presented as more of an art than science to optimize for recovery needs over business requirements.
The Pudding is a digital publication that is the best internet rabbit hole.
Webgl Fluid Simulation
QWANJI
Flying planes in Microsoft Flight Simulator with a JavaScript autopilot
So you think you know box shadows?

News

How a Rogue Developer Ruined Millions of Software (happened this weekend)
Everything on Amazon is becoming an ad
A Year-End Letter from our Executive Director

Let’s Encrypt surpassed 300,000,000 websites served.
random distribution visual

Web scraping is legal, US appeals court reaffirms

Good news for archivists, academics, researchers and journalists: Scraping publicly accessible data is legal, according to a U.S. appeals court ruling.

https://news.ycombinator.com/item?id=31075396

SymSpell

Spelling correction & Fuzzy search: 1 million times faster through Symmetric Delete spelling correction algorithm

1000x Faster Spelling Correction algorithm

AI

DALL·E 2 is a new AI system that can create realistic images and art from a description in natural language.
DALL-E 2 - Pytorch - Implementation of DALL-E 2, OpenAI's updated text-to-image synthesis neural network, in Pytorch.
- https://news.ycombinator.com/item?id=31228710
carefree-creator - An AI-powered creator for everyone.
Domain Name Search with AI - https://news.ycombinator.com/item?id=33882497
Neural Networks: Zero to Hero #bookmark
- https://github.com/karpathy/nanoGPT
Use GPT-3 incorrectly: reduce costs 40x and increase speed by 5x
- To reduce costs and increase speed, they developed a technique to generate a moderately sized corpus of completions made by a larger model, and fine-tune a smaller model to do the same task. This can reduce costs by 40x and increase speed by 5x. This is an interesting approach to using GPT-3 to reduce costs and increase speed, and could be useful for other applications.
- Knowledge distillation
I made ChatGPT and Bing AI have a conversation (and they are friends now)