Search the archive

NEW TO THE WEB PLATFORM IN JUNE

The web platform saw new features and improvements in June, such as new JavaScript Set methods and color interpolation in CSS gradients.

CHANGESETS

Changesets is a tool to manage versioning and changelogs. It's focused on multi-package repositories.

TYPESPEC

TypeSpec is a language created by Microsoft for defining cloud service APIs and shapes. It's an extensible language with primitives that can describe API shapes common among REST, OpenAPI, gRPC, and other protocols.

WHY AI INFRASTRUCTURE STARTUPS ARE INSANELY HARD TO BUILD

Building venture-scale AI infrastructure startups is extremely difficult because startups lack the differentiation and capital needed to compete with established players like GCP, AWS, Vercel, Databricks, and Datadog, who are all striving to create end-to-end AI platforms. The open-source community quickly replicates any promising innovations, further eroding the competitive advantage of startups. To survive, startups must either focus on a very narrow niche, raise substantial VC funding, or remain bootstrapped.

TOTAL ANNIHILATION GRAPHICS ENGINE

The Total Annihilation graphics engine was developed before hardware acceleration was widely available. The engine used a mix of fixed and floating-point math, with some assembly code optimizations. Due to the 8-bit palette limitation, various tricks like lookup tables were used for effects like alpha blending and shadows. To improve performance, unit rendering was optimized by caching 3D models as sprites and using z-buffering for self-intersection. This article continues to go through the various innovations of the graphics engine back then.

HOW TO HIRE AI ENGINEERS

AI engineers don't need deep machine learning expertise. Instead, they should have practical experience in building complex systems (on the web or otherwise) and also have a genuine curiosity for working with LLMs. A good hiring process for AI engineers focuses on assessing candidates' technical skills in scalable system development, their proficiency in asynchronous programming and distributed systems, and their readiness to tackle challenges in integrating LLMs into applications.

I RECEIVED AN AI EMAIL

This author received a generic email from a company promoting its product without any personalized effort as part of an AI-powered mass email campaign.

BOOTING LINUX OFF OF GOOGLE DRIVE

A developer successfully booted Linux off a Google Drive root with the aim of creating a "Cloud Native Computer" by overcoming various technical challenges, including manually switching roots and resolving dependency issues.

HACKER NEWS SEARCH ENGINE

A search engine for Hacker News.

LATENCY AND THROUGHPUT

Latency and throughput can be optimized through strategies like caching data and prioritizing network traffic.

JAVASCRIPT BLOAT IN 2024

JavaScript bundles on modern websites are much bigger than they need to be, even for simple functionalities. Popular websites, like ChatGPT and YouTube, have a high amount of client-side JavaScript bloat that doesn't always match the functionality the page offers. Certain websites, like adult websites, prioritize fast loading speeds and have small JavaScript bundles.

PROBLEMATIC SECOND

Leap seconds, added to account for variations in the Earth's rotation, occur infrequently. They can lead to unforeseen software errors and system failures due to inadequate testing and preparation. While various approaches exist to handle leap seconds, Google's "leap smear" method, which gradually distributes the extra second throughout the day, offers a smoother and more predictable solution. Leap seconds are planned to be abolished by 2035.

HOW SQL QUERIES WORK: SQL QUERY EXECUTION ORDER FOR TECH INTERVIEWS

A database management system parses, optimizes, and executes a SQL query in a specific order, starting with `FROM` and `JOIN` before `SELECT`. The process includes steps like tokenization, execution plan generation, data retrieval, filtering, sorting, and aggregation. Understanding this execution flow helps in writing efficient queries and optimizing database performance.

HOW I PLAN MY WEEK AS A SENIOR ENGINEER IN BIG TECH

This senior engineer uses an Excel spreadsheet to prioritize tasks based on organizational goals, feedback, and time estimates. Over three months, this helped them achieve 90%+ of their daily goals while providing a clear record of accomplishments and reducing overplanning.

PERSPECTIVE JUMPING: HOW TO FOCUS AS A SOLO-DEVELOPER

Design problems, particularly for solo developers, can be prone to fragmentation into smaller sub-problems, leading to a loss of focus on the original issue. Timeboxing alone is insufficient to combat this, as it lacks the necessary context and reasoning to refocus the developer. To effectively manage this, a solo developer can write from the perspective of different roles, like a marketer, manager, or developer, to prioritize tasks properly.

NEW WEB DEVELOPMENT. OR, WHY COPILOTS AND CHATBOTS ARE PARTICULARLY BAD FOR MODERN WEB DEV

A paradigm shift is happening towards simpler web development that utilizes built-in browser APIs. While this shift promises better performance and cost savings, it requires a skill set that many organizations and developers lack. LLMs are being used as a workaround, but they are not well-suited to this new shift due to a lack of relevant training data.

Star Dusk on GitHub

DOGGO

Doggo is a command-line DNS client with human-readable output and JSON output options. It supports multiple transport protocols, reverse DNS lookups, and more.

PDFSLICK

PDFSlick is a library for viewing and interacting with PDF documents in React, SolidJS, Svelte, and JavaScript applications. It uses Zustand to provide a reactive store for loaded documents.

MY FINETUNED MODELS BEAT OPENAI'S GPT-4

This author evaluated several fine-tuned LLMs on structured data extraction from press releases and found that they outperformed OpenAI's GPT models. The evaluation process was challenging due to the different locations and interfaces of the models. While the author was pleased with the results, they also acknowledge the tradeoffs of fine-tuning LLMs, such as the need for managing data and model versions.

HOW GOOGLE MIGRATED BILLIONS OF LINES OF CODE FROM PERFORCE TO PIPER

Google's source control system initially relied on a single Perforce server and faced scalability issues as the company grew. To address this, Google engineers developed Piper, a distributed source control system designed for massive scale. The migration from Perforce to Piper took over four years due to the deep integration of Perforce into Google's software ecosystem and the need to avoid disrupting production. However, the migration was successful and Piper is still used at Google today.

A GIT STORY: NOT SO FUN THIS TIME

In 1998, Linus Torvalds, creator of Linux, was overwhelmed by the growing number of code changes from developers. The solution they found was BitKeeper, a distributed version control system. Linux adopted BitKeeper in 2002, but its proprietary nature caused friction with free software advocates. Frustrated, Torvalds started developing Git, which gained popularity and became mainstream due to GitHub.

WHAT ACTUALLY HAPPENS WHEN A USER REJECTS COOKIES? (REDDIT THREAD)

Cookie Consent Manager platforms allow websites to categorize tracking scripts and obtain user consent before loading them, similar to how Tag managers function.

FOUR LINES OF CODE … IT WAS FOUR LINES OF CODE

THE OARB FRAMEWORK: WHY YOU SHOULD APPEAL TO SELF-INTEREST WHEN GIVING FEEDBACK

The OARB framework (Observation, Assertion, Repercussion, Benefit) is a technique for giving feedback that focuses on appealing to the recipient's self-interest by highlighting how their behavior negatively impacts them and how changing it would benefit them.

SOFTWARE ENGINEER PAY BY INDUSTRY

Software engineers in the tech and finance industries have average earnings ranging from $120,000 to $160,000 and $110,000 to $150,000 respectively.

UNDERSTANDING REACT COMPILER

React Compiler optimizes React components by automatically adding memoization to improve performance, reducing the need for manual optimization by developers. It transforms code into a more efficient version that caches values and memoizes results. However, this optimization comes at the cost of increased memory usage, which can be a problem when dealing with large amounts of data.

NOTES ON IMPLEMENTING DARK MODE

This blog post discusses implementing a tri-state dark mode feature for websites. To avoid theme flicker on load, developers should use inline JavaScript and respond to OS-level theme changes with the matchMedia() API. Local storage should be used to synchronize theme changes across tabs.

THE SUPER EFFECTIVENESS OF POKÉMON EMBEDDINGS USING ONLY RAW JSON AND IMAGES

Embeddings are numerical representations of objects often used in text analysis and search applications. LLMs like BERT and newer models with larger context windows can generate embeddings for text data, taking into account word position and punctuation for better contextual understanding. These embeddings can be used to identify similar items. This author explores embeddings by encoding Pokemon data, including both text and image data, into embeddings and analyzing their similarities. By combining text and image embeddings and using dimensionality reduction techniques like UMAP, the author was able to visualize and analyze relationships between Pokemon.

ASK HN: WHAT IS THE BEST CODE BASE YOU EVER WORKED ON? (HACKER NEWS THREAD)

Some of the best code bases developers have worked at Google, AOL, NetBSD, and other companies. These code bases all had great developer tooling, detailed code reviews, standardized style practices, and good test coverage.

A BUNCH OF PROGRAMMING ADVICE I'D GIVE TO MYSELF 15 YEARS AGO

To be a more efficient programmer, focus on improving the tools and processes you use. It's best to prioritize shipping speed over perfect code, as bad code can provide valuable feedback that can lead to improvements. Developers should also ask for help judiciously from teammates.

HTMX SUCKS

The team behind htmx published an essay about htmx's shortcomings, such as its lack of TypeScript support and the absence of a formal build process. However, these don't stand in the way of htmx's main goal, which is its simplicity and direct use of HTML to create interactive web applications without the overhead of frameworks.

CHROME IS ADDING `WINDOW.AI`

Chrome is adding `window.ai`, providing access to a Gemini Nano AI model right inside your browser. window.ai has a wide range of use cases like smart auto-completion, error correction and validation, natural language filtering, auto-filling forms, UI suggestions, first-pass summarization, and search.

SMOOTHMQ

SmoothMQ is a drop-in replacement for Amazon SQS (message queueing) with a smooth developer experience. It has a functional UI, observability, tracing, message scheduling, and rate-limiting. SmoothMQ lets you run a private SQS instance on any cloud.

THE FUTURE OF CACHING WITH NEXT.JS

Next.js v15 will introduce big updates to caching behavior, moving away from default caching for `fetch` requests and route handlers. It changes up partial prerendering, allowing developers to choose between static and dynamic rendering based on their application's needs.

THE ECONOMICS OF WRITING TECHNICAL BOOKS

Technical authors can make a decent income from their books, but they should not write them solely for monetary reasons. The amount of money earned varies depending on the distribution channel, with self-publishing platforms like Gumroad and Leanpub offering higher royalties than traditional publishers. Technical books can generate passive income over a long period, with some titles remaining relevant for decades. However, most technical books sell less than 5,000 copies per year.

BYTECODE BREAKDOWN: UNRAVELING FACTORIO'S LUA SECURITY FLAWS

Factorio, a popular game, uses Lua for game logic and modding, allowing Lua code execution on the client-side for multiplayer synchronization. However, the Lua implementation in Factorio is vulnerable to exploitation through malicious custom maps that bypass the bytecode verifier, allowing arbitrary code execution. The vulnerability, which enables attackers to create malicious bytecode and execute it on the client, is caused by the absence of an official bytecode verifier in Lua 5.2.1 and the presence of off-by-one issues in Factorio's custom implementation.

5 LESSONS FROM HELPING 9 STARTUPS MOVE TO OUTCOME-DRIVEN WORK

In their journey as a solopreneur guiding startups, this author learned to focus on a niche market and iterate on ideal customer profiles. Startups should also have a balanced approach to experimentation to have both sustainable growth and new innovations.

HOW VERCEL WORKS 🔍 — WITH MALTE UBL (60 MINUTE VIDEO)

Malte Ubl, the CTO of Vercel, talks about how Vercel works as a team and the future of AI-enabled applications.

CODE GALAXIES

An interactive visualization of language packages, like NPM and PyPI, as explorable galaxies.

DOCMOST

Docmost is an open source collaborative documentation and wiki software similar to Confluence and Notion.

WHAT DOES WORK-LIFE BALANCE EVEN MEAN?

Work-life balance is influenced by intrinsic motivation, opportunity cost, and health.

WHY GOOGLE SHEETS PORTED ITS CALCULATION WORKER FROM JAVASCRIPT TO WASMGC

Google Sheets initially used a Java-based calculation engine, but later switched to JavaScript for browser-based calculations. Due to performance limitations of JavaScript compared to Java, Google used WasmGC, an extension to WebAssembly that allows for efficient compilation of garbage-collected languages like Java. The initial prototype was slower than JavaScript, but optimizations like replicating JVM/V8 optimizations, using browser APIs, and making code more platform-agnostic resulted in huge performance improvements.

USING SHORT LIVED POSTGRES SERVERS FOR TESTING

Postgres installations consist of binaries and a data directory, allowing for easy setup of multiple instances on a server. A new instance can be created by setting up a new data directory and using the "pg_ctl" command. Existing data can be quickly populated into a new instance by cloning from a stable instance using the "pg_basebackup" command.

GRPC: THE BAD PARTS

gRPC is a high-performance RPC framework that has been successful but also has its downsides. It has a steep learning curve due to non-standard terms and complex implementations, compatibility issues with the web, lack of standardized JSON mapping initially, and finite message sizes requiring custom chunking for large files.

A RANT ABOUT FRONT-END DEVELOPMENT

Front-end development has become unnecessarily complex due to poor choices, such as over-reliance on JavaScript frameworks and ignoring fundamental principles like content importance and CSS capabilities. This complexity leads to frustration and inefficiency. Developers should prioritize simplicity.

A DEV'S THOUGHTS ON DEVELOPER PRODUCTIVITY

The software development process consists of two nested loops: an outer loop that aligns with the SDLC and an inner loop of reading, writing, and running code. The inner loop is crucial for productivity and achieving flow state, but is often disrupted by internal and external context switches. Developer productivity should be measured by the frequency of inner loop iterations rather than lines of code or commits.

software. Explore the Demo Center now.