Skip to main content

Rohan Deshpande

498 curated articles on engineering, startups, and the interesting corners of the internet.

GitHub Twitter LinkedIn Inbox
Tags
  • ''Give away your Legos' and other commandments for scaling startups

    Molly Graham, drawing from her experience scaling teams at Google, Facebook, and Quip, explains why rapid company growth is emotionally difficult and how to manage it. Her central metaphor is 'giving away your Legos' — the counterintuitive need to hand off responsibilities you built from scratch as new people join, fighting the instinct to hoard or micromanage. She breaks company scaling into phases: 30-50 employees (when you shift from family to company and must write down values), 50-200 (the foundation where hiring quality and firing quickly matter most), 200-750 (managing cultural growing pains like a teenager), and 750+ (where identity shifts to teams and politics emerge). Her key advice includes over-communicating mission and values, prioritizing principles over process, and helping people see the bigger opportunities that come with letting go.

    source|StartupsCareer
  • 'Stateless' sync using version vectors

    This article describes the architecture behind Good Spirits, a drink-tracking iOS app that uses stateless, conflict-free data syncing inspired by CRDTs. The author built a persistence layer using SQLite (via the GRDB framework) with two tables: one for check-in data with Lamport timestamps for last-writer-wins conflict resolution, and an operation log enabling delta updates via version vectors. The design allows any subsystem (UI, cloud, HealthKit, Untappd) to stay in sync by simply storing a version vector token and requesting changes since that point. Although CloudKit sync wasn't implemented before release, the architecture proved valuable for local development by keeping the data layer completely isolated from UI and network concerns.

    source|CRDTsCareer
  • $1 Microcontrollers

    This comprehensive review by embedded design consultant Jay Carlson evaluates 21 different microcontroller architectures that can be purchased for around one dollar, comparing their performance, peripherals, development tools, and power consumption. The article benchmarks each chip on standardized tasks including GPIO toggling, PWM generation, ADC sampling, and DMX-512 bit-banging, providing detailed measurements and oscilloscope captures. It covers architectures ranging from legacy 8-bit parts like the ATmega and PIC16 to modern Arm Cortex-M0+ processors, concluding that newer Arm-based chips often match or beat traditional 8-bit MCUs on price while offering superior tooling and performance.

    source|Hardware
  • 5000x faster CRDTs: An adventure in optimization

    Seph Gentle describes his quest to build an extremely fast CRDT implementation after being bothered by a benchmark paper that showed poor performance for the OT algorithm he used in Google Wave. He explains how naive CRDT implementations are slow because they process each character individually, and details his optimization journey using Rust, including run-length encoding, skip lists and B-trees for efficient lookups, and internal string representation choices. His Diamond Types implementation achieves roughly 5000x the speed of reference CRDT implementations, processing a real-world editing trace of 260,000 edits in about 56 milliseconds.

    source|CRDTsVC & Fundraising
  • 55 Pre-Launch Tips for Your Startup

    Peter Schroeder's article provides 55 actionable pre-launch marketing tips for startups, organized to help founders build buzz before their product officially launches. The tips cover a wide range of strategies including creating a landing page, building an email list, leveraging social media, content marketing, networking, and getting early press coverage. The guide is aimed at founders who may be unsure how to generate traction when all they have is an idea and the internet.

    source|Startups
  • A collection of outlandish human-computer interaction papers

    This article curates unusual and creative academic papers exploring unconventional human-computer interaction approaches and interfaces. It celebrates research creativity. The piece showcases HCI diversity. It documents experimental interfaces.

    source|Distributed Systems
  • A Comprehensive Study of Convergent and Commutative Replicated Data Types

    This technical report by Shapiro et al. presents a comprehensive study of Conflict-free Replicated Data Types (CRDTs), which enable distributed systems to replicate shared data across multiple sites without requiring coordination or consensus. The paper defines two forms of CRDTs: state-based convergent types (CvRDTs) that propagate by merging local state, and operation-based commutative types (CmRDTs) that propagate by transmitting operations. The authors provide a formal framework proving that both approaches achieve Strong Eventual Consistency, and present a catalog of useful CRDTs including counters, sets, registers, and graphs with their mathematical properties.

    source
  • A contentEditable pasted garbage and caret placement walk into a pub

    This technical article explores the quirks and challenges of implementing contentEditable in web browsers, particularly focusing on handling pasted content and caret positioning issues. The author documents frustrating browser inconsistencies and edge cases that developers encounter when building rich text editors. The piece serves as both a technical deep dive and a frustration vent for web developers.

    source|CRDTs
  • A crash course in just-in-time compilers

    This Mozilla Hacks article explains how JavaScript engines use Just-In-Time (JIT) compilation to bridge the gap between interpreters and compilers, achieving both fast startup and optimized execution. The JIT uses a monitor to identify warm and hot code paths, sending them first to a baseline compiler that creates stubs indexed by line number and variable type, then to an optimizing compiler that makes assumptions about types to generate faster machine code. When those assumptions prove wrong, the engine deoptimizes back to baseline code, and browsers add limits to prevent optimization/deoptimization cycles from degrading performance. The article serves as background for understanding WebAssembly, noting that while JITs dramatically improved JavaScript speed, they add runtime overhead that WebAssembly can avoid.

    source|CompilersLearning
  • A detailed look at the router provided by my ISP

    This article provides technical deep-dive into ISP-provided home router hardware, firmware, and capabilities, examining what services run and security implications. It explores consumer networking hardware from engineer perspective. The piece educates about ISP hardware. It demonstrates reverse-engineering approach.

    source
  • A Flexible Type System for Fearless Concurrency

    This PLDI 2022 paper by Milano, Turcotti, and Myers proposes a new type system for concurrent programs that allows threads to safely exchange complex object graphs without destructive data races. Unlike existing solutions that either enforce strict heap invariants prohibiting natural programming patterns or require pervasive annotations, their system called Gallifrey uses flexible regions and a novel borrowing mechanism to track object ownership across threads. The type system supports common patterns like sharing mutable data between threads with minimal annotation burden, and the authors prove it sound and implement it as an extension to Java.

    source|Rust
  • A free video streaming service that runs on a ESP32

    ESPFLIX is an open-source video streaming service running on ESP32 microcontroller with only 6KB of buffer memory, implementing a complete stack including MPEG1 software decoder, SBC audio codec, and delta-sigma modulation. Built with AWS CloudFront backend support, it demonstrates sophisticated media streaming on resource-constrained hardware through clever codec optimization. Video playback is controllable via AppleTV remote signals.

    source|Virtual MachinesHardware
  • A future for SQL on the web

    James Long introduces absurd-sql, a persistent backend for sql.js that enables SQLite to read and write data in small blocks through IndexedDB on the web. The project overcomes IndexedDB's severe performance limitations by intercepting SQLite's read/write requests and efficiently batching them, achieving 10x or better performance improvements over raw IndexedDB operations. Key technical tricks include using SharedArrayBuffer and Atomics.wait to turn async IndexedDB calls into synchronous ones that SQLite's C API requires, keeping long-lived IndexedDB transactions alive by blocking the worker thread, and intelligently switching to cursor-based iteration for sequential reads. The article also details the file locking and transactional semantics that prevent database corruption across multiple tabs.

    source|Databases
  • A generalised solution to distributed consensus

    Adrian Colyer reviews a paper by Heidi Howard and Richard Mortier that reframes distributed consensus using write-once immutable registers and configurable quorum sets, aiming to find a unifying protocol that subsumes existing algorithms. The approach allows different quorum configurations per register set, enabling flexible trade-offs between performance and fault tolerance, and proves that both Paxos and Fast Paxos are conservative instances of this generalized framework. The paper demonstrates that traditional quorum intersection requirements can be substantially weakened while still maintaining safety, opening up a diverse array of possible consensus algorithms beyond what the literature has explored.

    source|Distributed Systems
  • A half-hour to learn Rust

    A fast-paced introduction to the Rust programming language designed for developers who learn best by reading code examples. The article walks through Rust's key concepts including ownership, borrowing, lifetimes, pattern matching, and error handling through concise, progressive code examples. Rather than building a project, it focuses on reading and understanding Rust syntax and idioms in rapid succession.

    source|Rust
  • A history of Hup

    A historical overview of Hup, a version control or project management tool. The article traces its development, adoption, and eventual fate. It provides context for understanding tool evolution in the software development ecosystem.

    source
  • A Japanese Pen Maker Anticipated the Fountain-Pen Renaissance

    A story about a Japanese pen manufacturer's prescient decision to invest in fountain pen design and quality as ballpoint pens dominated. The article explores how the company's long-term vision resulted in growth as consumers rediscovered fountain pens. It demonstrates the value of counter-trend thinking.

    source
  • A military technique for falling asleep in two minutes

    A fitness expert named Justin Agustin went viral on TikTok sharing a military-developed technique for falling asleep in two minutes. The method, detailed in the book Relax and Win: Championship Performance, involves progressively relaxing muscles from forehead to toes, taking deep breaths, and then clearing the mind by visualizing calming scenarios like lying in a canoe on a calm lake or in a black velvet hammock. Originally developed for fighter pilots who needed peak reflexes, the technique reportedly works for 96% of people who practice it nightly for six weeks. The article also summarizes NHS recommendations about sleep requirements and health risks of sleep deprivation.

    source|Culture
  • A minimal stack based VM in C

    liblg is a minimal stack-based virtual machine library in C designed to be reasonably fast while maintaining simplicity, with computed goto architecture for instruction dispatch and tagged unions for values. Significantly outpaces Python3 but trails Go by roughly an order of magnitude in Fibonacci benchmarks, with potential applications in implementing custom domain-specific languages. MIT-licensed with 64 commits spanning C, Go, Python, and CMake.

    source|Virtual Machines
  • A Modern JavaScript Tutorial

    The Modern JavaScript Tutorial at javascript.info is a comprehensive, free online resource for learning JavaScript from scratch through advanced topics. It is organized into three parts: the JavaScript language itself (covering fundamentals, objects, data types, functions, classes, promises, async/await, generators, modules, and more), browser-specific topics (DOM manipulation, events, forms, resource loading), and additional articles on frames, binary data, network requests, storage, animations, web components, and regular expressions. The tutorial includes runnable examples and is community-translated into dozens of languages.

    source|Programming
  • A Mouse is a Database

    Erik Meijer's ACM Queue article argues that web and mobile applications increasingly rely on asynchronous and realtime streaming data, positioning this as a form of big data with 'positive velocity.' He derives the Rx (Reactive Extensions) IObservable<T> and IObserver<T> interfaces by crossing Java's Future<T> with GWT's AsyncCallback<T>, creating first-class representations of asynchronous data streams. The article demonstrates how UI events like mouse moves and text field changes can be treated as push-based streaming databases, composed using fluent query operators like Where, Select, SelectMany, Switch, and Throttle to build reactive applications such as an Ajax dictionary suggest feature.

    source|Databases
  • A Pi-Powered Plan 9 Cluster

    This project guide walks through building a self-contained 4-node Raspberry Pi cluster running Plan 9, the distributed operating system from Bell Labs that takes UNIX principles further by presenting everything as network-accessible files via the 9P protocol. The cluster is housed in a custom laser-cut acrylic enclosure with an integrated Ethernet switch, panel-mount USB and HDMI connections, and an LED-illuminated etching of Glenda, the Plan 9 mascot. The article explains Plan 9 key features including transparent distributed resource access, per-process namespaces, and the elimination of a superuser, noting that ideas from Plan 9 influenced the Go programming language and other modern systems.

    source|SystemsHardware
  • A Plane That Accidentally Circumnavigated the World

    After the attack on Pearl Harbor in December 1941, Pan Am flight 18602—a Boeing 314 Clipper—found itself unable to return to New York via its normal Pacific route due to the war. Instead, Captain Robert Ford and his crew were forced to fly westward around the world, hopping from Auckland through remote and dangerous locations across the South Pacific, Indian Ocean, Africa, and South America before finally arriving at LaGuardia on January 6, 1942. The bewildered air traffic controller who received their call had no idea a plane was inbound from New Zealand, making this the first commercial flight to accidentally circumnavigate the globe.

    source|Longform
  • A Primer on Database Replication

    This primer on database replication explains the core concepts and trade-offs of replicating databases across multiple servers. It covers single-leader replication (where one node handles writes and followers replicate data for read scaling), multi-leader replication (for geographically distributed writes), and leaderless replication (where any node can accept writes using quorum-based approaches). The article discusses synchronous versus asynchronous replication, consistency challenges like replication lag, and conflict resolution strategies, providing practical context from AlphaSights' experience scaling their systems across three continents.

    source|Distributed SystemsDatabases
  • A Review of Consensus Protocols

    Thomas Vilhena reviews four major consensus protocols — Chandra-Toueg, Ben-Or, Basic Paxos, and Nakamoto Consensus — through hands-on implementation in a shared codebase called ConsensusKit. The article covers each protocol theoretical algorithm and key implementation details, from Chandra-Toueg failure detector-based rotating coordinator approach to Ben-Or decentralized binary consensus with randomization, Paxos proposal numbering and two-phase commit, and Nakamoto proof-of-work longest chain rule. The author successfully extracted a common base structure across all four protocols consisting of Protocol, Instance, and Process classes that handle message routing and quorum management, highlighting structural similarities despite their different approaches to fault tolerance.

    source|Distributed Systems
  • A Second Conversation with Werner Vogels

    This article captures an extended conversation with AWS CTO Werner Vogels discussing distributed systems design, organizational scaling, and technology philosophy. The discussion covers lessons learned from building Amazon's infrastructure, approaches to consensus in distributed systems, and how organizational culture influences technical decisions. Vogels shares insights on balancing innovation with stability, managing complexity in large-scale systems, and the relationship between system design and organizational structure. The conversation provides perspective from one of the industry's most influential thinkers on systems design.

    source
  • A simple way to build collaborative web apps

    This article walks through building a collaborative real-time web app (a todo list called Todo Light) using React, Replicache for client-side state management and sync, Fly.io for global server deployment, and Postgres for persistence. The author explores how Replicache provides optimistic UI, offline support, and real-time collaboration by maintaining a persistent browser store that syncs with the server through push/pull endpoints. The backend implementation covers mutation handling, row-version-based diffing for efficient pulls, and WebSocket notifications via Ably. The article also discusses the challenge of achieving sub-100ms latency globally through distributed deployment.

    source|CRDTsWeb Dev
  • A successful Git branching model

    Vincent Driessen's influential 2010 post presents the git-flow branching model, which uses two permanent branches (master for production-ready code, develop for integration) and three types of supporting branches: feature branches for new development, release branches for preparing production releases, and hotfix branches for urgent production fixes. Each branch type has strict rules about where it branches from and where it merges back to, with --no-ff merges to preserve feature history. A 2020 retrospective note acknowledges that git-flow may be overkill for continuously delivered web apps, recommending GitHub flow instead, while noting git-flow remains appropriate for explicitly versioned software supporting multiple versions in the wild.

    source|Dev Tools
  • A tactical guide to kickstarting a community

    This guide provides practical tactics for building and growing an engaged community around a product, project, or brand. It covers initial activation strategies, member retention, community governance, and scaling dynamics. The article draws from real case studies to illustrate how to cultivate organic growth and long-term participation.

    source
  • A taste of CoreData – A graph framework

    This article explores using Apple's Core Data framework as a graph database, examining how to model complex relationships and queries using the framework. It demonstrates techniques for working with graph structures despite Core Data's relational roots, and discusses performance implications. The piece is useful for iOS developers building relationship-heavy applications.

    source
  • A walkthrough tutorial of TLA+ and its tools: analyzing a blocking queue

    This tutorial-style repository demonstrates formal verification of concurrent systems using TLA+ by modeling a blocking queue data structure progressively through git commits. Includes multiple specification variants with poison pill termination patterns, fairness properties, and refinement mappings along with TLAPS proofs and model checker configurations. Teaches that weeks of debugging can save hours of TLA+ by catching concurrency bugs before they occur.

    source|Distributed Systems
  • A Web Design Crash Course: From Developer to Developer

    This guide teaches web design fundamentals to developers, covering design principles, tools, and best practices from technical perspective. It helps technical people improve design skills. The piece demystifies web design. It enables designer-developers.

    source|Web DevLearning
  • Abusing AWS Lambda to make an Aussie search engine

    Ben Boyter built an Australian search engine called Bonzamate by exploiting AWS Lambda storage limits — compiling the search index directly into Lambda binaries, effectively getting free index storage within the 75GB Lambda deployment limit. The engine uses bloom filter-based indexing similar to Microsoft Bing bitfunnel approach, with each Lambda containing ~70,000 documents in a compiled Go binary, and a controller Lambda that fans out searches across 250+ worker Lambdas and merges results. The project includes a custom crawler for Australian .au domains, BM25 ranking with domain popularity as a cheap PageRank substitute, adult content filtering, and a sophisticated snippet extraction algorithm. The author argues for more competition in the search space and demonstrates that a single person with no budget can build a reasonable proof-of-concept search engine indexing 12 million pages.

    source|Cloud
  • Achieving an open-source implementation of Apple Code Signing and notarization

    Gregory Szorc announces a milestone for the apple-codesign Rust crate: a pure open-source implementation of Apple code signing and notarization that works on Linux, Windows, and other non-macOS platforms. The project replaced its dependency on Apple proprietary Transporter tool with a native Rust client for Apple new Notary API announced at WWDC 2022, enabling fully self-contained signing and notarization via the rcodesign CLI executable. This removes a long-standing barrier for the thousands of companies and developers who wanted to release Apple software from non-macOS CI/CD environments.

    source
  • Action Plan for a New CTO

    A strategic roadmap for someone newly appointed to Chief Technology Officer role at a technology company. The article outlines priorities like team assessment, technical strategy definition, and establishing systems for decision-making. It addresses both immediate actions and longer-term leadership initiatives for technical leaders.

    source|Career
  • Actuarial Life Table

    An explanation of actuarial life tables, the statistical tools used in insurance and pension planning to model mortality rates and life expectancy. The article covers how these tables are constructed from population data and used to calculate insurance premiums and pension obligations. It provides practical insights into the mathematical foundations of insurance industry.

    source|Finance
  • Adding directives to the GraphQL schema when there's no SDL

    This LogRocket article explains how to implement schema-type directives in code-first GraphQL servers, where the schema is generated programmatically rather than written in SDL. It covers the challenge that schema directives like @deprecated are straightforward in SDL-first approaches but require different strategies in code-first servers, walking through solutions using IFTTT-style directives that apply transformations through nested directive pipelines. The article demonstrates implementations using the GraphQL by PoP server in PHP, showing how directives can compose to create powerful field-level behaviors.

    source|Web Dev
  • Advanced Compilers: Self-Guided Online Course

    CS 6120 from Cornell is a doctoral-level advanced compilers course by Adrian Sampson offered as a self-guided online experience, covering topics including local analysis and optimization, dead code elimination, and local value numbering. The course provides structured lessons with video lectures and reading materials for self-paced learning outside the traditional classroom setting.

    source|Compilers
  • Advanced Data Structures

    MIT 6.851 Advanced Data Structures is a graduate-level course taught by Erik Demaine covering cutting-edge results and open research in data structures including persistence and retroactivity (time travel), geometric data structures, dynamic optimality of binary search trees, memory hierarchy optimization, hashing, integer data structures, dynamic graphs, string searching, and succinct data structures. The course uses an inverted lecture format with pre-recorded video lectures watched at home and in-class time devoted to collaborative problem solving, including open research problems that have led to over a dozen published papers. It requires a strong algorithms background equivalent to MIT 6.046 and includes weekly problem sets, scribe note revisions, and a research-oriented final project.

    source|Algorithms
  • Advanced Distributed Systems

    CS 525 at the University of Illinois is a graduate course on advanced distributed systems taught by Prof. Indranil Gupta that bridges research across cloud computing, peer-to-peer systems, distributed algorithms, and sensor networks. The course centers on a semester-long project aimed at producing either an entrepreneurial effort or a conference-quality research paper, with a strong track record of student projects being published at venues like ICDCS, INFOCOM, and ACM SoCC.

    source|Distributed Systems
  • Aggressive Chess Openings

    This article examines chess openings emphasizing immediate attack and advantage-seeking rather than positional consolidation. It provides chess strategy perspective. The piece documents chess theory. It analyzes opening principles.

    source|Culture
  • Amazon's ML University is making its online courses available to the public

    Amazon Science announces that Amazon's internal Machine Learning University (MLU), founded in 2016 to train Amazon employees, is making its courses freely available to the public. The initial release includes three accelerated online courses covering natural language processing, computer vision, and tabular data, with nine more in-depth courses planned. Classes are taught by Amazon scientists and include coding materials on GitHub, with video lectures recorded at home studios during the pandemic. The initiative aims to democratize ML education and address the global shortage of ML practitioners, complementing the textbook Dive into Deep Learning written by Amazon scientists.

    source|ML & AI
  • An Engineer’s Guide to Stock Options

    Alex MacCaw wrote a plain-English guide to help engineers understand startup stock options, covering how shares work, vesting schedules with cliff periods, exercise strategies, and tax implications including 409A valuations and AMT. The article walks through concrete scenarios showing the real cost of exercising options and advises engineers to ask key questions upfront about total shares outstanding, exercise price, and preferred share price. It also covers financing options for those who cannot afford to exercise.

    source|Finance
  • An Illustrated Guide to Masked Wrestlers

    A visual guide showcasing the diverse wrestling personas and signature masks of professional wrestlers from lucha libre and other traditions. The article explores the artistry and cultural significance of wrestling masks as performance and identity markers. It celebrates the visual creativity in professional wrestling.

    source|Culture
  • An illustrated guide to plastic straws

    Civil engineer BJ Campbell argues with data that banning plastic straws in the US is misguided because zero plastic thrown in American garbage cans enters the ocean. The Great Pacific Garbage Patch is overwhelmingly fed by rivers in developing countries, with the Philippines being the largest contributor through its 4,820 short rivers lacking waste collection. The article shows that US plastic recycling actually makes the problem worse since recycled plastic shipped overseas gets improperly dumped at rates of 55-80%.

    source|CloudCulture
  • An Introduction to Conflict-Free Replicated Data Types

    Lars Hupel's interactive tutorial series introduces Conflict-Free Replicated Data Types (CRDTs) with a focus on the mathematical foundations. The first installment covers preliminaries including the motivation for CRDTs in distributed web applications where devices may go offline or have slow connections. The series uses live JavaScript code snippets with property-based testing via fast-check to demonstrate concepts. Hupel focuses on the algebraic background (lattices, partial orderings) that powers CRDT implementations, with the series progressing through algebras, contracts, lattices, combinators, tombstones, registers, and deletion.

    source|CRDTs
  • An Introduction to Limit Order Books

    This comprehensive guide explains the fundamentals of limit order books and how electronic securities trading works. It covers the structure of order books with bid and ask sides, price/time priority ordering, and core operations like placing, canceling, and amending orders. The article walks through trade execution mechanics including single-level trades, multi-level trades, and leaving remainders, with detailed diagrams. It also covers order types (limit, market, stop), time-in-force options, market making, iceberg and hidden orders, market data levels (L1/L2/L3), and auction mechanisms used at trading session opens.

    source|Finance
  • An opinionated guide for developers getting things done using the Nix ecosystem

    nix.dev is the official documentation site for the Nix ecosystem, maintained by the Nix documentation team. It provides tutorials, guides, reference material, and conceptual explanations for learning and using Nix. The site highlights Nix's key capabilities including reproducible development environments, declarative Linux machine specification, transparent build caching via binary caches, atomic upgrades and rollbacks, cross-compilation support, and remote builds and deployments. It targets developers, DevOps engineers, system administrators, data scientists, and anyone who needs computers to behave reproducibly.

    source|Dev Tools
  • Analog TV Station on ESP8266

    Channel3 transforms an ESP8266 microcontroller into an analog television transmitter broadcasting on Channel 3 using clever 1-bit dithering to encode NTSC/PAL-compliant video signals. It leverages the I2S bus at 80 MHz with DMA buffers and includes nine different demo screens with 3D graphics capabilities and a web interface for real-time color customization. The project demonstrates practical RF engineering at the edge of Nyquist limits, creatively exploiting frequency mirroring effects.

    source|Hardware
  • Applied Monotonicity: A Brief History of CRDTs in Riak

    Christopher Meiklejohn traces the history of CRDT development at Basho/Riak, focusing on the evolution from the expensive Observed-Remove Set (which stored tombstones for every deletion, requiring O(n) space proportional to total operations) to the Observed-Remove Set Without Tombstones that achieved O(n) space proportional only to current elements. The key insight was encoding causal history information directly into the data structure using dotted version vectors, enabling safe garbage collection without the causal delivery guarantees that operation-based CRDTs relied on. The article also details a subtle bug where an optimization that skipped merges based on clock comparison alone failed because removals do not advance the clock, and discusses broader work on monotonic programming and type systems for verifying CRDT correctness.

    source|CRDTs
  • Are CRDTs suitable for shared editing?

    Kevin Jahns, author of the Yjs CRDT library, demonstrates through extensive benchmarks that CRDTs are absolutely suitable for shared editing on the web, countering claims from CodeMirror and Xi Editor developers about prohibitive overhead. The key optimization is a compound representation where consecutive insertions are merged into single Item objects, reducing the metadata from millions of objects to just thousands for real documents — Martin Kleppmann's 260k-operation editing trace of a conference paper requires only 11k Item objects and 19.7MB of memory. Yjs handles up to 26 million changes in 220MB of memory with linear parse time scaling, making it practical even for extremely large documents while enabling peer-to-peer collaboration without a central server.

    source|CRDTs
  • Arguments against JSON-driven development

    A critical examination of using JSON as the primary data format throughout development, from configuration to APIs to storage. The article argues for alternatives when JSON's flexibility becomes a liability, leading to poorly validated schemas and type confusion. It advocates for type-safe and validated data formats in appropriate contexts.

    source
  • Arpanet Part 3: The Subnet

    This detailed history covers the political and technical challenges of building the ARPANET subnet in the late 1960s. It explains how Wes Clark proposed using dedicated minicomputers (Interface Message Processors/IMPs) at each site to handle networking, relieving host institutions of the burden. BBN won the contract to build the IMPs using Honeywell DDP-516 computers, delivering the first one to UCLA in August 1969. The article also covers the Network Working Group's development of host protocols (NCP, Telnet, FTP) and the early struggles with low network utilization and resource sharing, noting that email would eventually save ARPANET from accusations of stagnation.

    source|Networking
  • Array with Constant Time Access and Fast Insertion and Deletion

    IgushArray is a C++ data structure combining array constant-time access with O(N^1/2) insertion and deletion performance using a two-level hierarchy of an array of pointers to fixed-size double-ended queues. Fully implements the std::vector interface while achieving speedups reaching 67x on 10-million-element structures for insertion-heavy workloads. Elements maintain O(1) access time while limiting insertion/erasure to square-root complexity.

    source
  • Assholes: A Probing Examination

    Rod Hilton's essay argues that hiring smart but socially toxic engineers is the most damaging staffing mistake a tech company can make. He defines asshole behavior as actions that make colleagues feel worse about themselves, cites research showing that removing an asshole boosts team productivity more than adding a superstar, and describes how incivility spirals cause asshole behavior to spread through organizations. He proposes a four-step FIBR method: Focus (declare zero-tolerance), Identify (have managers observe behavior), Badger (consistently correct every incident), and Represent (make the anti-asshole culture clear to candidates so self-aware assholes self-select out).

    source|Career
  • Automerge: JSON-like data structure for building collaborative apps

    Automerge provides fast CRDT implementations, a compact compression format, and an efficient sync protocol for networked collaborative applications. Features a Rust core compiled to WebAssembly for JavaScript with C language bindings, achieving approximately 10x memory usage reduction in version 3. Enables local-first applications with conflict-free data synchronization similar to how relational databases support server applications.

    source|CRDTsAlgorithms
  • Backend for Frontend pattern

    Nicolas Frankel explains the Backend For Frontend (BFF) pattern, which addresses the complexity of serving different clients (web, mobile, tablet) from microservices. The core problem is that each client needs different data subsets, and making every microservice filter data per client creates exponential coupling. BFF solves this by inserting a dedicated component per client type that fetches data from required microservices, extracts relevant parts, aggregates them, and returns client-specific responses. The same team develops both the client and its BFF, and the pattern can be implemented as a separate deployment unit or as an API Gateway plugin.

    source
  • Becoming a CTO

    Juozas Kaziukenas argues that being a CTO is fundamentally about business strategy rather than coding. A CTO defines the technology vision, protects the engineering team from becoming a pure execution arm, and communicates technology decisions in plain English to non-technical stakeholders. The role requires balancing innovation with pragmatism—avoiding costly rewrites that deliver zero customer value, making ROI-based decisions following the 80-20 rule, and building sustainable teams. The author emphasizes that CTOs must have real authority over technical decisions, understand business needs deeply, and create environments where teams can do their best work.

    source|Career
  • Becoming a Startup Finance Quarterback with Eric Lesser

    An article about becoming a Finance Quarterback at a startup, covering how non-finance founders can develop financial acumen and manage the financial aspects of scaling. It discusses working with finance professionals, understanding key metrics, and making informed financial decisions. The piece is valuable for founders transitioning into broader business leadership.

    source|Startups
  • Ben Evans' presentations

    Benedict Evans' presentations page collects his semi-annual big-picture analyses of macro and strategic trends in the tech industry. The page archives presentations from 2013 to 2025, tracking the evolution of his focus from 'Mobile is Eating the World' (2013-2016), through 'Ten Year Futures' and 'The End of the Beginning' (2017-2018), to 'The Great Unbundling' (2020), 'The New Gatekeepers' (2022), and most recently 'AI Eats the World' (2023-2025). Evans has presented for major companies including Alphabet, Amazon, LVMH, and Vodafone.

    source
  • Beyond the Bitcoin bubble

    An article examining the Bitcoin phenomenon beyond hype, analyzing its technological innovations, economic implications, and lessons for digital currencies. The piece provides balanced perspective on cryptocurrency's potential and limitations. It explores what remains valuable after speculative bubbles deflate.

    source|Finance
  • Big List of Naughty Strings

    A curated collection of test strings designed to expose input validation vulnerabilities in software applications, provided in multiple formats (txt, json, base64) for accessibility. Includes implementations across Node.js, .NET, PHP, and C++, serving both automated QA systems and manual testers. Addresses a real gap in software testing by targeting Unicode anomalies, problematic character sequences, and edge cases missed by major companies.

    source|Culture
  • Blockchains from a Distributed Computing Perspective

    Maurice Herlihy's paper provides a tutorial on blockchains from the perspective of distributed computing, arguing that much of the blockchain world is a disguised mirror-image of classical distributed computing problems. He covers the basic notions and mechanisms underlying blockchains including consensus protocols, smart contracts, and the relationship between traditional Byzantine fault tolerance and blockchain consensus mechanisms. The paper bridges the gap between the cryptocurrency world and academic distributed computing, helping researchers in each field understand the other's contributions and challenges.

    source|Distributed SystemsFinance
  • Blockchains from the ground up: Part 1

    An introductory guide to blockchain technology from first principles, building understanding from cryptographic primitives through distributed consensus to complete blockchain systems. The article demystifies blockchain beyond cryptocurrency hype, explaining core mechanisms and design choices. It serves as a gentle introduction to foundational blockchain concepts.

    source|Finance
  • Bloom Filters by Example

    This interactive tutorial explains Bloom filters, a probabilistic data structure that efficiently tests set membership. Using a live bit vector visualization, readers can add strings and see how hash functions (Fnv and Murmur) set bits in the vector, then test membership queries. The tutorial covers practical considerations including how to choose the number of hash functions (optimal k = (m/n)ln(2)), filter sizing to control false positive rates, and O(k) time complexity for insertions and lookups. It also surveys real-world implementations in Chromium, Redis, Apache Spark, RocksDB, ScyllaDB, and SQLite.

    source|Algorithms
  • Bloom filters debunked: Dispelling 30 Years of bad math with Coq

    Gopiandcode describes research presented at CAV2020 that produced the first certified proof of the false positive rate of Bloom filters using the Coq proof assistant. The article explains how the widely cited expression for Bloom filter false positive probability, dating back to Bloom's 1970 paper, contains an error in assuming that selected bits are independent. Bose et al. corrected this in 2008 with an alternative derivation using a balls-into-bins analogy, but their proof also contained errors found two years later. The Ceramist project finally settled the matter by producing a machine-verified proof from first principles in Coq, confirming Bose et al.'s corrected bound.

    source|Algorithms
  • Bonneville can be a tough place

    An article about the Bonneville Salt Flats in Utah, exploring their geology, history, and the conditions that make them a challenging environment. The piece covers the physical characteristics and human experiences at this natural wonder.

    source|Culture
  • Break a mirrored volume in Windows 2008 R2

    A technical guide for breaking a mirrored RAID volume in Windows Server 2008 R2, covering the procedures and considerations for removing one side of a mirrored pair. The article provides step-by-step instructions for this system administration task.

    source
  • Breaking a myth: Data shows you don't actually need a co-founder

    This TechCrunch article uses CrunchBase data to challenge the widely held belief that startups need co-founders. The data shows that more than half of startups with an exit had just a single founder, with the average being 1.72 founders. The analysis examines two measures of success—companies that raised over $10 million in funding and companies that had successful exits—finding that solo founders are far more common among successful startups than conventional wisdom suggests. The article argues that while co-founders can be valuable, the data breaks the myth that they are essential.

    source|Startups
  • Build a Two-Way Pager with Lora

    This IEEE Spectrum hands-on article details building a two-way pager using LoRa, a low-power radio protocol designed for IoT that provides 2-15 km range at 0.3-27 kbps data rates. The author progressed from a breadboard proof-of-concept with AI-Thinker Ra-02 modules and ATmega328 microcontrollers to a refined PCB design with a 128x64 LCD, Atmel SAMD21 Cortex M0 processor, and RFM95W transceivers. The article covers practical hardware challenges including antenna impedance matching, lithium-ion battery cold-weather failures, PCB color affecting reflow soldering, and power-on timing issues with the on/off controller.

    source|Hardware
  • Build a VM step by step with Rust

    This is the prelude to a tutorial series on building a language virtual machine in Rust. Part 00 provides a crash course in computer hardware fundamentals needed for the rest of the series, explaining what a language VM is (like Python's interpreter or the Java JVM), how source code gets compiled through assembly to binary, and the role of registers in CPU execution. The tutorial sets up the project of writing an application that pretends to be a CPU and executes programs, which will also require inventing a custom language and bytecode format.

    source|Virtual MachinesRust
  • Build Your Own React

    Rodrigo Pombo walks through rewriting React from scratch, step by step, following the architecture of real React 16.8 code but without optimizations. The tutorial covers eight steps: implementing createElement, render, concurrent mode, fibers, render and commit phases, reconciliation, function components, and hooks. By building each piece incrementally, readers gain deep understanding of React's internal workings including the fiber tree structure, work loop scheduling, and how hooks like useState maintain state across re-renders.

    source|Web Dev
  • Build Your Own Text Editor

    This step-by-step tutorial guides readers through building a text editor in C, based on antirez's kilo editor. The final result is about 1000 lines of C in a single file with no dependencies, implementing basic editing features, syntax highlighting, and search. The tutorial walks through 184 incremental steps across chapters covering setup, entering raw mode, raw input/output, building a text viewer, then a full text editor, adding search functionality, and implementing syntax highlighting. Each step adds or changes a few lines of code with immediate observable results.

    source
  • Building a React Native App for Your Christmas Lights

    This tutorial demonstrates building a local-network-only IoT system to control Christmas lights using a Jetson Nano, a Z-Wave USB stick, and waterproof outdoor light switches. The author creates a Flask API that interfaces with the Z-Wave network via the openzwave Python library to toggle switches on and off, then builds a simple React Native mobile app using Expo that sends GET and POST requests to the local server. The entire system stays off the public internet, communicating only over the home network, and the author emphasizes keeping IoT devices private rather than cloud-connected.

    source|Web Dev
  • Building ClickHouse Cloud From Scratch in a Year

    The ClickHouse team describes how they built ClickHouse Cloud, a managed serverless SaaS offering on top of the popular OLAP database, from the ground up in under a year. The post covers their planning process, architecture and design decisions, security and compliance considerations, and how they achieved global scalability and reliability in the cloud. They share the unconventional timeline and milestones of taking a well-established open-source database and transforming it into a production cloud service, along with lessons learned throughout the process.

    source|Cloud
  • Building Your Color Palette

    This Refactoring UI excerpt explains how to build a comprehensive color palette for real UI design, arguing that fancy color palette generators producing five hex codes are insufficient. A practical palette needs three categories: greys (8-10 shades for text, backgrounds, panels), primary colors (5-10 shades for actions and navigation), and accent colors (multiple shades for semantic states like error red, warning yellow, success green). The recommended process is to pick a base color, define the darkest and lightest shades based on actual use cases, then systematically fill in the gaps using a 9-shade scale (100-900).

    source|Design
  • Building your own shell using Rust

    Josh McGuigan provides a hands-on tutorial for building a Unix shell in Rust, progressively adding features in under 100 lines of code. Starting with reading a single command from stdin, the tutorial adds multi-command support, argument parsing, shell built-ins (cd, exit), error handling for invalid commands, and pipe support using Rust's Command API with Stdio redirection. The tutorial explains key concepts like why cd must be a shell built-in (it modifies the shell's own state), and how pipes chain processes by redirecting stdout of one command to stdin of the next.

    source|Rust
  • Cache Poisoning at Scale

    This security research paper details techniques used to report over 70 web cache poisoning vulnerabilities across various bug bounty programs. The author describes exploiting inconsistencies in CDNs like Cloudflare, Fastly, and Akamai, including capitalized host headers, URL fragment handling in Apache Traffic Server (CVE-2021-27577), x-forwarded-scheme header abuse for redirect loops, and storage bucket misconfigurations. Notable findings include a $7,500 GitHub CP-DoS, a $6,300 Shopify single-request cache poisoning, stored XSS across 21 subdomains, and Fastly host header injection enabling cross-origin attacks.

    source|Security
  • CAP is Only Part of the Story

    Inel Pandzic explains why the CAP theorem alone is insufficient for describing distributed system tradeoffs and introduces PACELC as a more complete framework. While CAP only describes the consistency/availability tradeoff during network partitions, PACELC adds the normal operation tradeoff between consistency and latency. The theorem states: during a Partition choose Availability or Consistency, Else choose Latency or Consistency. The article categorizes real databases (MySQL, Cassandra, Kafka, Zookeeper) into four PACELC categories and explains why understanding these tradeoffs matters when choosing a datastore.

    source|Distributed Systems
  • Case Study: Npm uses Rust for its CPU-bound bottlenecks

    NPM's adoption of Rust for performance-critical operations demonstrates how systems languages can solve bottleneck problems in interpreted language ecosystems. The article examines specific npm operations that benefited from Rust's performance characteristics, the integration challenges faced when combining JavaScript with compiled modules, and the resulting performance improvements. It discusses the trade-offs between pure JavaScript solutions and native compilation, including maintenance burden and deployment considerations. This case study provides practical guidance for teams evaluating similar polyglot approaches to performance optimization.

    source|Rust
  • Casio F-91W: The strangely ubiquitous watch

    This BBC article explores the surprising ubiquity and cultural significance of the Casio F-91W digital watch, launched in 1989 and retailing for as little as seven pounds. The watch gained unexpected notoriety when leaked US documents reportedly revealed that possession of the F-91W was viewed with suspicion at Guantanamo Bay as a possible sign of terrorist links, with over 50 detainee reports referencing it. Despite this, the watch remains a global bestseller due to its reliability, affordability, and seven-year battery life, and has experienced a resurgence among trendsetters and collectors drawn to its unpretentious design.

    source|Culture
  • Catalog of Patterns of Enterprise Application Architecture

    A reference catalog of design patterns used in enterprise application architecture, based on Fowler's seminal work. The catalog covers patterns for enterprise integration, persistence, and distribution. It's a valuable reference for architects designing large-scale systems.

    source
  • Chess Tactics Explained in Plain English

    Ward Farnsworth's 'Predator at the Chessboard' is a comprehensive free online resource for learning chess tactics explained in plain English. The site provides detailed instruction on tactical patterns including forks, pins, skewers, discovered attacks, and other combinations, using carefully selected positions with thorough written explanations rather than just notation. The resource is designed to build tactical vision systematically from basic to advanced concepts.

    source|Culture
  • CLARK cybersecurity curriculum

    CLARK (Cybersecurity Labs and Resource Knowledge-base) is the largest platform providing free cybersecurity curriculum, housing over 2,110 learning objects and 2,740 resources created by top educators. The platform offers modular, quality-reviewed curriculum organized as learning objects with clearly identified learning outcomes based on Bloom's Taxonomy. Resources cover undergraduate through postgraduate levels on topics including AI law and policy, cyber foundations, network security, and secure software. CLARK is supported by California State University, NSF, NSA, and Towson University, with all content freely available under CC BY-NC-SA 4.0 license.

    source|Security
  • Clocks and Causality - Ordering Events in Distributed Systems

    Giridhar Manepalli provides a comprehensive guide to logical clocks in distributed systems, covering Lamport Clocks, Lamport Origin Clocks, Vector Clocks (including Dotted Vector Clocks), Version Vectors, and Lamport Causal Clocks. Each clock design makes different tradeoffs between space complexity (O(1) for Lamport vs O(n) for vector clocks) and functionality like real-time ordering, concurrent event detection, and total ordering capability. The article explains how these clocks are foundational for multi-leader storage systems, conflict resolution, CRDTs, and collaborative editing applications, with detailed comparison algorithms and event ordering techniques.

    source|Distributed Systems
  • Collaborative Software That’s Wary of the Cloud

    An article exploring software tools designed for collaboration while maintaining skepticism about cloud dependency and centralization. It covers alternatives to traditional SaaS collaboration tools that prioritize user control. The piece advocates for decentralized collaboration approaches.

    source|CRDTsCloud
  • Color: From Hex codes to Eyeballs

    This article provides a deep dive into color science, explaining how colors are represented digitally and perceived by human vision. It covers the electromagnetic spectrum, how cone cells in the retina respond to different wavelengths, and how tristimulus color values form the basis of color models like CIE XYZ. The piece walks through gamma correction, sRGB color space, and how hex codes ultimately map to specific wavelengths of light emitted by monitor subpixels. It also explores topics like color mixing, metamerism, and why certain color transformations require careful handling of perceptual uniformity.

    source
  • Computer Science textbooks that are freely available online

    A curated collection of high-quality computer science textbooks that are available for free online. The article lists resources covering algorithms, data structures, operating systems, and other core CS topics. It serves as a comprehensive guide for students and self-learners to access quality educational materials without cost.

    source|Learning
  • crdt-richtext: Rust implementation of Peritext and Fugue

    This article introduces crdt-richtext, a Rust crate that combines the Peritext and Fugue CRDT algorithms for collaborative rich text editing. Peritext handles merging concurrent formatting changes (bold, italic, comments) while preserving user intent, and Fugue minimizes interleaving problems when multiple users type concurrently. The implementation uses B-Trees extensively for efficient lookups and updates, columnar encoding for compact storage, and is heavily tested with libFuzzer. Benchmarks show it outperforms Automerge and Yjs on real-world editing datasets, and the crate is designed to be incorporated into the Loro general-purpose CRDT library.

    source|CRDTsRust
  • CRDTs are the future

    In this essay, Seph Gentle (former Google Wave engineer) reflects on his shift from advocating Operational Transform (OT) to embracing CRDTs for collaborative editing. He recounts how OT powered Google Wave and ShareJS but was limited by its dependency on a centralized server, making federation difficult. After seeing Martin Kleppmann's advances in CRDT performance, encoding size, and features, he concludes that CRDTs have overcome their historical drawbacks and are now viable replacements for OT. He argues that CRDTs enable a local-first software future where users own their data, and commits to building high-quality CRDT implementations.

    source|CRDTs
  • CRDTs: The Hard Parts

    This page presents Martin Kleppmann's talk 'CRDTs: The Hard Parts' delivered at the Hydra distributed computing conference in 2020. The talk moves beyond basic CRDT introductions to address challenging problems that arise in real-world CRDT implementations, including issues with interleaving of concurrent text edits, handling moving elements in ordered lists and trees, and managing large metadata overhead. Kleppmann presents novel algorithms developed with colleagues to address these challenges while maintaining the consistency guarantees that make CRDTs attractive for collaborative editing applications.

    source|CRDTs
  • Create Your Own Programming Language

    This is a landing page for an ebook and course system by Marc-André Cournoyer that teaches how to create your own programming language. The 100-page PDF covers core compiler and interpreter concepts and applies them to building a custom language in Ruby, with full source code for three languages in Ruby and Java, exercises with solutions, and a screencast. Endorsed by Matz (creator of Ruby) and Jeremy Ashkenas (creator of CoffeeScript), the system is positioned as a practical, affordable alternative to traditional compiler textbooks.

    source|Programming
  • Creating a Collaborative Editor

    This tutorial walks through building a collaborative text editor using CRDTs (Conflict-Free Replicated Data Types). The author explains how CRDTs guarantee eventual consistency through mathematical properties of commutativity, associativity, and idempotence, and implements an operation-based CRDT sequence where each character has a unique global index. The editor uses React with the Quill rich-text editor on the frontend, WebSockets for real-time communication, and a Node.js/Express server as a central relay. The article covers handling concurrent inserts via UUID-based tiebreaking and deletions via tombstones, with source code provided for both client and server.

    source|CRDTs
  • Crsql - Multi-writer and CRDT support for SQLite

    cr-sqlite is a SQLite extension enabling multi-writer database replication through CRDTs, functioning as Git for your data with support for offline editing, real-time collaboration, and cross-device synchronization. Automatically reconciles divergent writes using CRDT semantics (last-write-wins, summation for counters, union for sets) while maintaining read performance. Implemented as a runtime-loadable extension compatible with JavaScript, Python, Rust, and WebAssembly.

    source|CRDTsDatabases
  • Cryptocurrencies – An Assessment by Reserve Bank of India

    An assessment of cryptocurrencies by India's Reserve Bank of India, examining their regulatory implications, risks, and economic impact. The official assessment provides perspective from a central bank on digital currency policy. It covers regulatory and institutional perspectives on cryptocurrency.

    source|Finance
  • Data structures and algorithms interview questions and their solutions

    This comprehensive resource provides 500 data structures and algorithms interview questions with complete solutions and explanations. The questions are organized by category and difficulty level, covering arrays, linked lists, trees, graphs, dynamic programming, and other core computer science topics. It serves as an extensive preparation resource for technical interviews at major tech companies.

    source|Algorithms
  • Data Structures Reference

    Interview Cake's data structures reference page provides a quick-reference cheat sheet covering the Big O costs and core properties of common data structures used in coding interviews and computer science. It covers arrays, dynamic arrays, linked lists, queues, stacks, hash tables, trees, binary search trees, graphs, tries, heaps, priority queues, bloom filters, and LRU caches, with brief descriptions of each structure's key characteristics and use cases.

    source|Algorithms
  • Data Transfer Project by Apple

    An explanation of the Data Transfer Project, a collaborative initiative by Apple and other tech companies enabling users to transfer personal data between services. The project addresses data portability and user control over digital information. It covers emerging standards for data interoperability.

    source
  • Decimating Array.Sort with AVX2

    Part 5 of a series on reimplementing array sorting using AVX2 SIMD intrinsics to dramatically outperform .NET's Array.Sort(). The multi-part series starts with a QuickSort refresher and comparison to Array.Sort(), introduces vectorized hardware intrinsics and vector types, walks through initial vectorized sorting code, and progressively optimizes the implementation. The author demonstrates how leveraging CPU vector instructions (processing multiple elements simultaneously) can achieve sorting performance far exceeding standard library implementations, while providing detailed benchmarks and explanations of the low-level optimization techniques involved.

    source
  • Deep learning papers reading roadmap

    A structured learning guide organizing seminal deep learning papers by topic and progression level, covering deep learning history, methodological advances, and applications in NLP, computer vision, reinforcement learning, and robotics. The roadmap includes rating indicators and brief context to guide learners through the field's evolution. Emphasizes progressing from foundational to cutting-edge work and state-of-the-art contributions.

    source|ML & AI
  • Designing Distributed Systems E-Book

    A free e-book resource on design patterns and principles for building distributed systems. The guide covers patterns for reliability, scalability, and efficiency in distributed applications. It serves as comprehensive reference material for system architecture.

    source|Distributed Systems
  • Developer's Guide to SaaS Compliance

    This guide by Aman Kandola explains the key compliance standards and certifications that SaaS developers need to understand to build trustworthy applications. It covers why compliance certifications matter for communicating data security practices to customers, common certifications for SaaS applications, compliance with legal requirements, and what compliance means specifically for developers in their day-to-day work. The article provides practical guidance on the road to getting certified and how handling sensitive user information securely is essential for long-term business success.

    source|Business
  • DIB Guide: Detecting Agile BS

    This short guide from the Defense Innovation Board provides DoD program executives and acquisition professionals with concrete questions to detect whether software projects are truly using agile development or are just waterfall in agile clothing. It maps agile values to DIB maxims, outlines key questions for program managers and developers (such as whether teams deliver working software to real users every iteration), and lists red flags and positive signs. The document emphasizes that genuine agile requires user engagement, iterative delivery, and empowered development teams rather than just adopting agile terminology.

    source|Career
  • Differential Dataflow for Mere Mortals

    Dida is a Zig library enabling streaming, incremental, iterative computation on time-varying collections with familiar operations like map, join, and loop. Designed for accessibility with core under 3,000 lines of code, using columnar batch storage and merge-join operations. Targets cross-language bindings and emphasizes exposing storage and scheduling details as an interpreter backend with planned graphical debuggers.

    source|Distributed Systems
  • Digital Tools I Wish Existed

    This post describes four categories of digital tools the author wishes existed: a queue management system for inbound content across formats and sources, a universal book log with recommendation and sharing capabilities, intelligent PDF/ebook/audiobook readers with better navigation and annotation features, and a centralized search interface (memex) across all personal digital interactions. The author highlights the high friction in finding, organizing, consuming, and re-finding digital content, and envisions tools that bridge the gap between our minds and devices while respecting privacy.

    source|Dev Tools
  • Distributed Systems Reading List

    A curated reading list of papers, articles, and books on distributed systems covering theory, practice, and applications. The list helps readers navigate the extensive distributed systems literature. It's a valuable study guide for learning distributed systems.

    source|Distributed SystemsLearning
  • DIY Dynamic DNS with Netlify API and Bash

    This blog post by Skyler Lewis describes how to set up a free DIY dynamic DNS solution using the Netlify API and a Bash script run via cron. The script checks the current external IP address, compares it against the existing DNS A record for a subdomain managed by Netlify, and updates the record if the IP has changed. The author provides the complete Bash script which uses jq for JSON processing and curl for API calls, along with optional IP caching to reduce unnecessary API requests.

    source
  • Do Elephants Have Souls?

    A philosophical and scientific exploration of whether elephants possess souls, examining cognitive abilities, emotional depth, and consciousness. The article blends philosophy, ethology, and spirituality in contemplating animal consciousness.

    source|Culture
  • Doing a Job – The Management Philosophy of Adm. Hyman G. Rickover

    This 1982 Columbia University speech by Admiral Hyman G. Rickover, the 'Father of the Nuclear Navy,' outlines his management philosophy built on personal responsibility, attention to detail, and determination. Rickover argues that people, not management systems, get things done, and advocates giving subordinates authority early, eliminating shared responsibility structures, requiring direct involvement from managers, and documenting important decisions. He criticizes the Defense Department's rotation system for breeding inexperience and nonaccountability, and emphasizes that hard work and common-sense principles matter more than modern management techniques taught in classrooms.

    source
  • Don't Be a Sucker

    An article examining how people become victims of scams and manipulation, identifying common patterns and vulnerabilities. The piece provides guidance for recognizing and avoiding becoming a sucker. It covers social engineering and psychological manipulation tactics.

    source|LongformCulture
  • Donald Knuth was framed

    Hillel Wayne's newsletter post re-examines the famous story of Donald Knuth's literate programming demonstration where Jon Bentley asked Knuth to write a word-frequency counting program in WEB, and Doug McIlroy responded with a six-line shell script. After reading the original paper, Wayne argues the story is misleading: Knuth was demonstrating his WEB literate programming tool (not competing with McIlroy), much of the code length came from Pascal limitations and hand-rolling a hash trie from scratch, and the shell script solution only works so cleanly because the problem perfectly suits Unix text processing. Wayne concludes it is not the damning indictment of literate programming that popular retelling suggests.

    source|Culture
  • Dotfile madness

    This article argues that programs have lost control of users' home directories by littering them with hidden dotfiles for configuration and data storage. The author advocates for the XDG Base Directory Specification, which defines environment variables like XDG_DATA_HOME, XDG_CONFIG_HOME, XDG_CACHE_HOME, and XDG_RUNTIME_DIR to organize program data into proper subdirectories instead of polluting the home directory. The article explains each XDG variable with practical examples and provides guidance on how developers can migrate existing programs to follow the standard.

    source|Dev Tools
  • Downsides of Offline First

    This RxDB documentation article provides a thorough examination of the downsides and limitations of the local-first/offline-first paradigm. It covers dataset size limitations (browser storage caps in IndexedDB vary by browser), non-persistent browser storage (Safari deletes after 7 days of inactivity), conflict resolution complexity, the misleading nature of 'realtime' replication, eventual consistency challenges, permission and authentication difficulties, client-side database migration complexity, performance overhead from multiple abstraction layers, unpredictable client device capabilities, and the lack of relational data support in offline-first databases.

    source|CRDTs
  • Dreaming of a Parser Generator for Language Design

    Jeff Walker examines why parser generators are rarely used in production compilers or by programming language designers despite decades of academic work on parsing algorithms. He identifies requirements for an ideal parser generator including a separate lexer, Unicode support, unambiguous grammars, flexible grammar specifications with disambiguation rules, AST and concrete syntax tree generation, intransitive operator precedence support, and most importantly excellent compiler error handling. The article argues that current tools fail language designers by offering poor error recovery, limited grammar flexibility, and inadequate usability, forcing most compiler writers to fall back on hand-written recursive descent parsers.

    source|Compilers
  • Earth and Sun

    Bartosz Ciechanowski's interactive blog post provides a detailed visual exploration of the relationship between Earth and Sun. Using interactive 3D demonstrations, it covers the relative sizes of Earth and Sun, elliptical orbits and Kepler's second law, the orbital plane, axial rotation and the difference between sidereal and solar days, axial tilt and its role in causing seasons, the analemma and equation of time, axial precession and its impact on defining tropical vs sidereal years, and the history of leap year corrections from the Julian to Gregorian calendar. The post emphasizes how two seemingly simple motions create a remarkably complex system.

    source|Science
  • Eat This Much – Automatic Meal Planner

    Eat This Much is an automatic meal planning web application that creates personalized meal plans based on users' food preferences, budget, calorie targets, and schedule. It supports various dietary styles including keto, Mediterranean, paleo, vegan, and vegetarian, and provides features like grocery lists, a virtual pantry to reduce food waste, restaurant meal logging, and integration with food delivery services. The platform is available on web and mobile apps.

    source|Culture
  • Effective Mockito

    A guide to using Mockito effectively for unit testing in Java applications. The article covers best practices for mocking objects, common pitfalls to avoid, and techniques for writing maintainable test code. It emphasizes when and how to use mocks appropriately rather than relying on them excessively.

    source|Dev Tools
  • Elastic TabStops: A Better Way to Indent and Align Code

    Nick Gravgaard proposes elastic tabstops as a solution to the long-standing tabs-versus-spaces debate in programming. Instead of tabstops at fixed N-character intervals, elastic tabstops redefine tab characters as cell delimiters (similar to TSV files) where column widths automatically adjust to fit the widest content in vertically adjacent cells. This allows different programmers to view properly aligned code regardless of their font or tab width preferences, and even enables the use of proportional fonts. The page lists numerous implementations across editors and programming tools, including adoption by Go's gofmt.

    source|Algorithms
  • Equity guide for employees at fast-growing companies

    A comprehensive guide for employees to understand equity compensation in startup environments. The article explains stock options, vesting schedules, dilution, and how to evaluate equity packages when joining fast-growing companies. It provides practical frameworks for assessing the true value of equity compensation.

    source|Finance
  • Evaluating Modest SaaS Business Ideas

    Dan Hulton presents a comprehensive framework for evaluating modest, bootstrapped SaaS business ideas. The guide walks through multiple evaluation lenses including personal excitement, problem significance, customer usage patterns, competitive moat, distribution strategy, and market characteristics. Hulton emphasizes that modest SaaS businesses differ from venture-backed startups in their goals and constraints, requiring different evaluation criteria. The framework helps founders systematically assess whether a business idea is worth pursuing by scoring it across these dimensions and identifying potential red flags early.

    source|Business
  • Exploring How and Why Trees 'Talk' to Each Other

    An examination of the mycorrhizal networks, often called the 'wood wide web,' through which trees exchange nutrients and information underground. The article explores the science behind tree communication, including how fungi facilitate resource sharing between trees. It discusses implications for forest ecology and forest management practices.

    source
  • Exploring performance differences between Amazon Aurora and vanilla MySQL

    Plaid's engineering team investigates a surprising performance issue where long-running transactions on Aurora MySQL read replicas degraded the writer instance's performance. The root cause lies in Aurora's shared storage architecture: unlike vanilla MySQL where replicas receive changes via binary log replication and maintain independent storage, Aurora replicas share the same underlying storage volume as the writer. This means long-running replica transactions prevent the writer from purging old undo log entries, causing the history list to grow and slowing down write operations. The article details Plaid's diagnostic process using MySQL internals like the undo log and purge system, and explains why this coupling between reader and writer performance doesn't exist in traditional MySQL replication setups.

    source|CloudDatabases
  • Falling in love with Rust

    Bryan Cantrill of Joyent writes a deeply personal account of his journey to adopting Rust, tracing his awareness of the language from its origins with Graydon Hoare through his initial skepticism about M:N threading to his eventual conversion. He highlights nine things he loves about Rust beyond the ownership model: its elegant error handling with the propagation operator, powerful hygienic macros, the format! macro, include_str!, the Serde serialization library, tuples, integrated testing, the inclusive community, and surprisingly fast performance. His naive Rust rewrite of a statemap visualization tool was 32% faster than his carefully optimized C implementation, fundamentally changing his view of Rust from a C augmentation language to a potential replacement for both C and dynamic languages.

    source|Rust
  • Faux86: A portable

    Faux86 is a portable, open-source 8086 PC emulator for bare metal Raspberry Pi operating without requiring an operating system on the device. Implements 8086/80186 instruction sets with CGA/EGA/VGA graphics emulation and audio capabilities for PC speaker, Adlib, and Soundblaster formats. The C++ project leverages the Circle library for Raspberry Pi hardware interaction and boots from floppy images.

    source|Virtual Machines
  • Feynman: I am burned out and I'll never accomplish anything

    An article discussing Richard Feynman's experience with academic burnout and how he recovered his passion for physics. The piece explores Feynman's struggles with the pressure of expectations and his path to rediscovering joy in scientific inquiry. It offers insights into managing burnout in intellectually demanding careers.

    source|CareerCulture
  • Finally on CBS the football matches the business cards

    John Teti's Doink-O-Rama column delivers a witty, detailed critique of CBS's 2021 NFL score bar redesign. He argues that while Fox and NBC design their football graphics to tell coherent visual stories about the sport, CBS's new flat design with the TT Norms font feels like a corporate branding exercise rather than something crafted for football. The previous understated CBS look honored a tradition of getting out of the viewer's way, but the 2021 revision applies a company-wide visual identity that strips away artful details and replaces them with harsh, bold color blocks. The column also features humorous segments about sideline moments, first-down markers, and NFL game picks.

    source
  • Financial Modeling for Startups: An Introduction

    Fivecast Financial's guide walks startup founders through building a basic financial model in Excel, focusing on tracking cash flows rather than full accrual accounting. The tutorial covers forecasting revenue using subscription-based modeling (new subscriptions, attrition rates, revenue per subscription), itemizing expenses with emphasis on personnel costs, calculating net income, projecting cash balance, and determining how much investment capital to raise. The guide includes a downloadable example model with three tabs (Model, Personnel, Summary) and walks through a concrete scenario where a company needs to raise $10M to bridge the gap to profitability in 2021.

    source|Startups
  • Financial Statements: A Beginner's Guide

    This guide explains how to read and interpret financial statements including income statements, balance sheets, and cash flow statements. It enables financial literacy. The piece demystifies financial reporting. It teaches financial analysis.

    source|Finance
  • Finding Harmony: A Love Story of Music and Connection

    Finding Harmony: A Love Story of Music and Connection Dravus and Hopseed had always been curious about classical music, but had never found the opportunity to explore it until that fateful evening when they went to the symphony for the first time. They were both enthralled by the music and the experience of being in the beautiful concert hall. Hopseed was especially moved by the power and bea...

    Culture
  • Firecracker: Start a VM in less than a second

    Julia Evans explores using AWS Firecracker from a DIY perspective for her command-line learning game that gives each player their own VM. She chose Firecracker over containers because she wanted users to have full root access to a realistic Linux environment, and over qemu because Firecracker can boot a VM in under a second versus 20+ seconds. The post provides complete hello-world shell scripts for starting Firecracker VMs, explains how to build custom filesystem images using Docker containers, and describes her Go HTTP service for managing VMs. She also covers practical details like running Firecracker with nested virtualization on DigitalOcean and connecting VMs to the Docker bridge network.

    source|Hardware
  • Fortune du jour
  • FoundationDB: A distributed unbundled transactional key value store

    A technical overview of FoundationDB, a distributed database system designed for high performance and ACID consistency. The article explains its unique architecture separating transaction processing from storage, and how it achieves scalability while maintaining strong consistency guarantees. It covers use cases where FoundationDB excels.

    source|Distributed Systems
  • Free and liberated e-books

    Standard Ebooks is a volunteer-driven project that produces new, professionally formatted editions of public domain ebooks that are free and open source. Starting from transcriptions by Project Gutenberg, they apply a rigorous style manual with modern typography (curly quotes, proper dashes), do complete proofreading against original page scans, add rich metadata and semantic markup, create unique covers from public domain fine art, and build ebooks with state-of-the-art features like hyphenation and popup footnotes. All work is tracked in Git and released into the public domain.

    source
  • Free Machine Learning crash course from Google

    Google's Machine Learning Crash Course is a free, self-paced introduction to machine learning featuring animated videos, interactive visualizations, and hands-on exercises. The course covers fundamental topics including linear regression, logistic regression, classification, working with numerical and categorical data, neural networks, embeddings, and large language models. It also addresses real-world considerations like production ML systems, AutoML, and ML fairness, with modules on identifying and mitigating biases in data and models.

    source|ProgrammingML & AILearning
  • From Nand to Tetris: Building a Modern Computer from First Principles

    Nand2Tetris is a free, open-source educational project by Noam Nisan and Shimon Schocken (MIT Press) that provides all the lectures, project materials, and tools needed to build a complete general-purpose computer system and modern software hierarchy from the ground up starting with just NAND gates. The course is taught at over 400 universities, high schools, and bootcamps worldwide and is split into two parts: Part I covers hardware (chapters 1-6) and Part II covers software (chapters 7-12). Students range from high schoolers to Ph.D. candidates to senior engineers.

    source|Systems
  • G1GC Fundamentals: Lessons from Taming Garbage Collection

    HubSpot's engineering team provides a comprehensive deep dive into Java's G1GC (Garbage First Garbage Collector) based on their experience tuning thousands of JVM-based microservices. The article covers G1GC's core architecture including regions, generational collection with Eden/Survivor/Tenured spaces, evacuation-style collections, and the Multi-Phase Concurrent Marking Cycle that enables Tenured space to be cleaned in batches. Key practical insights include that Eden size doesn't affect individual STW pause times but halving it doubles overall GC time, that InitiatingHeapOccupancyPercent should be tuned relative to the working set size, and that Humongous objects are G1GC's biggest weakness due to their impact on Free space and Full GC risk.

    source
  • Gentle introduction to GPUs inner workings

    An accessible introduction to how GPUs (Graphics Processing Units) work at a hardware and algorithmic level. The article explains parallel processing, SIMD operations, and why GPUs excel at certain computational tasks. It covers the evolution from graphics-specific to general-purpose GPU computing.

    source
  • Get Started Making Music

    Ableton's Learning Music is a free interactive web-based course that teaches the basics of music making directly in the browser with no prior experience or equipment required. The site covers beats, notes and scales, chords, basslines, melodies, and song structure through hands-on exercises where users click boxes to toggle musical patterns on and off. Advanced topics include building major and minor scales, modes, pentatonic scales, diatonic triads, inversions, voicings, and seventh chords. The course is available in multiple languages.

    source|Culture
  • Get your work recognized: write a brag document

    Julia Evans advocates for maintaining a 'brag document' that lists your work accomplishments throughout the year, rather than scrambling to remember them at performance review time. The document helps you and your manager recall important work like mentoring interns, security projects, and migrations that might otherwise be forgotten. She recommends sharing it with managers and peer reviewers, using it to notice patterns in your work, and including fuzzy contributions like improving code review culture. The post includes a detailed template covering projects, collaboration, mentorship, design docs, company building, and skills learned, plus ideas for running brag workshops with teammates.

    source|Career
  • Getting more than what you've asked for: The Next Stage of Prompt Hacking

    This repository demonstrates indirect prompt injection vulnerabilities affecting language models integrated with external applications, showing how LLMs can be compromised when processing retrieved content from websites, emails, or repositories. Provides proof-of-concept demonstrations across multiple attack vectors including remote control, data exfiltration, and persistent compromise. Shows that prompt injections can be as powerful as arbitrary code execution when integrated with retrieval capabilities.

    source|Security
  • Getting Past 'Ampersand-Driven Development' in Rust

    Evan Schwartz from Fiberplane explains Rust's ownership and borrowing system using intuitive analogies of children sharing toys. Shared references (&variable) are like lending a toy with the rule 'look but don't touch, and give it back'; mutable references (&mut variable) are like lending a coloring book to color one page; owned values are like giving a toy away permanently; and Rc/Arc reference-counted pointers are like party decorations that stay up until the last guest leaves. The article emphasizes that mutable references are exclusive to prevent unexpected mutations, and that with practice, choosing between these types becomes intuitive rather than the trial-and-error 'ampersand-driven development' that new Rust developers experience.

    source|Rust
  • Getting Started with Firecracker on Raspberry Pi

    This tutorial walks through setting up AWS Firecracker microVMs on a Raspberry Pi 4B running 64-bit Ubuntu. It covers installing Firecracker, Jailer, and Firectl (compiling Firectl from source with Go 1.14), downloading pre-built kernel and filesystem images, configuring tap networking with IP forwarding and iptables rules, and launching a first microVM. The article also addresses practical issues like resizing the root filesystem to install packages, fixing apt configuration problems in the minimal Ubuntu image, and setting up DNS resolution inside the VM.

    source|Hardware
  • Gitfs: Version Controlled File System

    A tool that presents a Git repository as a mountable filesystem where different commits appear as directories. Users can navigate Git history as if browsing versioned directories. The project demonstrates innovative applications of FUSE (Filesystem in Userspace) for Git integration.

    source
  • GitHub Issues-only project management

    A guide to using GitHub Issues as a complete project management system instead of separate tools. The article covers issue templates, labels, milestones, and workflows for team coordination. It demonstrates lightweight project management.

    source|Dev Tools
  • Gitlet.js – Git implemented in 1k lines of JavaScript

    Gitlet is a heavily commented JavaScript implementation of Git written by Mary Rose Cook to demonstrate how Git works under the covers. The project includes a companion essay 'Git in six hundred words' that explains Git fundamentals through a step-by-step walkthrough of init, add, commit, branch, checkout, and merge operations, showing how Git stores data as blob, tree, and commit objects. The annotated source code implements core Git commands as readable JavaScript functions, serving as both a working Git implementation and an educational resource for understanding Git's internal object model, index, refs, and merge algorithms.

    source
  • Google Reader

    Google's official blog post from March 2013 announcing the shutdown of Google Reader, effective July 1, 2013. Software Engineer Alan Green explains the two reasons for the decision: declining usage of Google Reader and Google's strategic focus on pouring energy into fewer products for a better user experience. The brief post offers a three-month sunset period for users to find alternatives and directs them to Google Takeout for exporting their subscription data.

    source|ProgrammingCulture
  • Graph2Plan: Learning Floorplan Generation from Layout Graphs

    Graph2Plan introduces a deep learning framework for automated floorplan generation that combines generative modeling with user-in-the-loop design. Users provide sparse constraints via a layout graph specifying room counts and relationships, which the system converts into a floorplan satisfying both layout and building boundary constraints. The core neural network uses graph neural networks for processing layout graphs and conventional image convolution for building boundaries. Trained on RPLAN, a dataset of 80K annotated floorplans, Graph2Plan first generates a raster floorplan image and then refines it into a set of room-representing boxes.

    source
  • Greatest Java apps

    This article catalogs notable applications built with Java, showcasing language's diversity of use cases from enterprise to creative projects. It celebrates Java ecosystem. The piece demonstrates Java's range. It documents application examples.

    source|Programming
  • Guide to JavaScript Frameworks

    This comprehensive living document by John Hannah catalogs all known front-end JavaScript frameworks, serving as a reference guide for developers trying to navigate the crowded landscape. While acknowledging that most developers will use one of the 'Big Three' (React, Angular, and Vue), the guide covers dozens of other frameworks and UI libraries, explaining their strengths, weaknesses, and use cases. It helps developers understand the broader ecosystem, discover niche frameworks that might be better suited for specific projects, and stay informed about emerging alternatives.

    source|Programming
  • Guide to Remote Work

    This guide covers strategies for remote work success, discussing setup, communication, productivity, and work-life balance in distributed environments. It helps remote workers thrive. The piece provides practical remote work guidance. It enables remote work transition.

    source|Career
  • Hacking Together an E-ink Dashboard

    Andrew Healey builds a small e-ink dashboard using a Waveshare 2.7-inch tri-color e-Paper HAT (~$20) connected to a Raspberry Pi. The dashboard displays current weather from Open Weather API and BBC news headlines from News API, rendered in black and red layers using Python's Pillow library and the epd-library. The article walks through the code for fetching weather data via PyOWM, parsing news headlines with manual text wrapping, creating image layers for the two-color display, and scheduling updates every 30 minutes during waking hours to avoid ghosting.

    source|Hardware
  • Halt and Catch Fire Syllabus

    Ashley Blewer created a semester-long curriculum of 15 classes built around the TV series Halt and Catch Fire (2014-2017), designed for self-forming 'watching clubs' to discuss technology history from the 1980s-1990s. Each class includes casual viewing appetizers, an RFC from the IETF for reflection, an emulated vintage computer in the browser, discussion themes and prompts, and supplementary readings. The syllabus provides episode summaries for those who can't watch every episode and includes content warnings for relevant episodes.

    source|Learning
  • Hard to discover tips and apps for macOS

    A collection of useful but lesser-known tips, tricks, and applications for macOS users. The article covers hidden features, productivity shortcuts, and niche software that enhance the Mac experience. It serves as a resource for power users and those wanting to improve their macOS workflow.

    source
  • Harvard Extension School: resume and cover letter guide

    Career advice from Harvard Extension School on crafting effective resumes and cover letters. The guide emphasizes clarity, accomplishment-focused writing, and tailoring applications to specific opportunities. It provides concrete examples and formatting recommendations for professional document creation.

    source|Career
  • Hash Array Mapped Tries (HAMT) to the Rescue

    This article introduces Hash Array Mapped Tries (HAMTs) as a memory-efficient data structure for key-value storage at scale, motivated by the author experience building a peer-block registry for IPFS Bitswap protocol where naive hash tables could not handle the volume of content requests. HAMTs combine hash tables with tries, using bitmap indexing with 32-bit bitmaps and 5-bit prefixes to achieve O(1) insert, search, and delete operations while maintaining a compact memory footprint. The author walks through the data structure from simple 2-bit examples to full 32-bit implementations, covering search and insert operations, and notes HAMTs are widely used in Web3 systems like Filecoin and in languages like Clojure.

    source|Algorithms
  • Hermes - Javascript engine optimized for React Native

    Hermes is a JavaScript engine developed by Facebook (Meta) that is optimized for fast start-up of React Native apps through ahead-of-time static optimization and compact bytecode. The GitHub README provides instructions for building the Hermes CLI tools from source on macOS/Linux and Windows using cmake and Ninja, and demonstrates running JavaScript through the compiled binary. The project is MIT licensed and encourages community contributions through a contributing guide and code of conduct.

    source|Web Dev
  • Hiring the first head of marketing at a startup

    A guide for founders on recruiting and hiring the first dedicated marketing executive for their startup. The article covers what skills to look for, how to structure the role, and expectations for early-stage marketing leadership. It provides insights into finding the right marketing hire and setting them up for success.

    source|StartupsBusiness
  • Horrible insurance kerfuffle gone good

    A story about a complex or frustrating insurance situation that eventually resolved favorably. The article likely details how persistence and negotiation led to a better outcome. It serves as an anecdote about navigating insurance challenges.

    source|ProgrammingFinance
  • Hosting SQLite Databases on GitHub Pages

    A technique for hosting and serving SQLite databases directly from GitHub Pages using WebAssembly. The article explains how to make databases available on static hosting without backend servers. It enables creative use cases like shareable databases and collaborative data projects.

    source|Databases
  • How a Kalman filter works

    This visual tutorial explains Kalman filters using colorful illustrations and progressive mathematical derivation. Starting with a robot navigation example, it shows how Kalman filters combine uncertain predictions about a system's state (position and velocity) with noisy sensor measurements to produce estimates more accurate than either source alone. The key insight is that both the prediction and measurement are modeled as Gaussian distributions, and multiplying two Gaussians produces a new Gaussian whose parameters can be computed via the Kalman gain matrix. The article derives the complete predict-update cycle equations from first principles, showing how covariance matrices track correlations between state variables and how external influences and process noise are incorporated.

    source|Science
  • How big tech runs tech projects and the curious absence of Scrum

    Gergely Orosz surveys project management practices across 100+ companies and finds that Big Tech companies notably avoid Scrum, instead relying on empowered autonomous teams where engineers lead projects and choose their own methodologies. The article explains how organizational factors like team autonomy, engineer-to-engineer communication, and first-class developer tooling reduce the need for heavyweight processes. It also defends Scrum's value in specific contexts like kitchen-sink teams, newly forming teams, and organizations shipping infrequently, while arguing that the key to effective project management is trust, autonomy, and iterative improvement rather than any single methodology.

    source
  • How Browsers Lay Out Web Pages

    A technical explanation of browser rendering engines' layout algorithm (reflow). The article explains how browsers calculate element positions and sizes based on CSS and the document tree. It provides insights into browser performance optimization and why certain CSS practices are slow.

    source
  • How browsers work

    Tali Garsiel's comprehensive primer on browser internals covers the architecture and rendering pipeline of modern web browsers, focusing on WebKit and Gecko engines. The article explains how browsers parse HTML and CSS, construct the DOM and render trees, perform layout calculations, and paint pixels to the screen. Originally published in 2011 and hosted on web.dev, it remains one of the most detailed publicly available explanations of the multi-step process browsers use to transform raw HTML, CSS, and JavaScript into rendered web pages.

    source
  • How Discount Brokerages Make Money

    Patrick McKenzie provides a detailed analysis of how discount brokerages generate revenue, revealing that trading commissions are a small and shrinking part of the business. The dominant revenue source is net interest income, accounting for 50-67% of revenue, earned from the spread between what brokerages pay customers on cash balances and what they earn investing that cash. Other revenue streams include asset management fees, securities lending, and the controversial practice of payment for order flow. The article explains why brokerages like Schwab and Robinhood can offer zero-commission trading while remaining highly profitable.

    source|Finance
  • How do startups get their content marketing to work?

    Julian Shapiro explains on TechCrunch how startups can make content marketing work by sharing data from Growth Machine and Bell Curve clients. The key insights are to write fewer but more in-depth articles optimized for reader engagement rather than chasing backlinks, since Google now reliably detects content quality through engagement signals. The article demonstrates that backlinks matter less than they used to, and that startups should focus on being the last site a searcher visits by covering topics comprehensively. It also covers conversion optimization strategies like naturally segueing product pitches into content and using retargeting ads on blog readers.

    source|StartupsBusiness
  • How Do Venture Capitalists Make Decisions?

    This Stanford/NBER research paper by Gompers, Gornall, Kaplan, and Strebulaev surveys 885 institutional venture capitalists at 681 firms to understand how they make decisions across eight areas including deal sourcing, investment selection, valuation, deal structure, and post-investment value-added. The key finding is that VCs consider the management team more important than business characteristics like product or technology when selecting investments, and they attribute more of the likelihood of success or failure to the team. Among deal sourcing, deal selection, and post-investment value-added, VCs rate deal selection as the most important contributor to value creation.

    source|VC & Fundraising
  • How it feels to learn JavaScript in 2016

    A satirical dialogue by Jose Aguinaga that humorously illustrates the overwhelming complexity of the modern JavaScript ecosystem in 2016. Written as a conversation between a bewildered developer who just wants to display data from a server and an enthusiastic front-end engineer, it lampoons the dizzying array of tools, frameworks, transpilers, module bundlers, and task runners (React, Babel, Webpack, SystemJS, TypeScript, etc.) required to accomplish simple web tasks. The piece went viral for perfectly capturing developer fatigue with JavaScript tooling churn.

    source|Programming
  • How my role as CTO has changed as we've grown from 1 to 100 engineers

    A personal narrative about evolving responsibilities as a CTO during company growth. The article discusses the transition from hands-on technical work to architecture and leadership. It provides insights into scaling technical organizations and adapting to new roles.

    source|Career
  • How OIDC Works

    This Teleport blog post explains how OpenID Connect (OIDC) works as an authentication layer built on top of the OAuth authorization protocol. It covers the distinction between authentication and authorization, compares OIDC with SAML, and explains how OIDC extends OAuth by issuing ID tokens (JWTs containing user identity claims) alongside access tokens. The article walks through standardized OIDC scopes and includes a detailed case study demonstrating the Authorization Code Grant flow using Teleport SSO with Okta as the identity provider.

    source|Security
  • How the biggest consumer apps got their first 1k users

    An analysis of how major consumer applications achieved their initial user base. The article examines strategies from Airbnb, Instagram, and other successful startups. It covers product-market fit, growth hacking, and viral mechanics.

    source|Programming
  • How to be a Manager – A step-by-step guide to leading a team

    How to be a Manager is a comprehensive step-by-step guide to leading a team effectively, covering responsibilities from hiring and onboarding to performance management and professional development. The guide provides practical frameworks for one-on-ones, delegation, conflict resolution, and building team culture. It serves as an accessible resource for first-time managers and experienced leaders seeking to improve their leadership skills.

    source
  • How to Build a Great Series A Pitch and Deck

    YC Series A Program Manager Janelle Tam compiles detailed advice on building a great Series A pitch deck based on hundreds of hours working with YC founders. The guide recommends starting with a fundraising vertebrae exercise of 10-15 bullet points, then building slides around each point covering title, traction teaser, problem, solution, in-depth traction, market, competition, vision, team, and use of funds. Key themes include optimizing for clarity over beauty, showing trends rather than point-in-time numbers, quantifying impact with concrete metrics, and presenting from the customer's perspective. The article also covers tips for video pitching and iteratively refining the deck based on investor feedback.

    source|VC & Fundraising
  • How to build an open source business

    An exploration of different business models for monetizing open source software. The article covers approaches like dual licensing, support services, hosted versions, and premium features. It provides frameworks for sustaining open source projects financially.

    source|Business
  • How to Create a Hex Tile Grid Map in Excel

    A step-by-step tutorial on creating hexagonal tile grid maps in Microsoft Excel using the built-in Shapes tools. The guide walks through inserting hexagon shapes, rotating them, adjusting size and colors, and arranging them into a grid layout to create geographic tile maps where each hex represents a state or region. The technique uses standard Excel features without requiring any add-ins or programming.

    source|Design
  • How to Create a Junction in Windows

    How to Create a Junction in Windows Create a junction in Windows using the MKLINK command. mklink /j source-path target-path Example: create a junction named C:\Volatile to the S:\Volatile folder.

  • How to De-Risk a Startup

    A strategic guide for founders on identifying and reducing risks in startup ventures. The article covers market risk, technical risk, team risk, and execution risk. It provides frameworks for validation and de-risking techniques before scaling.

    source|Startups
  • How to Destroy the Earth

    A tongue-in-cheek scientific essay by qntm that rigorously examines various hypothetical methods for literally destroying planet Earth, defined as reducing it to something other than a planet. The author emphasizes that Earth is a 4.55-billion-year-old, nearly 6 sextillion tonne ball of iron that has survived countless asteroid impacts, making actual destruction far harder than movies suggest. The essay catalogs methods ranging from the theoretically possible (like hurling it into the Sun or disassembling it with von Neumann machines) to the purely fictional, while distinguishing between merely making Earth uninhabitable versus truly destroying it as a celestial body.

    source|Culture
  • How to disagree with someone more powerful than you

    Amy Gallo's Harvard Business Review article addresses the challenge of disagreeing with someone who holds more power than you, such as a boss or senior colleague. The piece provides practical advice for deciding whether it's worth speaking up when you think an initiative won't work or a timeline is unrealistic, and offers strategies for how to express disagreement constructively while maintaining the relationship and your professional standing.

    source|Career
  • How to do distributed locking

    Martin Kleppmann critiques the Redlock distributed locking algorithm built on Redis, arguing it is unsafe for correctness-critical use cases. He distinguishes between locks used for efficiency (where a single Redis instance suffices) and locks needed for correctness (where Redlock's timing assumptions about bounded network delay, process pauses, and clock drift can be violated). The article demonstrates how GC pauses, network delays, and clock jumps can cause two clients to simultaneously believe they hold the same lock, and explains how fencing tokens solve this problem - a feature Redlock lacks. Kleppmann recommends using proper consensus systems like ZooKeeper for correctness-critical locking instead.

    source|Distributed Systems
  • How to hack Hacker News (and consistently hit the front page)

    The co-owner of Simple Analytics shares a repeatable strategy for consistently reaching the Hacker News front page, which they call 'The War Room.' The three-step process involves: building a newsbot that monitors relevant topics using the HN API and Google Alerts with ChatGPT scoring, dropping everything to investigate and write about breaking news when alerted (War Room mode), and carefully adding minimal self-promotion only in the final thoughts section after providing substantial value. Their key formula is timeliness plus added perspective equals front page placement, and the strategy produced significant SEO side effects with articles earning hundreds of backlinks from dozens of domains.

    source
  • How to have a billion dollar exit with zero capital gains tax

    An extensive Axiom Alpha guide explaining how U.S. Opportunity Zone tax laws can be used to eliminate capital gains taxes on business exits and investments. The article details how Qualified Opportunity Funds (QOFs) and Qualified Opportunity Zone Businesses (QOZBs) work, showing that most asset-light tech startups, e-commerce companies, and small businesses can qualify by operating from offices in designated low-income census tracts. It walks through detailed examples including a billion-dollar startup exit with zero capital gains tax, explains the double benefit of combining OZ rules with real estate depreciation, and covers practical considerations like cloud infrastructure agreements, acquisitions, and stock buybacks.

    source
  • How to Learn Advanced Mathematics Without Heading to University - Part 3

    Part three of QuantStart series on self-studying advanced mathematics covers the third-year curriculum of a four-year masters-level mathematics degree with an applied focus relevant to quantitative finance. The article covers topics including Measure Theory, Functional Analysis, Partial Differential Equations, and advanced Probability Theory that prepare students for Stochastic Calculus, Probabilistic Machine Learning, and Bayesian Econometrics. The guide recommends specific textbooks for each subject and explains how these abstract mathematical foundations connect to practical quantitative applications.

    source|Science
  • How to make 'localhost' slightly less local

    Bryce Wray explains how to test a locally-developed website on other devices on your LAN by pointing them to the dev machine's IP address. The article covers finding your IP address on macOS (ifconfig), Windows (ipconfig), and Linux (hostname -I), then provides specific configuration instructions for multiple static site generators and tools including Eleventy, Hugo, Astro, Next.js, Gatsby, Nuxt.js, Jekyll, and VS Code's Live Server extension. He recommends using shell scripts to avoid repeatedly typing complex dev server commands.

    source|Networking
  • How to pick the least wrong colors

    A guide to making thoughtful color choices for design and visualization. The article covers color theory, accessibility considerations, and emotional impact. It provides practical frameworks for color selection in various contexts.

    source|Design
  • How to present to executives

    Will Larson's guide explains why presenting to executives is uniquely challenging: each executive has a particular way of consuming information, and mismatches in communication style can derail meetings. He recommends writing structured documents using the SCQA format (Situation, Complication, Question, Answer) and optionally the Minto Pyramid Principle, then gathering stakeholder feedback before the meeting. The article also covers common mistakes like fighting feedback, evading problems, presenting questions without proposed answers, and fixating on preferred outcomes.

    source|Career
  • How to professionally say

    A humorous and practical reference tool by Akash Rajpurohit that translates blunt workplace thoughts into professional corporate language. For each entry, it shows what you might feel like saying (e.g., 'That sounds like a horrible idea' or 'Stop micromanaging') alongside a polished professional alternative. The site covers dozens of common workplace scenarios including pushing back on extra work, asking for raises, handling credit-taking, setting boundaries around working hours, and dealing with micromanagement.

    source|Career
  • How to put machine learning models into production

    A Stack Overflow blog post discussing why 87-90% of data science projects never make it to production and how to bridge the gap between model building and deployment. The article identifies three critical areas to address before starting any ML project: data storage and retrieval (batch vs. real-time, cloud vs. on-premise), frameworks and tooling (evaluating efficiency, popularity, platform support), and feedback and iteration (monitoring model decay, continuous integration). It walks through a detailed case study of an ad-click prediction system using Python, TensorFlow, TFX pipelines, and Google Cloud Platform.

    source|ML & AI
  • How to Read a Log Scale

    This Datawrapper blog post explains how to read logarithmic scales on charts, focusing on the key insight that equal distances on a log scale represent equal percentage changes rather than equal absolute changes. Using New Zealand tourism data as an example, the author demonstrates how log scales reveal relative growth rates and proportional drops that linear scales obscure — such as showing that the WWII drop in arrivals was proportionally almost as steep as the subsequent decades-long recovery. The article is the first in a three-part series on understanding log scales in data visualization.

    source|Science
  • How to Recalculate a Spreadsheet

    How to Recalculate a Spreadsheet explains the algorithms and engineering challenges behind spreadsheet recalculation, covering dependency tracking, topological sorting, and incremental updates. The article demonstrates that even seemingly simple tools require sophisticated systems engineering. It provides insights valuable for developers building spreadsheet-like applications or data analysis tools.

    source|Databases
  • How to Recognize the Real Senior Developer?

    Ivana Veljovic argues that years of experience alone don't make a senior developer, and identifies key qualities that distinguish true seniors from developers who merely have seniority. These qualities include deep internalization of programming principles beyond syntax, the ability to write code independently without constantly relying on Stack Overflow, thinking objectively rather than being bound to personal preferences, balancing working software with code quality and technical debt, and the ability to teach and mentor others. The article emphasizes that senior developers see the bigger picture and make decisions based on minimizing negative outcomes.

    source
  • How to say no

    A guide on declining requests and saying no effectively in professional and personal contexts. The article covers techniques for refusal without guilt, protecting time and boundaries. It emphasizes the importance of selective commitments for productivity.

    source|Career
  • How to Self-Publish a Novel in 2017

    A practical guide for authors self-publishing novels including editing, cover design, formatting, and distribution channels. The article covers platforms like Amazon KDP and strategies for reaching readers. It demystifies the self-publishing process.

    source
  • How to Ship Side Projects

    Advice on completing and releasing personal projects and side hustles. The article covers maintaining momentum, managing scope, and launching effectively. It addresses the common challenge of side projects remaining incomplete.

    source|Startups
  • How to start a startup without ruining your life

    Rik Lomas, founder of SuperHi and former co-founder of code school Steer, shares practical advice for first-time startup founders based on six years of working at startups and four years of running his own. The guide covers validating ideas through customer development before building anything, finding co-founders you trust, hiring staff vs. freelancers vs. agencies (with real salary ranges for London and New York), the importance of design, keeping your product simple, and effective sales and marketing strategies. He also addresses the emotional roller coaster of startup life, emphasizing the importance of maintaining work-life balance, having hobbies outside work, and pushing through low periods.

    source|Startups
  • How to stay focused while working on your startup and having a 9 to 5

    Fernando Pessagno shares practical advice for balancing a full-time job with a side business, drawn from years of working on ResumeMaker.online alongside regular employment. The six strategies include blocking out dedicated time for each activity, making intentional sacrifices in leisure time, planning the week ahead on Sundays, sticking to a consistent productivity framework, taking breaks to avoid burnout, and joining communities of like-minded entrepreneurs. The article pushes back against the all-or-nothing startup mentality, arguing that slow and strategic progress is a perfectly valid path to building a business.

    source|Startups
  • How to teach yourself hard things

    Julia Evans outlines specific learning skills she has developed over time for teaching herself hard programming topics. The key skills include identifying what you don't understand by turning vague confusion into specific questions, having confidence in knowledge you're certain about, asking good questions to the right communities, and doing effective research across documentation, books, source code, and mailing lists. She emphasizes that translating 'I'm confused' into concrete questions is the most important skill, and illustrates with her experience learning Rust references and borrowing.

    source
  • How to unc0ver a 0-day in 4 hours or less

    An article exploring the topic of how to unc0ver a 0-day in 4 hours or less. The piece examines key concepts and practical applications relevant to the subject matter. It provides insights and guidance for readers interested in understanding this area better.

    source|Security
  • How to validate your startup idea quickly

    A practical framework for rapidly testing startup assumptions before investing heavily. The article covers customer interviews, prototype testing, and metric collection. It provides lean startup methodology for idea validation.

    source|StartupsNetworking
  • How to Write a Computer Emulator

    This article covers writing computer emulators from scratch, discussing architecture simulation and instruction set implementation. It provides emulator development guidance. The piece demystifies emulation. It enables emulation projects.

    source|Virtual MachinesCareer
  • How to Write a Spelling Corrector

    Peter Norvig demonstrates how to build a functional spelling corrector in about 21 lines of Python, written during a single plane ride. The program uses Bayes' theorem to find the most probable correction by combining a language model (word frequency from a million-word corpus) with a simple error model based on edit distance. It generates candidate corrections within one or two edits (deletions, transpositions, replacements, insertions) and selects the most probable known word, achieving 75% accuracy on a development test set at 41 words per second.

    source|Career
  • How to write a tiny compiler

    This article provides concise tutorial for building a minimal compiler, covering essential concepts without excessive complexity. It demystifies compiler creation. The piece enables compiler learning. It teaches compiler fundamentals.

    source|CompilersCareer
  • How to write a toy JVM

    This tutorial walks through implementing a toy Java Virtual Machine in Go, starting from parsing the class file format (the famous CAFEBABE magic bytes, constant pool, fields, methods, and attributes) through to executing bytecode. The author demonstrates how JVM works as a stack machine by implementing a simple add() method, showing how iload, iadd, and ireturn instructions manipulate the operand stack. The article also outlines what a complete JVM would need: all 200+ instructions, object model, method dispatching, garbage collection, exception handling, and runtime classes.

    source|Career
  • How to Write Articles and Essays Quickly and Expertly

    Stephen Downes explains a strategy for writing articles and essays quickly by making your first draft your final draft. The method centers on identifying which of four types of discursive writing you're doing - argument, explanation, definition, or description - each of which has a distinct structure with specific indicator words. You start by writing your second paragraph (the lede that summarizes your whole piece), then expand each sentence into its own section using the appropriate structural pattern. The article details how complex forms emerge when each element of one type contains nested elements of another type, allowing writers to go deeper into structure until they've said enough.

    source|NetworkingCareer
  • How to write technical posts so people will read them

    Sandy Maguire offers practical advice for writing technical blog posts that people will actually read. The key principle is that readers don't inherently care about your content — you get about four sentences to convince them it's worth their time. The article recommends structuring posts as a tree with descriptive headings, writing strong motivations framed as solutions to problems, keeping one idea per paragraph with the key point in the first sentence, and anticipating conceptual stumbling blocks. The author emphasizes giving lots of examples, making content easy to skim, and ending on a high note.

    source|Career
  • How we use Rust in our mobile SDK

    FullStory's experience integrating Rust into their mobile SDK demonstrates pragmatic approaches to leveraging systems language performance while maintaining compatibility with mobile platforms. The article details specific performance-critical sections where Rust provides measurable improvements over traditional mobile development approaches, along with challenges encountered during integration. It covers tooling considerations, dependency management, and strategies for shipping Rust code within mobile SDKs distributed to thousands of applications. The case study provides valuable lessons for teams considering Rust adoption in constrained mobile environments.

    source|Rust
  • Hyperscan: High-performance multiple regex matching library from Intel

    Hyperscan is a high-performance regular expression matching library from Intel, released as open source under a BSD license. It supports PCRE syntax, simultaneous multi-pattern matching, and streaming mode for cross-packet matching in networking scenarios. The library is optimized using Intel SIMD instructions and is designed for use cases like deep packet inspection, intrusion detection/prevention systems, and firewalls, having been integrated into open-source security tools like Snort and Suricata. Performance benchmarks show near-linear scalability across CPU cores and high throughput in gigabits per second.

    source
  • I accidentally loaned all my money to the US government

    A humorous account of the author trying to buy US government I-bonds on treasurydirect.gov during a window offering ~8.5% guaranteed returns, only to accidentally trigger two rejected $10,000 purchases due to the site terrible UX. The Treasury Department took the money immediately but needed 8-10 weeks to process refunds, leaving the author with $173.12 in savings and just $25 in actual I-bonds while locked out of the purchase window. The post highlights the frustratingly poor user interface of treasurydirect.gov, including a virtual keyboard for password entry and no visible interest rates when selecting securities.

    source|ProgrammingLongform
  • I bricked then recovered my reMarkable 2

    The author describes how they accidentally bricked their reMarkable 2 tablet by copying ARM library files to the root filesystem instead of /opt while trying to install rsync without using the official package manager. The misplaced shared libraries caused the device to enter a boot loop. Recovery involved using the reMarkable 2's pogo pins and a USB-C breakout board to trigger the device's recovery mode via its Freescale SoC, which exposed the internal storage as a USB mass storage device. By imaging the disk, mounting the partitions, and restoring the good library files from the inactive backup partition to the active one, the author successfully recovered the device.

    source
  • I made an app that lets you split a file into horcruxes

    Horcrux-UI enables users to split sensitive files into encrypted fragments called horcruxes without password memorization, using Shamir's Secret Sharing Scheme adapted from HashiCorp's Vault. Built with Fyne framework in Go, it requires only a threshold number of fragments to reconstruct the original file. Users can disperse fragments across multiple secure locations or channels, making interception substantially harder.

    source
  • I Quit Being So Accommodating

    This anonymous 1922 essay from The American Magazine tells the story of a man who spent 35 years being everyone's 'Good Fellow' — running errands, attending to favors, and accommodating every request — just as his druggist father had done before him. After being passed over for a promotion because his boss recognized he was too busy pleasing everyone to be effective, he resolved to stop being indiscriminately accommodating. He concludes that one's chief loyalty must be to family and work, that indiscriminate helpfulness pauperizes recipients, and that real achievement requires commanding one's own time.

    source|Career
  • I started a paper website business

    An article about starting a business that creates paper-based websites and printed digital experiences.

    source|Business
  • I wasted $40k on a fantastic startup idea

    The author recounts spending $40,000 and nine months building GlacierMD, a startup that used structured clinical trial data to provide evidence-based answers to common medical questions through automated meta-analyses. Despite enthusiastic responses from doctors and consumers, no one was willing to pay — consumers expected free health info, ad revenue was too low at ~$0.50/user/year, and doctors saw no direct financial benefit from prescribing better treatments. The key lesson was that creating value for users alone is not enough; a viable business must create value for all stakeholders, and validating the business model should come before building the product.

    source|Startups
  • I wouldn't invest in open-source companies

    This article argues against investing in pure open-source businesses, discussing business model challenges and why venture returns are difficult. It provides skeptical investment perspective. The piece examines open-source business models. It questions investment assumptions.

    source
  • I, Pencil

    Leonard E. Read's classic 1958 essay, narrated from the perspective of a pencil, illustrates the extraordinary complexity of producing even the simplest everyday object. The pencil traces its genealogy through cedar logging in Oregon, graphite mining in Sri Lanka, clay from Mississippi, lacquer ingredients, brass ferrules, and erasers made with rapeseed oil from Indonesia, demonstrating that no single person on earth knows how to make a pencil. The essay uses this observation to argue for free markets and against central planning, showing how millions of people cooperate through the price system without any master mind directing them. It includes an afterword by Milton Friedman praising the essay's illustration of Adam Smith's invisible hand.

    source|Culture
  • I've compiled the best SaaS Landing pages and broke down all their secrets

    Pedro Cortes compiles the best SaaS landing page examples he has seen across thousands of reviews, focusing specifically on conversion effectiveness rather than just aesthetics. The article breaks down actionable examples organized by page section: clear headers for strong first impressions, social proof placement, feature presentation, pricing page design, and call-to-action optimization. Each example includes analysis of why it works and how readers can apply the techniques to their own SaaS websites to convert more visitors into customers.

    source|Business
  • Identity Theft

    An article exploring identity theft: how it occurs, consequences for victims, and prevention strategies. The article covers types of identity theft, recovery processes, and protective measures. It raises awareness about this prevalent crime.

    source|Security
  • Implementing a Debugger: The Fundamentals

    A guide to building a software debugger covering breakpoints, stepping, and inspecting program state. The article explores debugger architecture and common implementation approaches. It provides foundation knowledge for debugger development.

    source|Dev Tools
  • In-Place Parallel Super Scalar Samplesort (IPS⁴o)

    IPS⁴o (In-place Parallel Super Scalar Samplesort) is a high-performance sorting algorithm combining blockwise in-place data distribution with branchless decision trees for sequential and parallel operations. Supports 16-byte atomic instructions, OpenMP and C++ threads, with dynamic adaptation based on input characteristics like duplicate elements. Substantially outperforms competing algorithms across multiple data types and architectures.

    source|Algorithms
  • Inside the weird get-rich-quick world of dropshipping

    An investigation into the dropshipping e-commerce model and its proliferation through Instagram advertisements. The article exposes the mechanics of how dropshipping businesses operate, including low margins, unreliable supply chains, and the marketing tactics used to attract participants. It reveals why dropshipping remains attractive to aspiring entrepreneurs despite its significant challenges and hidden costs.

    source|BusinessNetworking
  • International Standard Paper Sizes

    An explanation of international standard paper sizes like A4, A3, B-series, and their mathematical properties. The article covers origins, relationships between sizes, and practical applications. It provides reference information for paper standards.

    source
  • Interval Tree Clocks

    Fred Hebert explores Interval Tree Clocks (ITCs) from the 2008 Almeida, Baquero, and Fonte paper, proposing a practical protocol for using them in key-value stores. ITCs improve on version vectors and vector clocks by separating node identity from event counters, allowing dynamic addition and removal of participants without coordination. The article defines concrete ID propagation and data synchronization protocols with both version vector and vector clock equivalence modes. Hebert discusses practical considerations including backups, cluster membership changes, and how ITCs elegantly handle scenarios that are problematic for traditional causality tracking mechanisms.

    source|Distributed Systems
  • Introducing Cloudflare’s IPFS Gateway

    Cloudflare introduces its IPFS Gateway, enabling easy access to InterPlanetary File System content without installing special software. IPFS is a peer-to-peer file system where content is addressed by cryptographic hash rather than location, meaning files are fetched from whichever node has them rather than a single server. The gateway at cloudflare-ipfs.com serves as a bridge between the traditional web and IPFS, supporting custom hostnames with free SSL certificates. The post explains IPFS fundamentals including content addressing, the Merkle DAG structure, and the DHT-based peer discovery system, positioning it as part of Cloudflare's broader Distributed Web Gateway project.

    source|Distributed SystemsCloudNetworking
  • Introduction and Quick Guide to GraphQL for BackEnd and FrontEnd

    A visual guide explaining QUIC, a new transport layer protocol improving upon TCP/UDP. The article uses diagrams to illustrate the handshake process, encryption, and features. It makes the complex protocol accessible through visual explanation.

    source|Web DevNetworking
  • IPFS and Their Gateways

    Daniel Stenberg (curl author) examines the security and privacy implications of IPFS HTTP gateways, which proxy IPFS content over HTTPS. He explains how applications like ffmpeg added IPFS support by rewriting ipfs:// URLs to HTTPS requests through remote gateways, but warns that this approach lets gateway operators inspect and tamper with all traffic passing through them. Stenberg argues against using default remote gateways, noting that some gateways even redirect users to other gateways without consent. For curl, he insisted on requiring users to explicitly configure a gateway rather than defaulting to a remote one, and his concerns prompted ffmpeg to also remove their default gateway.

    source|Networking
  • Is this Mahler? This sounds like Mahler

    A software engineer describes building a system to display what's currently playing on WQXR, NYC's classical radio station, without interrupting her activities. She discovered WQXR's website makes periodic API calls to a whats_on endpoint returning JSON data about current programming. She first displayed this in her tmux status bar, then built a dedicated e-ink display using a Raspberry Pi Zero W and Pimoroni's inkywHAT display. The project involved formatting text to fit the display, creating a decorative frame using ImageMagick, and running updates via cron, resulting in a mantel-mounted display readable from anywhere in her living room.

    source|Culture
  • It's virtually impossible to read old iMessages and they take up tons of storage

    This article exposes a long-standing iMessage frustration: users pay for iCloud storage holding gigabytes of old messages they effectively cannot access, since the only method to reach early conversations is scrolling backward — which crashes the app or beach-balls the computer long before reaching messages from years ago. The author criticizes Apple for lacking basic features like date-based search, a timeline scrubber, or selective conversation deletion, while the system actively recommends users permanently delete all old conversations to free space. The piece argues this represents both a UX failure and a subtle revenue strategy, as inaccessible messages consume storage that pushes users toward paid iCloud tiers.

    source
  • Japan's Paper Culture

    JetPens explores the deep significance of paper in Japanese culture, from traditional washi-making techniques dating back to 610 C.E. to modern stationery innovation. The article profiles major Japanese paper companies including Kokuyo (Campus notebooks, PERPANEP line), Tomoegawa (ultra-thin Tomoe River paper), and Midori (MD Paper), highlighting how each designs paper with specific pen interactions in mind. It covers Japan's techo planner culture, the history of washi and its UNESCO Intangible Cultural Heritage status, and modern washi applications from automotive windscreens to festival floats.

    source
  • John Resig: Introducing the GraphQL Guide

    An introduction to GraphQL, a query language for APIs, by John Resig. The article explains GraphQL concepts, benefits over REST, and how to get started. It provides foundational knowledge for adopting GraphQL.

    source|Web Dev
  • JPMorgan CEO Jamie Dimon’s morning routine: Wake up 5am and 'read tons of stuff'

    CNBC profiles JPMorgan Chase CEO Jamie Dimon's daily morning routine based on his appearance on the Coffee with The Greats podcast. Dimon wakes around 5 a.m. and spends one and a half to two hours reading physical newspapers including the NY Daily News, NY Post, NY Times, Wall Street Journal, and Financial Times, plus economic publications. He prefers print because online reading tends to narrow what you consume. He then exercises for about 45 minutes around 7 a.m. and skips breakfast, just having coffee before heading to work.

    source|Culture
  • JVM Internals

    A technical deep dive into Java Virtual Machine internals including memory management, class loading, and bytecode execution. The article covers how JVM achieves portability and performance optimization. It's valuable for Java developers understanding how their code runs.

    source|Programming
  • Keeping Calm: When Distributed Consistency Is Easy

    Hellerstein and Alvaro present the CALM theorem (Consistency As Logical Monotonicity), which establishes a theoretical boundary for when distributed systems require coordination and when they don't. The key insight is that monotonic programs—those that only accumulate information and never retract it—can achieve consistency without coordination protocols like Paxos or Two-Phase Commit. The article explains how coordination is often a limiting factor in performance, scalability, and availability, and provides a computability result that helps programmers determine which parts of their distributed systems truly need coordination mechanisms and which can safely operate coordination-free.

    source|Distributed Systems
  • Kindergarten Quantum Mechanics

    These lecture notes by Bob Coecke survey joint work with Samson Abramsky on doing quantum mechanics using only pictures of lines, squares, triangles and diamonds. This diagrammatic picture calculus substantially extends Dirac's notation and has a purely algebraic counterpart in Strongly Compact Closed Categories. The work subsumes Coecke's Logic of Entanglement and was presented at several conferences in 2005, including talks at the Perimeter Institute, QTRF-III, Google, and the Kestrel Institute.

    source|Science
  • Kinematics of Reverse Angle Parking

    Volodymyr Agafonkin (creator of Leaflet) built an interactive Observable notebook to determine whether his car could fit into a challenging reverse angle parking spot. Using the Ackermann steering geometry model from 1817, he simulates the path a car traces during reverse parking given adjustable parameters like steering angle, starting position, and heading. The visualization shows the swept area of the car during the maneuver using actual specs from his Hyundai i30 and measured parking lot dimensions. He concludes the parking is possible but not easy, and the exercise helped him understand car movement patterns during reverse turns.

    source
  • Last minute tips for YC Interviewees

    Garry Tan of Initialized Capital shares advice for Y Combinator interview candidates based on his nearly 5 years as a YC partner conducting interviews. Key tips include using plain-spoken concrete language rather than startup jargon, starting with specific traction then zooming out to show the big vision, knowing your unit economics and costs thoroughly, and being able to articulate what you understand that competitors don't. He emphasizes that with only 10 minutes there's no margin for error, and that traction gets you in the door while the potential for a big future gets you the investment.

    source|Startups
  • Learn how to design large-scale systems

    A comprehensive open-source resource for mastering large-scale system design principles covering scalability, availability, consistency, caching, databases, load balancing, and asynchronous processing. Includes multiple learning formats: structured study guides, Anki flashcard decks for spaced repetition, and hands-on exercises with complete solutions for common interview questions. Provides both theoretical foundations and practical architectural patterns used by major tech companies.

    source
  • Learn how to write an emulator

    This article provides guidance for building computer emulators that simulate hardware behavior, covering CPU instruction sets and memory simulation. It enables emulation projects. The piece teaches emulation. It preserves technology knowledge.

    source|Virtual MachinesCareer
  • Learn TLA+

    Learn TLA+ is a comprehensive guide to the TLA+ formal specification language by Hillel Wayne. It teaches how to write formal specifications to verify complex systems like concurrent algorithms, distributed protocols, and data structures before implementation. The guide progresses from foundational concepts like modules and operators through advanced topics such as refinement proofs and liveness properties, making it valuable for engineers seeking to catch bugs early and understand system behavior at a deeper level.

    source|Distributed Systems
  • Learn to write your first OS kernel

    mkernel is an educational minimalist kernel project demonstrating fundamental OS development concepts, comprising x86 assembly and C code compiled with NASM, GCC, and a linker script. Adheres to the multiboot specification for compatibility with GRUB bootloaders on emulated and real hardware. The accompanying blog post provides educational context for kernel development fundamentals, valuable for beginners exploring low-level systems programming.

    source|Systems
  • Learning Golang – From Zero to Hero

    This tutorial provides a structured learning path for developers new to Go programming language, covering everything from basic syntax and data structures to advanced concepts like goroutines and interfaces. The course emphasizes practical understanding through hands-on examples and gradually increases complexity, enabling learners to progress from simple programs to building concurrent applications. It establishes Go's key strengths in simplicity, performance, and concurrency as essential skills for modern backend development.

    source|Programming
  • Learning hardware programming as a software engineer

    A software engineer shares lessons learned from their first foray into embedded programming after a friend asked for help building an RGB strip controller with Arduino Nanos. The article covers common pitfalls including misreading pinouts, accidentally using input-only pins for output, and destroying hardware. It explains different pin types (digital, analog, PWM, touch, DAC) and their proper usage on ESP32 boards, with practical advice on verifying board-specific pin definitions and understanding platform-specific limitations that pinout diagrams may not show.

    source
  • Learning Rust via Advent of Code

    This article describes using Advent of Code programming challenges as a practical method to learn Rust efficiently. By solving puzzles of increasing difficulty, learners develop real programming skills while absorbing Rust's ownership system, pattern matching, and functional programming paradigms. The approach provides concrete examples and immediate feedback, making it an effective alternative to traditional tutorials for understanding Rust's unique features and safety guarantees.

    source|Rust
  • Lessons learned from creating a real-time collaborative rich-text editor

    CKEditor describes the massive engineering effort behind building CKEditor 5 with real-time collaboration as a first-class architectural concern. The team abandoned CKEditor 4's 50+ person-year codebase to build a new architecture using Operational Transformation extended to support tree-based data structures rather than the typical linear model. They added five new operations (rename, split, merge, insert text, marker) beyond OT's standard three to properly handle rich text conflict resolution scenarios. The project took over 42 person-years with 25+ developers, 5,700 tickets closed, 12,500 tests at 100% code coverage.

    source|CRDTs
  • Let's Build a Compiler

    Let's Build a Compiler is an educational series that walks through implementing a compiler from scratch, explaining lexing, parsing, code generation, and optimization. The guide uses clear explanations and progressive examples to demystify compiler construction, making the topic accessible to developers without formal compiler theory background. It serves as both a learning resource and practical reference for understanding how programming languages are translated into machine instructions.

    source|Compilers
  • Let's Build a Compiler

    The first installment of a tutorial series on building an optimizing Scheme compiler from scratch, targeting 32-bit x86 assembly. The author sets up a basic framework using GNU Guile where a compile-program function outputs assembly that gets linked with a C runtime to produce executables. The article implements support for constant integers, booleans, and characters using tagged pointer representation, where the lowest bits of a 32-bit word encode type information—integers use the lowest 2 bits as 00, while booleans, characters, and heap pointers use distinct tag patterns. In 87 lines of code, the compiler can generate executables that represent and display primitive Scheme types.

    source|Compilers
  • Let's Build a Simple Interpreter. Part 14: Nested Scopes

    Let's Build a Simple Interpreter. Part 14: Nested Scopes continues a tutorial series on building an interpreter, focusing on implementing variable scoping within nested function definitions. The article explains symbol table management, scope resolution, and handling variable lookups across nested contexts. It advances learners' understanding of how interpreters manage program state and variable bindings.

    source|Compilers
  • Let’s implement a Bloom Filter

    Onat Mercan walks through implementing a Bloom filter in Rust as part of a series on building a distributed datastore. The article explains Bloom filters as space-efficient probabilistic data structures that answer set membership queries with possible false positives but no false negatives. It covers the math behind optimal bit array size and hash function count, then implements a StandardBloomFilter struct using the Kirsch-Mitzenmacher optimization of simulating k hash functions with just two. The implementation uses Rust's bit-vec crate and DefaultHasher, with insert and contains operations that calculate indices via g_i(x) = h1(x) + i*h2(x).

    source|Algorithms
  • Let's make a Teeny Tiny compiler

    This article provides tutorial for building a simple compiler, covering lexing, parsing, and code generation fundamentals. It demystifies compiler construction through practical implementation. The piece enables compiler learning through hands-on building. It teaches compiler basics.

    source|Compilers
  • Let's Talk SkipList

    Ketan Singh provides a comprehensive guide to skip lists, implementing one in Go with generics. The article explains skip lists as collections of sorted linked lists arranged in levels that enable logarithmic search by skipping nodes, serving as alternatives to balanced trees and hash maps. It covers find, insert, and delete operations with visual diagrams, then discusses performance optimizations including cache-friendly memory allocation, unrolled skip lists, and level balancing. Real-world examples include RocksDB's memtable implementation, Redis sorted sets (with Antirez's rationale for choosing skip lists), and the MuQSS Linux CPU scheduler.

    source|Algorithms
  • Let's write a compiler

    Brian Robert Callahan begins a tutorial series on writing a compiler from scratch in C. Part 1 covers selecting PL/0 (a Pascal-like language designed by Niklaus Wirth specifically for teaching compiler construction) as the source language and C as the implementation language, with plans to eventually create a self-hosting compiler. The post explains the three main compiler components—lexer, parser, and code generator—and how they will interleave in a single-pass compiler that outputs C code. The series aims to go from zero to a working compiler to a self-hosted compiler written in the language it compiles.

    source|Compilers
  • LibreTexts – Free The Textbook

    LibreTexts is an online platform providing free, peer-reviewed educational textbooks and materials across diverse subjects from chemistry to engineering. The platform emphasizes accessibility, affordability, and collaborative authorship, enabling students worldwide to access quality educational resources without cost barriers. It represents a significant effort to democratize education and challenge traditional expensive textbook markets.

    source|Learning
  • Linux Inside - How the Linux Kernel Works

    Linux Insides is an open-source book-in-progress by @0xAX that shares knowledge about the internals of the Linux kernel. The project covers low-level subject matter related to how the Linux kernel works and is available in multiple languages including Portuguese, Chinese, Japanese, Korean, Russian, Spanish, and Turkish. It is community-driven with contributions welcome via GitHub, and has an associated Google Group mailing list for discussion about learning the kernel source code.

    source|Systems
  • List of All Current TLDs

    This article catalogs top-level domains currently available for domain registration, providing reference for domain options. It documents internet infrastructure. The piece serves as TLD reference. It catalogs domain space.

    source|Networking
  • Living APIs and the Case for GraphQL

    Brandur makes the case for GraphQL as a next-generation API technology, arguing it goes beyond surface-level benefits like reduced payload sizes. Key advantages include explicit field contracts that enable living APIs which evolve gradually rather than through abrupt versioning, built-in introspection that powers tools like GraphiQL for API exploration, native batch operations without custom specifications, and typed schemas with nullability baked in. He notes GraphQL's biggest challenge may be that while it's better than REST, it's not dramatically enough better to overcome REST's momentum and widespread adoption. A 2025 update adds that HTTP/3 plus SDK-driven design may be a better compromise.

    source|Web Dev
  • Local area network push notifications

    Local Area Network Push Notifications describes implementing a push notification system within a local network, enabling devices to send updates and alerts without relying on internet infrastructure. The article covers protocol selection, network broadcast techniques, and client-server communication patterns for LAN-based notification delivery. It provides practical solutions for developers building offline-capable systems and local applications.

    source
  • Local First Tuple Database

    Tuple Database is a local-first, embedded TypeScript database with schemaless architecture where developers manipulate typed indexes as tuples instead of using SQL queries. Supports both synchronous in-memory and asynchronous persistent storage (SQLite, LevelDb) with reactive updates across all queries automatically. Enables transactional operations with arbitrary application logic and supports many-to-many queries without JOIN overhead.

    source|CRDTsDatabases
  • Local Sheriff – Browser extension to show PII leaks to third-parties

    Local Sheriff is a browser web-extension that identifies when personally identifiable information leaks to third-party trackers during browsing by monitoring network traffic for sensitive data in URLs, headers, or query parameters. Uses the open-source WhoTracks.me database to map tracker hostnames to companies with findings displayed in a control panel. Available on Chrome, Firefox, and Opera with all data stored locally.

    source|Security
  • Local-first software

    Ink & Switch's seminal essay introduces the concept of local-first software, which combines the ownership and control of traditional desktop applications with the collaboration benefits of cloud apps. The authors argue that today's cloud apps leave users at the mercy of service providers who can shut down, raise prices, or lose data. They propose seven ideals for local-first software: fast performance, multi-device support, offline capability, collaboration, longevity, privacy, and user ownership of data. The essay evaluates existing technologies like CRDTs (Conflict-Free Replicated Data Types) as a promising foundation for building local-first applications that store data primarily on the user's device while still enabling real-time collaboration.

    source|CRDTs
  • Lode Runner

    Lode Runner is a historical analysis of the classic 1983 arcade game by The Digital Antiquarian, exploring its innovative level design, gameplay mechanics, and cultural impact. The article discusses how the game pioneered concepts like player-created content through level editors and influenced subsequent platformer design. It captures the technical and creative achievements of early computer gaming.

    source|Culture
  • Machine Learning Crash Course

    Google's Machine Learning Crash Course is a free, self-paced introduction to machine learning featuring animated videos, interactive visualizations, and hands-on exercises. The course covers fundamental topics including linear regression, logistic regression, classification, working with numerical and categorical data, neural networks, embeddings, and large language models. It also addresses real-world considerations like production ML systems, AutoML, and ML fairness, with modules on identifying and mitigating biases in data and models.

    source|ML & AILearning
  • macOS internals

    Failed to fetch - network blocked

    source|Systems
  • Make Your Python CLI Tools Pop with Rich

    Rich is a Python library by Will McGugan for creating visually appealing CLI output using ANSI escape codes. It supports text formatting with colors and styles, progress bars, markdown rendering, syntax highlighting, tables, and tree displays. A standout feature is the inspect function, which uses reflection to display detailed information about any Python object for debugging. Rich also provides beautifully formatted tracebacks with local variable context. The library serves as the foundation for Textual, a TUI framework that enables developers to build GUI-like interfaces directly in the terminal.

    source|Programming
  • Making CRDTs Byzantine Fault Tolerant

    SKIP:already-in-all-summaries

    source|CRDTs
  • Mapping the unknown - Steps to map any industry

    Mapping the Unknown: Steps to Map Any Industry by Steve Blank provides a systematic methodology for analyzing and understanding any unfamiliar industry. The ten-step framework guides entrepreneurs through customer discovery, value chain analysis, and market structure identification. It emphasizes that successful industry entry requires thorough understanding rather than assumptions, enabling strategic decision-making.

    source
  • Master web development with over 9000 tricks

    David Gilbertson introduces Know It All (know-it-all.io), a web tool that presents a hierarchical tree of roughly 10,000 web development topics spanning CSS, HTML, JavaScript, SVG, and the DOM. Developers can browse the tree and rate their knowledge level on each topic to discover gaps in their expertise. Gilbertson built the tool after reading through 80 web specifications while recovering from illness, entering all data using Microsoft Word's outline mode and converting it via Excel to JSON. The application works offline as a progressive web app and aims to help developers systematically identify what they don't yet know.

    source|Web Dev
  • Mastering the Basics of Icon Design

    French graphic designer Adrien Coquet shares his process for creating simple, consistent icons in Adobe Illustrator in this Noun Project interview. Coquet relies heavily on the Pen tool, Shape Builder, and Pathfinder to construct icons from basic shapes and strokes, then refines them iteratively by duplicating and comparing versions. He emphasizes consistency in stroke weights, fill styles, miter limits, and corners across an icon collection to achieve a cohesive visual language. Coquet transitioned from pharmacy studies to freelance design after discovering his passion for creativity, and his typeface Panama was even featured on a national television show.

    source|Design
  • Matrix-CRDT – real-time collaborative apps using Matrix as backend

    Matrix-CRDT enables using Matrix as a backend for distributed, real-time collaborative web applications that sync automatically by functioning as a sync provider for Yjs. Treats CRDT updates like chat messages, sending base64-encoded document changes as custom Matrix events while leveraging Matrix's authentication, access control, encryption, and federation. Includes periodic snapshot functionality and experimental WebRTC for peer-to-peer synchronization.

    source|CRDTs
  • Maximizing Developer Effectiveness

    Tim Cochran's article on martinfowler.com introduces a framework for maximizing developer effectiveness by identifying and optimizing key developer feedback loops. He contrasts highly effective environments (where developers make incremental progress with fast feedback) against low-effectiveness ones (plagued by blockers, bureaucracy, and long wait times). The article highlights micro-feedback loops that developers execute hundreds of times daily, presents case studies from Spotify (which built the Backstage developer portal) and Etsy (which tracks lead time and developer NPS), and argues that organizations must invest in developer experience as a product through platform thinking and continuous improvement.

    source|Career
  • Memory Techniques from Vedic Learning

    Memory Techniques from Vedic Learning presents ancient Sanskrit learning techniques for memorizing and retaining large amounts of information through structured repetition and visualization. The document explains how Vedic education systems developed sophisticated memory methods used for centuries to preserve extensive textual knowledge. It provides modern practitioners with powerful mnemonic frameworks applicable to contemporary learning challenges.

    source
  • Mergeable replicated data types – Part I

    This Morning Paper post reviews the OOPSLA'19 paper on Mergeable Replicated Data Types (MRDTs), which are similar in spirit to CRDTs but with the key advantage of composability. MRDTs use a three-way merge approach that takes two concurrently modified versions and their lowest common ancestor to resolve conflicts. The core technique maps data type values to a relational domain using invertible specifications, performs merges in that domain using standard set operations, then maps back to the concrete data type. The paper demonstrates this with counter and queue examples, showing how characteristic relations can be defined for various data types to automatically derive correct merge functions.

    source|CRDTs
  • Microsoft Coffee

    In 1996, a group of Microsoft employees pulled an elaborate April Fools prank by designing, printing, and shrink-wrapping hundreds of realistic boxes for a fake product called 'Microsoft Coffee,' poking fun at the company's reputation as an imitator of Sun's Java. They planted the boxes in retail stores across the Seattle area, complete with fake barcodes and easter egg jokes, and sent fake press releases that got local TV and radio coverage. Microsoft's PR department panicked and tried to suppress the story, collecting remaining boxes from stores and denying corporate involvement, which had a lasting chilling effect on employee pranks at the company.

    source|Culture
  • Microsoft Space Simulator

    The Digital Antiquarian chronicles the story of Microsoft Space Simulator (1994), created by Charles Guy, an astronomy-obsessed programmer who previously worked on flight sims at subLogic. The software was a unique hybrid: half realistic spaceflight simulator supporting historical craft like Apollo and the Space Shuttle, and half virtual galaxy explorer encompassing the entire Milky Way in Super VGA graphics. Despite Microsoft's enthusiastic backing with lavish manuals and strategy guides, the product failed commercially because its vast, goal-less exploration proved too intimidating for most users compared to the more accessible Flight Simulator. Guy never helmed another game and died of cancer in 2004 at a young age.

    source|Culture
  • Mini HTTP Guide for Developers

    A concise guide to the HTTP protocol essentials for web developers. The article covers request/response structure, status codes, headers, and common patterns. It serves as a quick reference for HTTP concepts.

    source|Networking
  • MiniCouchDB in Rust

    MiniCouchDB in Rust describes building a minimal CouchDB implementation in Rust during a hack week project, demonstrating key database concepts like document storage, HTTP APIs, and eventual consistency. The article documents architectural decisions and implementation challenges when building a database from scratch. It serves as an educational example of database design for developers interested in systems programming.

    source|RustDatabases
  • Modern Javascript: Everything you missed over the last 10 years

    Sandro Turriate's concise cheatsheet covers all the major JavaScript features added over the past decade (up to ECMAScript 2020) for developers who avoided modern syntax requiring transpilers. It covers array functions (map, filter, reduce, etc.), const/let block scoping, nullish coalescing and optional chaining operators, async/await, arrow functions, classes with getters/setters, destructuring, the spread operator, template literals, Proxies, and ES modules. Each feature includes runnable code examples and links to further documentation.

    source|Programming
  • Modern Software Over-Engineering Mistakes

    This article identifies 10 common software over-engineering mistakes, including trying to outsmart business requirements, creating overly reusable business logic, abusing generics, writing shallow wrappers around good libraries, and blindly applying quality concepts like SOLID. The author argues that business requirements only diverge over time, making premature abstraction wasteful, and advocates for vertical splitting of functionality over horizontal layering. The piece also warns against vague quality attributes like configurability that go unchallenged, in-house library inventions that become legacy, following the status quo without refactoring, and bad estimation that destroys code quality.

    source|Dev Tools
  • Monica - open source personal CRM

    Monica is an open-source personal CRM designed to help people organize and remember details about their personal relationships rather than business contacts. It lets users store information about family members, work details, pets, activities, reminders for important dates, gift ideas, debts, and tasks associated with each contact. The application includes a journal feature for tracking daily reflections and is built with privacy in mind — users can self-host it on their own server for free, and the code is publicly available for community scrutiny and contribution.

    source
  • Moreutils: A collection of Unix tools that nobody thought to write long ago

    Moreutils is a collection of Unix tools that provides missing utilities for common command-line tasks, filling gaps in standard Unix toolkits.

    source|ProgrammingDev Tools
  • Mozart’s Infinite Riches

    An article examining Mozart's vast compositional legacy and the depths of his musical catalog.

    source|Culture
  • My Business Card Runs Linux

    An embedded systems engineer built a business card that runs Linux using the obscure Allwinner F1C100s processor, which integrates both CPU and RAM in a single package. The card costs under $3 to produce and boots a custom Buildroot Linux in about 6 seconds when plugged into a USB port, appearing as both a flash drive (containing a resume and photography) and a virtual serial port with a shell. The shell includes classic Unix programs like rogue, fortune, a 2048 game, and MicroPython, all packed into an 8MB flash chip with proper wear leveling via UBI.

    source|Business
  • My Life in Pens (2009)

    A personal essay exploring the author's relationship with different pen tools and writing instruments.

    source|Culture
  • My Show HN Project got acquired after 6 months

    Josh Howarth details how he built Trennd.co, a trend-detection tool, from a personal CLI tool into a web app that reached #1 on Hacker News Show HN and grew to over 2,000 email subscribers in six months. After validating the concept through Reddit, Hacker News, and organic Twitter buzz from figures like Rand Fishkin and Pieter Levels, he struggled to find the right monetization model. The project was acquired by Brian Dean of Backlinko, rebranded as ExplodingTopics.com, and refocused on serving professional bloggers who need to discover trending topics early. The story illustrates how building in public, solving your own problem, and organic word-of-mouth can lead to acquisition even without revenue.

    source|ProgrammingStartups
  • My Top Free Font Superfamilies

    A guide highlighting the best free font families available for designers, covering versatile typefaces for various projects.

    source|Design
  • Myth: Eric Brewer On Why Banks Are BASE Not ACID

    This article examines Eric Brewer's explanation of why banks use BASE (Basically Available, Soft State, Eventually Consistent) rather than ACID transactions. The key insight is that availability correlates with revenue while consistency generally does not—if an ATM is down, the bank loses money. Banks have historically never had perfect consistency because communication was always partitioned, dating back to Renaissance banking conducted via letters on horseback. ATMs use commutative operations (increment/decrement) that can be reordered and reconciled later, with overdraft penalties managing the rare risk of inconsistency. The real-world philosophy is to bound and manage risk while keeping all operations available.

    source|Finance
  • Never accept a counter-offer

    Vardan Torosyan argues that employees should never accept a counter-offer from their current employer when resigning. His core reasoning is that if a problem triggered the desire to leave, and the employer only resolves it during a resignation meeting after previously failing to address it, there is no guarantee future problems will be handled differently. He notes that even when the trigger is just a vague feeling rather than a concrete issue, a counter-offer won't resolve it, and surveys suggest most employees who accept counter-offers leave within 15 months anyway.

    source|Career
  • New particle discovered - the Ridiculon

    A humorous or satirical post about the discovery of a fictional particle called the Ridiculon.

    source|Science
  • Nim Apocrypha

    An in-depth exploration of Nim programming language concepts and lesser-known features.

    source|Programming
  • Nyxt 2.2.0 - the hacker's power browser

    Nyxt 2.2.0 is a release of the Nyxt web browser, a keyboard-driven, extensible browser written in Common Lisp. This release introduces panel buffers for displaying side information, a macro editor for interactively recording command macros, QR code generation for URLs, a view-source command, and an element hinting overhaul with a new DOM parser replacing much of the former Parenscript code with pure Lisp. The prompt buffer received significant performance improvements, and the status area now allows clicking on modes to describe them and toggling modes with a '+' button.

    source
  • OKRs vs. KPIs: understanding measurements for your SaaS business

    An article comparing OKRs (Objectives and Key Results) and KPIs (Key Performance Indicators), explaining how they differ in measuring business success.

    source|Business
  • Olaf – Acoustic Fingerprinting on the ESP32 and in the Browser

    Olaf is a system for acoustic fingerprinting that works on both ESP32 microcontrollers and in web browsers for music identification.

    source|Hardware
  • Old CSS New CSS

    Eevee's comprehensive history of CSS traces web design from the late 1990s through the modern era, using the original Space Jam website as a case study for early techniques. The article covers the progression from table-based layouts and spacer GIFs through the CSS zen garden movement, the IE6 dark ages, the Web 2.0 era of glossy gradients and rounded corners, and into modern CSS with flexbox, grid, custom properties, and media queries. It serves as both a nostalgic retrospective for veterans and an educational overview for newer developers who may not know how much CSS has evolved.

    source|Web Dev
  • On Being a Senior Engineer

    John Allspaw's influential essay defines what it means to be a mature senior engineer, arguing that technical skill alone is insufficient. Key characteristics include seeking constructive criticism of designs, understanding non-technical perception, making trade-offs explicit, being comfortable with estimates under uncertainty, lifting the skills of those around them, practicing empathy across stakeholders, and being aware of cognitive biases like hindsight bias and self-serving bias. The post emphasizes that the degree to which others want to work with you directly indicates career success, and includes the Ten Commandments of Egoless Programming and the Dreyfus model of skill acquisition.

    source|Career
  • One Billion Apples’ Secret Sauce: Recipe for Apple Wireless Direct Link Protocol

    This paper reverse-engineers Apple Wireless Direct Link (AWDL), a proprietary undocumented IEEE 802.11-based ad hoc protocol that Apple introduced around 2014 and integrated into its entire product line. AWDL powers popular applications like AirPlay and AirDrop on over one billion devices. Through binary and runtime analysis, the authors reveal how AWDL nodes announce Availability Windows for communication, synchronized by an elected master node. They study the master election process, synchronization accuracy, channel hopping dynamics, and achievable throughput, and publish an open source Wireshark dissector for the protocol.

    source
  • One Order of Operations for Starting a Startup

    A guide outlining a recommended sequence of steps and priorities for entrepreneurs launching a new startup.

    source|Startups
  • Open source applications for macOS

    A curated GitHub repository cataloging 689 free, open-source macOS applications across 49 categories, with Swift, Objective-C, and JavaScript as dominant languages. Includes structured JSON-based data files, automated GitHub Actions workflows, and comprehensive badges showing release status and activity level. Provides developers and users an accessible way to discover maintained community-driven alternatives to proprietary software.

    source
  • Open source collaborative text editors

    Jure Triglav surveys the landscape of open-source real-time collaborative web-based rich text editors, comparing CKEditor 5 and Atlassian's Atlaskit Editor using a kiwi-based scoring system across criteria like licensing, feature support, collaboration capabilities, and production readiness. Both editors scored 8.5/11 initially, but the author tipped the scales by building the missing open-source collaboration server for Atlaskit (backed by PostgreSQL and PubSweet) and replacing Atlassian's restrictively licensed icons with the open Feather set, bringing it to 10.5/11. The article also discusses the underlying technologies of operational transforms and CRDTs that power collaborative editing.

    source|CRDTs
  • OpenAI's GPT-3 may be the biggest thing since Bitcoin

    Manuel Araoz demonstrates GPT-3's capabilities through a clever meta-experiment: the entire blog post claiming to share early GPT-3 experiments on bitcointalk.org was itself written entirely by GPT-3. The AI-generated article convincingly describes fake experiments, makes predictions about GPT-3's impact, and mimics Araoz's writing style, only revealing at the end that it was all machine-generated from a brief prompt containing the author's bio and a title. The post serves as a compelling demonstration of GPT-3's ability to produce coherent, persuasive long-form content that could fool readers.

    source|ML & AIFinance
  • OpenDrop: An open Apple AirDrop implementation written in Python

    OpenDrop is a command-line tool enabling cross-platform file sharing over Wi-Fi with protocol compatibility to Apple's AirDrop system, written in Python. Allows sending files and URLs to iOS and macOS devices with support for everyone discovery mode and contacts-only devices. Requires Apple Wireless Direct Link (AWDL) support and includes limitations including lack of Bluetooth LE triggering and multi-file sending capabilities.

    source|Networking
  • Oppenheimer and the Gita

    Nuclear historian Alex Wellerstein analyzes Oppenheimer's famous quote from the Bhagavad Gita at the Trinity test, drawing heavily on James Hijiya's scholarly article. Wellerstein explains that in the Gita, the god Krishna reveals his terrifying cosmic form to the warrior prince Arjuna to convince him to fulfill his duty in battle against his own relatives. Crucially, Oppenheimer identified not as the god Krishna but as the human Arjuna — a reluctant participant compelled by cosmic forces to wage war despite personal misgivings, reconciling himself to his duty as a 'prince of physics' upon witnessing the bomb's terrible power.

    source|Culture
  • Optimizing Lambda Cost with Multi-Threading

    A presentation from AWS Re:Invent 2020 discussing techniques for reducing AWS Lambda costs through multi-threading and efficient resource utilization. The talk covers how to maximize the CPU and memory concurrency available in a Lambda invocation. It provides practical strategies for optimizing serverless computing economics.

    source|Cloud
  • Original Bessemer Investment Memos for Shopify

    Bessemer Venture Partners publishes their original investment memos from early-stage decisions in companies that became major successes. The page serves as a learning tool to identify patterns in successful venture investments, featuring memos from investments in companies like Shopify, Twilio, Pinterest, LinkedIn, and many others. BVP shares these alongside their famous 'Anti-Portfolio' of companies they passed on, offering rare transparency into the VC decision-making process.

    source|VC & Fundraising
  • OS development series

    A series covering operating system development, providing tutorials and resources for learning to build OSes from scratch.

    source|Systems
  • OT and CRDT trade-offs for Real-Time collaboration

    This TinyMCE blog post compares Operational Transformation (OT) and Conflict-free Replicated Data Types (CRDT) for building real-time collaborative editing. While CRDT offers appealing peer-to-peer capabilities and offline support, the author argues it struggles with rich text editing because it operates at too low a level and loses user intent, leading to poor experiences like cursor jumping during formatting changes. TinyMCE chose server-based OT for its collaboration feature because OT, despite being more complex, captures editing intent better and has proven reliability for rich text editors, while CRDT remains better suited for simpler data structures like plain text and JSON.

    source|CRDTs
  • Oura CLI

    A command-line interface tool for the Oura health and wellness tracking platform.

    source
  • Overview of Consistency Levels in Database Systems

    An article exploring the different consistency models in distributed database systems and their tradeoffs.

    source|Distributed SystemsDatabases
  • Paginating Requests in APIs

    Ignacio Chiazzo compares three common API pagination strategies: page-based (offset), keyset-based, and cursor-based pagination. Page-based is simplest and allows jumping to any page but suffers from poor SQL OFFSET performance on large datasets and can return duplicates when records are inserted during pagination. Keyset-based uses a delimiter key like since_id for more efficient SQL WHERE queries but is tied to sort order. Cursor-based pagination uses opaque tokens and is the most performant approach, favored by Stripe, Facebook, Slack, and Shopify, though it cannot skip pages or support parallel batch requests.

    source|Networking
  • Palindrome Day 20200202

    This mathematical analysis explores palindromic dates across different date formats (ISO YYYYMMDD, American MMDDYYYY, and German DDMMYYYY). The date 2020-02-02 was special because it reads as a palindrome in all three major formats simultaneously, the first such occurrence since 2011-11-02 in ISO format. The author catalogs all 331 palindromic dates in YYYYMMDD format and 335 in DDMMYYYY format between the years 1000 and 9999, noting the next cluster won't occur until 2030 and providing complete reference tables.

    source|Culture
  • Papers every developer should read at least twice

    Michael Feathers curates 10 foundational computer science papers every developer should read, spanning modular system design, distributed computing, and programming language theory. The selections include Parnas on decomposing systems into modules, Waldo et al. on why distributed computing cannot be made transparent, Landin on expression-oriented programming, Backus arguing for functional programming, and Thompson on the fundamental trust problem in compiled code. Each paper gets a brief synopsis explaining its continued relevance, emphasizing that the software industry reinvents its notations while important foundational ideas risk being lost.

    source
  • Papers We Love

    Papers We Love is a community-driven repository of academic computer science papers and a global network of local chapters that organize meetups where members present and discuss influential research papers. With chapters in dozens of cities worldwide including New York, San Francisco, London, Berlin, and Tokyo, the community hosts regular talks on topics ranging from data structures and distributed systems to machine learning and formal verification. Recent presentations have covered papers on temporal knowledge graphs, transformer architectures, bin packing algorithms, and write-optimized B-epsilon trees.

    source|Learning
  • Papers with Code

    Papers With Code was a free and open resource that linked machine learning papers with their code implementations, datasets, and evaluation benchmarks. The site provided a structured way to find state-of-the-art results across thousands of ML tasks, making it easier for researchers to reproduce and build upon published work. It was acquired by Hugging Face and now redirects to Hugging Face's trending papers page.

    source|Learning
  • Parsing JSON at the CLI: A Practical Introduction to jq and more

    This practical tutorial by Sequoia McDowell covers jq, a command-line tool for parsing and modifying JSON data. It walks through basics like pretty-printing, filtering by key and value, and piping filters together, then demonstrates real-world debugging of Prometheus metrics involving 316K lines of JSON. The article also introduces yq for YAML parsing and pup for HTML parsing, showing them applied to Kubernetes configurations and web scraping tasks.

    source|CompilersDev Tools
  • Pathfinding Demystified

    An article demystifying pathfinding algorithms like A* and Dijkstra, explaining core concepts and implementations.

    source|Algorithms
  • Pattern: Backends For Frontends

    Sam Newman describes the Backend For Frontend (BFF) architectural pattern, where instead of a single general-purpose API backend, each user interface type (mobile, desktop, third-party) gets its own dedicated server-side component. Each BFF is tightly coupled to its specific UI, maintained by the same frontend team, and handles aggregation of downstream microservice calls. The pattern was popularized at SoundCloud and REA Group, with Newman recommending one BFF per user experience to avoid the bloat that comes from trying to serve all clients with a single API gateway.

    source
  • Paxos vs. Raft: Have we reached consensus on distributed consensus?

    Heidi Howard and Richard Mortier compare the two dominant distributed consensus algorithms, Paxos and Raft, by describing a simplified Paxos using Raft's terminology. They find both algorithms take very similar approaches, differing primarily in leader election: Raft only allows servers with up-to-date logs to become leaders, while Paxos allows any server to lead if it updates its log afterward. Raft's approach proves surprisingly efficient since it avoids exchanging log entries during leader election. The authors conclude that much of Raft's perceived understandability comes from its paper's clear presentation rather than fundamental algorithmic differences.

    source|Distributed Systems
  • Paxos vs. Raft: Have we reached consensus on distributed consensus?

    Aleksey Charapko's reading group discusses the paper comparing Multi-Paxos and Raft consensus protocols, noting that while they are very similar in happy-case operations, key differences emerge in leader election: Raft requires the most up-to-date replica as leader while Paxos can pick any node. The discussion covers why Raft became dominant in industry (reference implementations, engineering-friendly description, existing libraries) while Google notably uses Multi-Paxos in Spanner. The group also debates whether Raft is essentially an implementation of Multi-Paxos, the teachability of both protocols, and how formal methods like TLA+ may bring them closer together.

    source|Distributed Systems
  • PC Assembly Book

    PC Assembly Language is a free online textbook by Paul Carter, written for an introductory assembly language course at the University of Central Oklahoma. The book covers 32-bit protected mode x86 assembly programming using the free NASM assembler and GNU gcc compiler, with extensive coverage of interfacing assembly and C code. It includes chapters on basic assembly, bit operations, subprograms, arrays, floating point, and C++ structures, and has been translated into French, German, Spanish, Italian, Korean, and Chinese.

    source
  • Playwright CLI – Open/inspect/emulate pages

    Playwright CLI is a command-line interface for browser automation providing token-efficient commands as skills for AI agents working with limited context windows. Enables comprehensive browser control including navigation, element interaction, storage management, network mocking, and debugging with snapshots to identify elements. Optimizes AI agent workflows by avoiding verbose accessibility trees in prompts while enabling high-throughput automation.

    source|Dev Tools
  • Pocket-sized cloud with a Raspberry Pi

    This article describes building cloud services and infrastructure using Raspberry Pi clusters, demonstrating accessible edge computing. It enables DIY infrastructure. The piece demonstrates hardware capability. It enables learning projects.

    source|HardwareCloud
  • Porting Firecracker to a Raspberry Pi 4

    The Cloudkernels team documents their work porting AWS Firecracker, the micro-VMM behind Lambda and Fargate, to run on the Raspberry Pi 4. The key missing piece was support for GICv2, the ARM interrupt controller version used by the RPi4, since Firecracker only supported GICv3. They introduced a GIC Trait abstraction in Rust to handle both versions transparently, implemented the GICv2-specific register mappings for the distributor and CPU groups, and updated the Flattened Device Tree creation for GICv2 compatibility. The article includes detailed code snippets and instructions for building and running Firecracker on the RPi4 with a tap network interface.

    source|Hardware
  • Porting USB applications to the web. Part 1: libusb

    This web.dev article by Ingvar Stepanyan explains how to port libusb, a popular C library for USB device communication, to the web using WebAssembly (Emscripten), Asyncify, and the WebUSB API. The author implemented a new Emscripten backend for libusb that maps its synchronous C operations to asynchronous WebUSB calls, handling challenges like event loop integration and device enumeration permissions. The work culminated in a demo that remotely controls a DSLR camera from a web browser, achieving 30+ FPS live preview after optimizing the async event handling mechanism.

    source
  • Positions - Profit and Loss

    This comprehensive introductory guide explains how trading positions work, covering brokers, exchanges, settlement, and the difference between long and short positions. It details how to calculate key position metrics including quantity, cost basis, average entry price, realised PnL, mark value, and unrealised PnL. The article also compares inventory valuation methods (FIFO, LIFO, and Average Cost) for calculating realised profit and loss, demonstrating that while they produce different intermediate results, they converge to the same final PnL when a position is fully closed.

    source
  • PR 101 for engineers

    Aneel's guide demystifies public relations for engineers and startup founders who come from non-marketing backgrounds. The article explains how to work effectively with journalists by understanding their incentives, preparing key messages and sound bites, practicing media interactions, and building genuine relationships with relevant press contacts. It also covers when to use PR agencies, how to be a useful source for reporters even when not pitching, and is part of a larger series on go-to-market topics for technical founders.

    source|Career
  • Practical Transformer Winding

    A practical guide to the techniques and methods for hand-winding electrical transformers.

    source|Hardware
  • Principles for Naming a Brand

    This guide from Mojomox (formerly MMarch) covers principles for creating effective brand names. It emphasizes that names exist within a broader context of taglines, logos, and color palettes, and should reflect a company's strategy, product, positioning, and target market. Key advice includes checking domain availability early, keeping names short and pronounceable, avoiding descriptive category keywords that competitors will copy, and testing names with target audiences for sound, meaning in other languages, and negative associations. The article also covers naming trends like vowel-dropping and suffix fads, and discusses spelling conventions for logos versus body text.

    source|Business
  • Project Gotham Racing
    ProgrammingCulture
  • Prompt injection explained with video, slides, and a transcript

    An explanation of prompt injection attacks on language models with videos, slides, and transcripts.

    source|Security
  • Protect domains that do not send email

    This UK Government Digital Service guidance explains how to protect domains that do not send email from being used in spoofing and phishing attacks. It provides step-by-step DNS configuration instructions for creating SPF records that deny all senders, DMARC records set to reject policy, empty DKIM key records, and null MX records. The guide also covers how to handle subdomains that do send email within a non-sending parent domain.

    source|Security
  • Prototyping a YC Startup Each Day

    Anvil, a platform for building web apps in Python, challenged itself to prototype a Y Combinator startup each day during Demo Day week. The first prototype was MeterFeeder (YC Winter 16), a smartphone parking payment system, built in just 90 minutes. The prototype included a public-facing app for customers to pay for parking via Stripe integration and check remaining time, plus an enforcement app for parking wardens to verify valid bookings by location or license plate. Anvil's visual interface designer, Python-based client and server code, and one-click cloud deployment made the rapid prototyping possible.

    source|Startups
  • Public speaking for introverts

    Dan Shipper shares personal techniques for overcoming public speaking anxiety as an introvert, drawn from hundreds of presentations including Fortune 100 pitches and live TV appearances. His key advice includes practicing with minimal slide notes rather than memorizing scripts word-for-word, adopting a sharing mindset focused on genuine communication rather than perfect phrasing, and keeping slides focused with large fonts. He emphasizes that the only real cure for public speaking insecurity is repeated experience, and encourages starting early rather than avoiding it.

    source|Career
  • QuickJS JavaScript Engine

    QuickJS is a small, embeddable JavaScript engine created by Fabrice Bellard that supports the ES2023 specification including modules, async generators, proxies, and BigInt. It features a fast interpreter with very low startup time, passes nearly 100% of the ECMAScript Test Suite, and can compile JavaScript sources to standalone executables with no external dependencies. The engine uses reference-counting garbage collection with cycle removal and includes sub-projects like libregexp, libunicode, and a float64 dtoa library, all released under the MIT license.

    source|ProgrammingNetworking
  • Raft Visualization

    Raft Visualization is an interactive tool that visualizes the Raft consensus algorithm, showing how distributed nodes reach agreement through leader election and log replication. The visualization helps developers understand Raft mechanics and properties like safety and liveness. It serves as an educational resource for understanding distributed consensus concepts.

    source|Distributed Systems
  • Raising VC funding for a solo-dev OSS project

    The ToolJet team shares their experience raising $1.55 million in VC seed funding within two weeks of launching their open-source low-code framework on Hacker News. The article discusses why they chose VC funding over bootstrapping or maintaining it as a side project, explaining that they needed to grow quickly and build dozens of integrations. It covers their timeline from first commit to closing the round, the fundraising process including cold outreach and inbound interest after launch, and the benefits VC funding provides to open-source projects including team building and customer confidence.

    source|VC & Fundraising
  • React from zero: a simple tutorial for React

    React From Zero is an educational tutorial teaching React fundamentals through 18 progressive lessons executable directly in browser without build tools or pre-compilation. Emphasizes vanilla JavaScript and minimal modern syntax to help beginners understand core concepts, covering topics from basic elements and JSX through components, lifecycle methods, and integration patterns. Available in multiple languages with 4,600+ GitHub stars.

    source|ProgrammingWeb Dev
  • Real Differences Between OT and CRDT for Co-Editors

    This paper provides a comprehensive comparison between Operational Transformation (OT) and Conflict-free Replicated Data Types (CRDT) for collaborative editing. While CRDTs have been positioned as superior post-OT techniques since 2006, the authors find that OT solutions remain the dominant choice in production co-editors. The paper refutes several CRDT-claimed advantages over OT regarding correctness, complexity, and simplicity, presenting evidence from a thorough review of representative solutions and working co-editors to clear up common myths and misconceptions in the co-editing community.

    source|CRDTs
  • Repository of paper airplane designs

    Fold 'N Fly is a database of 56 paper airplane designs with step-by-step folding instructions, video tutorials, and printable folding plans. Each design is categorized by type (distance, time aloft, acrobatic, decorative), difficulty level (easy to expert), and whether scissors are required. The site has been teaching people worldwide how to make paper airplanes for over 10 years and includes expert tips, tricks, world records information, and a paper airplane launcher guide.

    source|Culture
  • Resources for Amateur Compiler Writers

    This curated resource page by Quentin Carbonneaux collects practical papers, books, and references for amateur compiler writers. It covers key topics including compiler descriptions (Ken Thompson's Plan 9 compiler, PCC, lcc), SSA form construction and dominators, optimizations like constant propagation and global value numbering, register allocation algorithms (linear scan and graph coloring), code generation, and machine-specific references for x64 and ARM64 ABIs. Each entry includes the author's personal remarks on practical usefulness.

    source|Compilers
  • Reverse Engineering for Beginners

    Reverse Engineering for Beginners is a comprehensive textbook teaching reverse engineering techniques for understanding compiled programs, covering assembly language, debugging, protection mechanisms, and analysis methods. The book progresses from fundamentals through advanced techniques applicable to security research and malware analysis. It serves as both educational resource and reference for developers entering the reverse engineering field.

    source|Security
  • Reverse-Engineering WebAssembly

    Reverse-Engineering WebAssembly is a technical document exploring methods for analyzing and understanding WebAssembly binaries, addressing the challenge of analyzing compiled web applications. The paper discusses tooling, analysis techniques, and the security implications of WASM's reduced human-readability. It serves as a reference for security researchers and developers working with WASM.

    source|Virtual MachinesSecurity
  • RFC 3339 vs. ISO 8601

    RFC 3339 vs. ISO 8601 explains the differences between these two timestamp standards, clarifying which to use for different purposes in APIs and data interchange. The interactive resource demonstrates real-world implications of format choices for string parsing, timezone handling, and international compatibility. It guides developers in making appropriate decisions for timestamping systems.

    source|Networking
  • Rhapsody Rabbit and Cat Concerto

    Rhapsody Rabbit and Cat Concerto are classic animated cartoons that present musical stories and performances through animation, representing important works in animation history. These pieces combine visual artistry with musical accompaniment, showcasing how animation can express narrative and emotion. They preserve cultural artifacts from the golden age of animation.

    source|Culture
  • Rising from the Ashes: A Family's Journey of Resilience and Hope

    Rising from the Ashes: A Family's Journey of Resilience and Hope It was a sunny day in early spring when Dravus and Hopseed received the news that would change their lives forever. The couple had just finished breakfast when they received a call from their bank informing them that their savings account had been emptied due to a fraudulent scheme. In addition, the home that they had been renti...

  • robocopy

    An article about robocopy, a powerful Windows file synchronization and backup tool. The piece covers practical usage patterns and options for robust file copying in enterprise environments.

    source
  • robocopy

    Robocopy is a Windows command-line tool for robust file copying that offers advanced features like multi-threaded copying, retry logic for network failures, and bandwidth throttling. The tool provides granular control over file attributes, timestamps, and permissions, making it essential for system administrators managing large-scale file transfers and backup operations. Robocopy supports sophisticated filtering options for selective copying and detailed logging capabilities to monitor transfer operations. Its reliability in handling network interruptions and ability to resume interrupted transfers have made it the de facto standard for Windows file synchronization tasks.

    source
  • rohand.com

    The ballad of Rohit Chapter 1: Introduction In the interplay between destiny and determination, sometimes an individual's journey creates symphonies. The life of Rohit Verma is a testament to such melodies. From the corporate towers of Wall Street to the fiery stoves of New York kitchens, his story is one of passion, persistence, and culinary brilliance.

  • Run Your Home on a Raspberry Pi

    A guide to running Python-based home automation and personal projects on Raspberry Pi.

    source|Hardware
  • Running macOS in a Virtual Machine on Apple Silicon Macs

    This Apple Developer sample code project demonstrates how to install and run macOS virtual machines on Apple silicon using the Virtualization framework. It includes two apps: InstallationTool (a CLI utility that downloads macOS restore images and creates VM bundles) and macOSVirtualMachineSampleApp (a GUI app that launches and controls the VM). The sample covers configuring VM properties like CPU count, memory, graphics, and storage, as well as saving and restoring VM state between app launches, with both Swift and Objective-C implementations provided.

    source|Virtual Machines
  • Rust on the ESP32

    This blog post documents the author's work on getting Rust to compile and run on the ESP32 microcontroller using the Xtensa architecture. It describes hacking the Rust compiler to use Espressif's LLVM Xtensa fork, creating built-in Rust targets for ESP32 and ESP8266, and building a bare-metal LED blinky program using no_std Rust with direct memory-mapped register writes. The post covers GPIO configuration, cycle-accurate delays using the CPU counter register, and discusses current limitations including debug info generation issues and the ongoing LLVM upstream effort.

    source|RustHardware
  • rvdc: Rohan's personal CLI

    rvdc: Rohan's personal CLI rvdc is Rohan's personal CLI and contains a smorgasbord of functionality that I've built over time.

  • SaaSy Math: A Resource of SaaS Metrics for Your Startup

    This comprehensive resource compiles essential SaaS metrics and their calculations, providing founders and operators with mathematical frameworks for measuring business health. The guide covers key performance indicators like MRR, ARR, churn rate, CAC payback period, and LTV, explaining both calculation methods and interpretation strategies. It contextualizes metrics within different growth stages, helping teams understand which metrics matter most at each phase of their development. The resource serves as a quick reference guide for investors reviewing startups and for founders tracking their own business performance.

    source|StartupsBusiness
  • Sam Altman's Startup Playbook

    Sam Altman's Startup Playbook is a comprehensive guide distilled from Y Combinator's advice to startups, covering the essential elements needed to build a successful company. It is organized into sections on finding a great idea with a clear market, building a great team starting with co-founders, developing a great product through tight feedback loops with users, and executing well through growth, focus, hiring, fundraising, and competition. The playbook emphasizes that the best startups solve a problem the founders themselves experience, and that relentless execution matters as much as the initial idea.

    source|Startups
  • Scaling Browser Automation with Puppeteer on AWS Lambda with Container Image Support

    This AWS Architecture Blog post demonstrates how to run Puppeteer headless Chrome browser automation inside AWS Lambda functions using container image support. The solution uses two Lambda functions: a Puppeteer function that takes a URL and screenshots it using headless Chrome, and a fan-out function that asynchronously invokes the Puppeteer function for each URL in a list. The article provides a complete Dockerfile based on the AWS Lambda Node.js base image with Chrome installed, and uses AWS CDK for infrastructure deployment. The approach enables horizontal scaling of browser automation tasks like web scraping, functional testing, and load testing with up to 100K simultaneous connections.

    source|Cloud
  • Schopenhauer’s 38 Stratagems

    This page presents Arthur Schopenhauer's 38 rhetorical stratagems from The Art of Controversy, first translated into English in 1896. The strategies cover techniques like exaggerating an opponent's position to create more objections, exploiting ambiguous word meanings, hiding your conclusion until all premises are admitted, making opponents angry to impair their judgment, and turning their own arguments against them. The collection serves as both a guide to persuasion and a catalog of logical fallacies and debate tricks that remain widely used today.

    source
  • SDSL – Succinct Data Structure Library 2.0

    SDSL-Lite is a C++11 library implementing succinct data structures storing data close to information-theoretic lower bounds while maintaining efficient operations, based on 40 research papers. Encompasses bitvectors with rank/select, wavelet trees, compressed suffix arrays, and suffix trees with ease-of-use modeled after STL. Includes serialization capabilities and extensive documentation with examples and tutorials under GPLv3.

    source|Algorithms
  • Separating gifted children hasn't led to better achievement

    An opinion piece arguing that separating gifted students from their peers has not improved educational outcomes.

    source|Career
  • Setting up a Pi Hole made my home network faster

    This article describes setting up Pi-hole on a Raspberry Pi as a network-wide DNS-based ad blocker running in Docker. By routing all DNS queries through the Pi-hole, ads are blocked at the network level for every device without needing browser extensions. After one month of use, the author found that 15-30% of all DNS queries were ad-related and blocked, noticeably improving browsing speed. The article includes step-by-step setup instructions for Docker deployment and configuration tips for optimal performance.

    source|Hardware
  • Shoelace 2.0 release: UI toolkit that works with all frameworks or none at all

    Shoelace is an open-source, framework-agnostic library of professionally designed UI web components that work with any frontend framework including React, Vue, and Angular. Built on web standards using Lit, it offers fully customizable components via CSS, includes dark theme support, built-in localization, and accessibility features. The project positions itself as a foundation for design systems that saves organizations from building common UI components from scratch, with the advantage that web components won't break when frameworks are upgraded or switched.

    source|Web Dev
  • Signed Char Lotte

    This article examines the concept of signed character representation in programming, discussing how different systems represent character data types and the implications for string handling and data processing. It explains nuances of character encoding and signed/unsigned distinctions relevant to low-level programming. The piece helps programmers understand character representation subtleties that affect compatibility and correctness. It addresses often-overlooked character encoding details.

    source
  • Simply explained: how does GPT work?

    An accessible explanation of how GPT (Generative Pre-trained Transformer) language models work. The article covers transformer architecture, training on text data, and why these models can generate coherent text. It makes deep learning concepts understandable to non-specialists.

    source|ML & AI
  • SkiftOS: Simple delightful operating system

    SkiftOS is an open-source operating system prioritizing modularity, simplicity, and modern design built primarily in C++ (90% of codebase), featuring integrated components like Karm core library, KarmUI reactive UI, Hideo desktop environment, and Vaev browser engine. Currently in early development stages, representing an artistic endeavor rather than commercial product and not ready for daily use. Licensed under LGPL-3.0 and combines C++, Python, and Assembly components.

    source|Systems
  • Smashing the Stack for Fun and Profit

    This classic security paper demonstrates buffer overflow vulnerabilities and stack-based attacks, explaining how overwriting return addresses enables code execution. It provides foundational understanding of memory safety issues and exploitation techniques. The piece is essential reading for understanding why memory safety matters and how exploits work. It documents fundamental security vulnerability class.

    source|Security
  • So Good They Can't Ignore You

    This Commoncog book summary covers Cal Newport's 2012 career strategy book 'So Good They Can't Ignore You,' which argues against the common advice to 'follow your passion.' Newport presents four key ideas: the passion hypothesis is flawed because passion usually develops after building rare and valuable skills; the craftsman mindset of deliberate practice and skill acquisition leads to better career outcomes than the passion mindset; autonomy is earned through career capital but requires avoiding control traps; and meaningful work emerges from mission, which itself requires expertise at the cutting edge of a field.

    source|ProgrammingCareer
  • So I took a corporation to arbitration

    Shu Chow recounts his experience taking a home warranty company to binding arbitration after they refused to reimburse him for an emergency plumber he hired when their computer systems went down and they couldn't dispatch a contractor or process a pre-authorization. He filed with the American Arbitration Association for $200, deliberately withholding evidence in his initial filing, and within two days received a 50% settlement offer. By carefully citing specific contract clauses the company violated, he counter-offered at 80% and they accepted the next day, demonstrating that reading your contract carefully and using arbitration can be effective against large corporations.

    source|Longform
  • So you want to build an embedded Linux system?

    Jay Carlson's comprehensive guide targets embedded engineers looking to move from microcontrollers to application processors running Linux. The article reviews numerous low-cost system-on-chip options from vendors like Allwinner, TI, NXP, Microchip, and others, comparing their features, software support, toolchains, and community ecosystems. It covers the practical challenges of designing with these processors including PCB layout, boot sequences, kernel configuration, and driver support, aiming to demystify what has traditionally been considered an expert-only domain.

    source|Hardware
  • So you want to study mathematics

    Susan Rigetti provides a comprehensive self-study curriculum for learning undergraduate mathematics, structured as a roadmap from calculus through PDEs and electives. The guide recommends specific textbooks for each topic—including Stewart's Calculus, Velleman's How to Prove It, Strang's Linear Algebra, Dummit and Foote's Abstract Algebra, and Abbott's Understanding Analysis—along with supplementary resources and free online courses. Rigetti emphasizes that anyone can learn mathematics regardless of perceived aptitude, and that the key is patience, working through problems, and studying in small manageable pieces. The curriculum covers nine core areas: calculus, introduction to proofs, linear algebra, algebra, real analysis, complex analysis, ODEs, PDEs, and electives.

    source|Science
  • Software Engineering: The Soft Parts

    This book addresses the human and organizational aspects of software engineering often overlooked in technical training, covering collaboration, communication, leadership, and team dynamics. It emphasizes that software engineering success depends as much on soft skills as technical ability. The work helps engineers develop interpersonal and organizational effectiveness. It elevates soft skills in engineering discourse.

    source|Career
  • Sol – a sunny little virtual machine

    Rasmus Andersson describes Sol, a register-based virtual machine he built as a learning project over a weekend. Sol features cooperative multitasking with multiple tasks managed by a scheduler, 32-bit encoded instructions in three layouts (ABC, ABx, Bxx), activation records with function prototypes and program counters, and a yield mechanism for task switching. The article provides detailed explanations of the instruction encoding, the scheduler's run queue, timer-based yielding, and an operation cost counter to prevent tasks from monopolizing CPU time, with inspiration drawn from Lua 5 and Erlang.

    source|Virtual Machines
  • Soviet Venus Images

    This article documents images captured by Soviet Venera probes during their Venus missions, some of the only images of Venus surface transmitted to Earth. It discusses the technical achievements of these early space missions despite challenging Venusian conditions. The piece celebrates Soviet space exploration contributions and the challenging engineering required for Venus missions. It documents significant space exploration history.

    source|Culture
  • Speeding up VSCode extensions in 2022

    This article provides performance optimization techniques for Visual Studio Code extensions, discussing common bottlenecks, profiling approaches, and optimization strategies. It covers extension architecture patterns that impact performance and best practices for efficient extension development. The piece helps extension developers improve user experience. It documents VS Code extension performance optimization.

    source|Web Dev
  • Stack Based Virtual Machines

    Andrea Bergia's introductory article explains stack-based virtual machines as abstractions of computers that interpret bytecode. The article walks through how basic operations work using a stack and instruction pointer, showing step-by-step how pushing values and executing an add instruction manipulates the stack. It also contrasts stack-based VMs (like the JVM and .NET CLR) with register-based VMs (like Lua), noting that while register-based VMs are generally faster, stack-based designs can still achieve high performance, as demonstrated by the JVM.

    source|Virtual Machines
  • Start your Subaru with a dead key fob
  • Startup Financial Modeling: What is a Financial Model?

    Troy Henikoff explains what a financial model is for startups in this first installment of a series co-written with Will Little. He argues that founders should build their financial models from scratch rather than using templates, because the process forces them to deeply understand their business levers. A financial model is defined as a simplified mathematical representation of a company's financial operations that helps founders plan, fundraise, and make strategic decisions. The article emphasizes that the real value lies in the modeling process itself, not just the final spreadsheet.

    source|Startups
  • Startup idea checklist

    Slava Akhmechet presents a structured checklist for evaluating startup ideas, organized into three sections: Product, Growth, and Meaning. Each section contains character-limited prompts that force concise thinking about target users, their dissatisfaction, market size, competitive moats, growth strategies, and personal commitment. The checklist draws on wisdom from investors and entrepreneurs like Don Valentine, Peter Drucker, and Jeff Bezos. It serves as a practical framework for founders to stress-test their ideas before committing significant time and resources.

    source|Startups
  • Startups in 13 Sentences

    Paul Graham distills startup wisdom into 13 essential sentences covering topics like picking good cofounders, launching fast, evolving ideas based on user feedback, and spending as little as possible. He emphasizes understanding users as the most critical principle, arguing that if he could keep only one sentence it would be that. The essay also covers practical advice on customer service as a competitive advantage, measuring what you make, achieving ramen profitability, avoiding distractions, and not giving up. Graham notes that deals falling through is normal and founders should maintain morale as a key management priority.

    source|Startups
  • Staying on the path to high performing teams

    Will Larson presents a framework for reasoning about team performance across four states: falling behind, treading water, repaying debt, and innovating. Each state requires a different system fix: hiring for falling-behind teams, adding process to consolidate efforts for treading-water teams, adding time for debt-repaying teams, and maintaining slack for innovating teams. He emphasizes focusing resources on one team at a time rather than spreading them thin, and argues that these slow systemic fixes create durable improvements that compound over time, contrasting them with quick tactical interventions that don't stick.

    source
  • Step by step guide for building a startup: from idea to exit

    A comprehensive step-by-step guide for entrepreneurs from initial idea through startup exit, covering validation, funding, team building, and growth. The guide breaks the startup journey into actionable stages with specific milestones. It serves as a practical roadmap for aspiring founders.

    source|Startups
  • Stock trading strategies using machine learning

    This repository provides comprehensive resources for applying machine learning techniques to asset management and trading, encompassing 15 trading strategy varieties with around 100 strategies and portfolio weight optimization using supervised, unsupervised, and reinforcement learning. Structured around published research papers in top journals, offering code and datasets for strategies from simple momentum to advanced deep learning approaches. Ranked in Top 1% SSRN paper downloads across multiple categories.

    source|ML & AI
  • Story of the Flash Fill Feature in Excel

    Sumit Gulwani tells the behind-the-scenes story of how Excel's Flash Fill feature was born, from his POPL 2011 paper that won the Most Influential Paper award. The inspiration came from a woman on a flight who asked him how to reformat names in Excel—a question he couldn't answer at the time. After researching Excel help forums, he discovered millions of users struggling with simple string transformations and developed a program synthesis approach using domain-specific languages and deductive search. The Excel team's strict requirements for sub-second performance and single-example accuracy pushed key technical innovations including program ranking techniques that combined logical reasoning with machine learning. Flash Fill now sees about a million invocations per week.

    source|ML & AI
  • Structure and Interpretation of Computer Programs (SICP) Book

    This is the companion website for Structure and Interpretation of Computer Programs (SICP) by Harold Abelson and Gerald Jay Sussman, the influential MIT computer science textbook known as the 'Wizard Book.' The site provides the complete text in HTML (licensed under Creative Commons), sample programming assignments from the MIT course, all source code from the book, an instructor's manual with exercise discussions and teaching suggestions, and information on obtaining Scheme implementations. SICP uses the Scheme dialect of Lisp and is widely used at universities for both introductory and advanced courses.

    source|Learning
  • Summary of all the MIT Introduction to Algorithms lectures

    This post summarizes the author's complete lecture notes from MIT's Introduction to Algorithms course (23 lectures across 14 blog posts). It covers fundamental algorithm topics including analysis of algorithms, divide and conquer, sorting (quicksort, radix sort), order statistics, hashing, search trees, augmenting data structures, skip lists, amortized analysis, dynamic programming, greedy algorithms, shortest path algorithms (Dijkstra, Bellman-Ford, Floyd-Warshall), parallel algorithms, and cache-oblivious algorithms. For each lecture, the author lists key topics and highlights the most interesting insights, such as how quicksort and BST sort produce identical decision trees, and how Dijkstra's algorithm reduces to BFS on unweighted graphs.

    source|Algorithms
  • Surviving Adversity: The Inspiring Story of Dravus and Hopseed

    Surviving Adversity: The Inspiring Story of Dravus and Hopseed Dravus and Hopseed were a hardworking couple who had three children: two daughters and a son. They lived in a small town where Dravus worked as a nurse at the local hospital, while Hopseed worked part-time as an accountant. One day, Dravus received the devastating news that the hospital was facing a decrease in patient admissions ...

  • Svelte and TypeScript

    This article covers integrating TypeScript into Svelte projects, discussing tooling, type definitions, and benefits of static typing for Svelte components. It helps developers add type safety to Svelte applications. The piece improves Svelte developer experience. It enables type-safe Svelte development.

    source|ProgrammingWeb Dev
  • Sync a GitHub fork with the upstream manually

    Sync a GitHub fork with the upstream manually This is specific to aws-sdk-react-native, since that is the primary fork I sync. # add upstream if it doesn't exist # https://help. github.

  • Take my app ideas

    Austin Henley, a professor at Carnegie Mellon, shares several app ideas from his list of 80+ concepts that he hasn't had time to build himself. The ideas include a service that auto-generates and updates a website from a Google Doc or Sheet, a platform for short-form solo audio podcasts (something between Twitter and traditional podcasts), a clipboard tracker that monitors and logs URLs, and a journal/todo app that asks what you plan to do and follows up on whether you did it. Henley argues that ideas are less valuable than execution and encourages others to take and build his concepts rather than let them go to waste.

    source|Startups
  • Tapster’s robots are built to poke touchscreens

    Tapster is a lean two-person startup in Oak Park, Illinois that builds small, affordable robots designed to automate touchscreen testing. Founded by former Google tester Jason Huggins, the company originated from a laser-cutting class project that evolved into a button-clicking robot capable of playing Angry Birds. Their big break came when Mercedes-Benz needed 10 robots for testing a self-parking app's touchscreen interface—at a fraction of the $100,000 per unit quote from industrial robotics companies. The company builds its robots using desktop 3D printers and has been intentionally bootstrapped with product sales and $100,000 from Indie.vc.

    source
  • Tauri - Electron Alternative Powered by Rust

    Tauri is a framework for building desktop applications with web technologies but using Rust for the backend, offering lighter weight and better performance than Electron. It enables web developers to build desktop apps with smaller binaries and lower resource consumption. The framework addresses key Electron limitations while maintaining web development paradigms. It provides viable Electron alternative.

    source|RustWeb Dev
  • Teensy 3.1 bare metal: Writing a USB driver

    This detailed tutorial walks through writing a bare-metal USB device driver for the Teensy 3.1's Freescale K20 ARM processor without relying on external libraries. The author covers four major parts: configuring the clock system (MCG and SIM) to provide the required 48MHz USB reference clock, the USB module initialization sequence including Buffer Descriptor Table setup with 512-byte memory alignment, implementing an interrupt-driven state machine for handling USB resets and token processing, and handling SETUP tokens to respond with device and configuration descriptors. The article emphasizes practical challenges like ensuring USB buffers are in RAM (not flash), proper DATA0/DATA1 toggling, and ping-pong buffer synchronization.

    source|Hardware
  • Ten Commandments of Salary Negotiation

    A guide covering ten commandments of salary negotiation to help professionals advocate for fair compensation.

    source|Career
  • Testing Distributed Systems

    This is a comprehensive, curated list of resources on testing distributed systems maintained by Andrey Satarin. It covers a wide range of testing approaches including Jepsen analyses, formal methods (TLA+), deterministic simulation, chaos engineering, fuzzing, lineage-driven fault injection, and property-based testing. The page catalogs specific testing strategies used by major companies including Google, Amazon, Netflix, Meta, FoundationDB, CockroachDB, MongoDB, and many others. It also includes sections on research papers about bugs in distributed systems, tools for network simulation and benchmarking, and resources for testing concurrent single-node systems.

    source|Distributed SystemsDev Tools
  • The “menu engineers” who optimize restaurant revenue

    Menu engineering is a niche consulting profession that blends design, psychology, and food knowledge to optimize restaurant menus for maximum profitability. Engineers like Michele Benesch and Gregg Rapp use techniques such as strategic item placement based on eye-tracking patterns, removing dollar signs from prices, using descriptive language for high-margin dishes, and deploying decoy items to influence customer spending. The COVID-19 pandemic forced a major shift as restaurants moved to delivery and QR code menus, requiring engineers to rethink fundamental assumptions about how diners interact with menus on smartphone screens rather than physical pages.

    source|Business
  • The Adam and Eve Story

    A retelling or analysis of the Adam and Eve story from the Bible, potentially exploring its origins, cultural significance, or theological interpretations. The article provides context for one of the most influential narratives in Western civilization.

    source|Longform
  • The Art of Logging

    A comprehensive guide to effective logging practices in software systems, covering what to log, log levels, structured logging, and performance considerations. The article explores how good logging design accelerates debugging and enables production monitoring. It emphasizes logging as a critical system design concern rather than an afterthought.

    source|Dev Tools
  • The Beirut Bank Job

    This Darknet Diaries podcast episode features infosec professional Jayson E. Street telling the story of when he was hired to physically break into a bank in Beirut, Lebanon as a penetration test, and everything went wrong. The episode explores physical security testing of banks, where the goal is not to steal cash but to gain access to teller computers. Street, who has nearly two decades of infosec experience doing both defensive and offensive security work, shares the incredible details of the engagement. The episode includes references to his related DEFCON talks on social engineering and physical penetration testing.

    source|FinanceLongform
  • The bible of doing business with the city of New York

    An article about New York City Record, a newspaper that serves as the official publication for government contracts and business with the city.

    source
  • The boring technology behind a one-person Internet company

    This article examines how a solo founder runs an Internet business using straightforward, unglamorous technology choices prioritizing reliability and maintainability. It argues that boring technology enables sustainable businesses by reducing operational burden. The piece advocates for pragmatism over trendiness. It demonstrates sustainable one-person business approach.

    source|Dev Tools
  • The brain has a 'low-power mode' that blunts our senses

    Neuroscientists at the University of Edinburgh discovered that when mice are food-deprived for weeks (losing 15-20% of body weight), neurons in their visual cortex reduce ATP usage at synapses by 29% through a low-power mode. The neurons save energy by decreasing sodium ion currents while maintaining firing rates through compensatory mechanisms like increased input resistance and raised resting membrane potentials. However, this comes at the cost of reduced precision in visual processing—food-deprived mice couldn't distinguish fine angular differences below 10 degrees. The effect is toggled by the hormone leptin: administering leptin restored normal cortical function without the mice eating any food, suggesting leptin signals energy availability to the brain.

    source
  • The Business of SaaS

    A comprehensive guide to the business model, economics, and management of Software-as-a-Service (SaaS) companies. The article covers metrics like MRR, churn, and lifetime value, along with strategies for growth and profitability. It's essential reading for anyone starting or managing a SaaS company.

    source|Business
  • The Chaos Engineering Book

    This book provides comprehensive introduction to chaos engineering—intentionally introducing failures into systems to test resilience—covering principles, tools, and case studies. It helps practitioners implement chaos engineering practices. The piece elevates chaos engineering as critical reliability practice. It documents chaos engineering methodology.

    source|Distributed Systems
  • The China Ship

    This multimedia feature by the South China Morning Post tells the story of the Spanish galleon trade routes known as the 'China Ships' that connected Asia and the Americas for over 250 years starting in the 16th century. It traces how globalisation began when Spanish navigators in the Philippines established the tornaviaje (return route) across the Pacific, enabling the Spanish silver dollar to become a transcontinental currency. The article covers the Magellan-Elcano circumnavigation that first discovered the route and the subsequent centuries of uninterrupted trade between Asia and the rest of the world.

    source|Longform
  • The Cognitive Style of PowerPoint

    Edward Tufte's critique of PowerPoint's slide-based presentation format, arguing it encourages oversimplification and reduces information transfer compared to alternative presentation formats. The analysis questions fundamental assumptions about presentations. The piece challenges presentation orthodoxy. It critiques information design.

    source|Design
  • The Complete GraphQL Security Guide

    This comprehensive guide covers security considerations when building GraphQL APIs, including authentication, authorization, injection attacks, and rate limiting. It helps developers build secure GraphQL systems. The piece addresses often-overlooked GraphQL security issues. It provides security guidance for GraphQL.

    source|Web DevSecurity
  • The data brokers quietly buying and selling your personal information

    Fast Company reports on a new Vermont law requiring companies that buy and sell third-party personal data to register with the Secretary of State, revealing 121 data brokers operating in the US. These range from people-search sites like Spokeo and Intelius, to credit bureaus like Equifax and Experian, to advertising giants like Acxiom and Oracle. The law notably excludes first-party data collectors like Amazon, Facebook, and Google, and doesn't require brokers to disclose what data they collect or give consumers access to their own data, though it does require some information about opt-out systems.

    source|Security
  • The Distributed Computing Manifesto

    Werner Vogels publishes Amazon's 1998 Distributed Computing Manifesto, a foundational internal document that laid the groundwork for transforming Amazon's monolithic two-tier architecture into a service-oriented and workflow-based distributed system. The manifesto proposes separating presentation, business logic, and data into a three-tier architecture with well-defined service interfaces, and introduces workflow-based processing where data moves between processing nodes rather than processes accessing a central database. It covers key concepts including synchronous vs asynchronous service methods, data domaining, state tracking, and handling changes to in-flight workflow elements through publish-subscribe models.

    source|Distributed Systems
  • The First Delta Force Trainee Class

    This article documents the history and experiences of the first official Delta Force training class, covering selection, training, and early operations. It explores military special operations history. The piece celebrates military achievement. It documents special forces history.

    source|Longform
  • The Founder’s Guide To Selling Your Company (2014)

    This guide provides practical advice for founders planning to sell their companies, covering valuation, negotiation, due diligence, and closing procedures. It helps founders maximize value and navigate M&A processes effectively. The piece demystifies company sales and acquisition procedures. It provides M&A guidance for sellers.

    source|StartupsBusiness
  • The grandmaster diet: How to lose weight while barely moving

    This ESPN feature explores the surprising physical toll of competitive chess, where grandmasters can burn up to 6,000 calories a day and lose 10-12 pounds over a 10-day tournament due to sustained mental stress, elevated blood pressure, and disrupted sleep. The article profiles how top players like Magnus Carlsen and Fabiano Caruana have adopted rigorous fitness regimens including running, tennis, and swimming to prepare their bodies for competition. Carlsen has optimized every detail from his diet (switching from orange juice to chocolate milk) to his sitting posture to maintain peak performance in the fifth hour of play.

    source|Culture
  • The Hot Wheels Design Studio: How a Real Car Gets Turned into a 1:64 Toy

    The Drive visited the Hot Wheels Design Studio in southern California to document the 9-month process of turning a real car into a 1:64 die-cast toy. The article follows the journey of Riley Stair's custom 1970 Pontiac Trans Am, the 2020 Legends Tour winner, through the full production pipeline. The process includes design sketching, digital clay sculpting on a medical surgery trainer rig, 3D prototyping, metal casting, track testing, and packaging. Hot Wheels designers balance capturing each car's unique character with the constraints of tiny scale manufacturing and the need for every model to perform well on orange track.

    source|Culture
  • The Hunt for the Death Valley Germans

    This article tells the story of the Death Valley Germans, a group of German settlers who became stranded in Death Valley during early twentieth century mining rushes. It explores historical desert survival and mystery. The piece documents historical exploration story. It examines American frontier history.

    source|Longform
  • The Illustrated QUIC Connection

    An interactive educational page that explains every byte of a QUIC connection, the secure UDP-based stream protocol that forms the basis of HTTP/3. The demonstration walks through a complete connection lifecycle: client hello, server hello and handshake with TLS 1.3 encryption, key exchange generation, sending 'ping', receiving 'pong', and connection termination. Each UDP datagram and its constituent packets are broken down byte-by-byte with detailed explanations of the cryptographic handshake, key derivation, and protocol mechanics.

    source|Networking
  • The Important Habit of Just Starting

    This article explores why we procrastinate on starting new projects and how to overcome the fear of beginning. It explains the psychology behind procrastination, including 'temporal myopia' (inability to envision your future self), time inconsistency (valuing immediate rewards over future ones), and how our 'motivationally toxic' environment encourages consumption over creation. The piece offers practical strategies for building a 'just start' habit: sweating the small stuff to avoid cumulative losses, using commitment devices (like Hugo locking away his clothes to force himself to write), creating distraction-free environments, and letting go of perfectionism. The core message is that the guilt of not starting is usually worse than the effort of actually doing the work.

    source|Career
  • The legacy of Pieter Hintjens

    This article honors Pieter Hintjens, discussing his contributions to ZeroMQ, open-source software, and unconventional thinking about technology and human systems. It celebrates his influence. The piece documents software community contributor. It honors technology advocate.

    source|Culture
  • The Little Fighter Who Overcame Cancer: A Story of Hope, Courage, and Community Support that won over Adversity

    The Little Fighter Who Overcame Cancer: A Story of Hope, Courage, and Community Support that won over Adversity Emma was just six years old when she was diagnosed with leukemia. Her parents were devastated, but they were determined to fight alongside their daughter. They took her to the best hospitals and doctors, but nothing seemed to work.

  • The longest train journey in the world in 2021

    Jon Worth meticulously investigates and corrects a viral claim about the longest train journey in the world, prompted by the opening of the Laos railway connecting Southeast Asia. He defines the journey as greatest geodesic distance, calculates precise coordinates, and plans a route from Portugal to Singapore using only trains (plus necessary walking segments). The route goes from Cascais/Praia das Maçãs through Spain, France, Germany, Sweden, Finland, Russia, China, Laos, Thailand, Malaysia, to Singapore—requiring about 30 trains, 4-7 gauge changes, and 24km of walking at three gaps where tracks exist but no passenger services run. The geodesic distance is approximately 11,905-11,929 km depending on start/end points.

    source
  • The Lost Art of Lacing Cable

    This article examines proper cable management and lacing techniques increasingly uncommon as technology infrastructure evolves. It preserves knowledge of data center practices. The piece documents infrastructure engineering practices. It celebrates technical craftsmanship.

    source
  • The Lottery Hackers

    The Huffington Post's long-form story follows Gerald Selbee, a retired Kellogg's cereal box engineer from Michigan, who discovered a mathematical loophole in the Massachusetts lottery game Cash WinFall. Selbee realized that when the jackpot rolled down to lower-tier prizes, the expected value of a ticket exceeded its cost, creating a guaranteed profit opportunity with enough volume. He and his wife Marge formed a corporation with friends and family, eventually buying hundreds of thousands of tickets and winning millions over several years, all completely legally through pure mathematical advantage rather than luck.

    source|Longform
  • The MBA Myth and the Cult of the CEO

    Dan Rasmussen and Haonan Li examine the legacy of Harvard Business School professor Michael Jensen's influential 1990 argument that CEO compensation should be tied to stock price performance. Jensen's theory rapidly transformed corporate pay structures, creating the modern cult of the highly compensated CEO justified by the need to attract 'the best and brightest.' The authors question whether three decades of evidence support Jensen's thesis, investigating whether stock-linked CEO pay actually improved corporate performance or simply enriched executives while the correlation between CEO quality and company outcomes remained weak.

    source
  • The Missing Semester of Your CS Education

    The Missing Semester of Your CS Education is an MIT class that teaches the practical tools and skills that CS curricula typically overlook, such as mastering the command line, using powerful text editors, version control with Git, debugging and profiling, and packaging and shipping code. The course argues that students spend hundreds of hours using these tools but rarely receive formal training in them. The 2026 edition integrates AI-enabled tools and workflows directly into each lecture rather than treating AI as a standalone topic, and covers newer subjects like agentic coding alongside traditional topics.

    source|Dev Tools
  • The Prosperous Software Consultant

    Nader Dabit shares seven strategies for building a successful software consulting career, drawn from years of experience earning $200K-$400K annually. The key principles are: building bridges (giving back through mentorship and content), specializing in a niche rather than being a generalist, creating content to establish authority, using flat-rate pricing based on value delivered rather than hourly billing, networking strategically, maintaining a professional image, and continuous learning. He emphasizes that specialists can charge more because they need less onboarding and work more efficiently, and that flat-rate pricing based on the business value you create breaks through the ceiling imposed by hourly billing.

    source
  • The Really Big One

    Kathryn Schulz's Pulitzer Prize-winning article for The New Yorker describes the catastrophic earthquake and tsunami that will someday strike the Pacific Northwest when the Cascadia subduction zone ruptures. The 700-mile fault line running from California to Canada has produced 41 major earthquakes in the past 10,000 years, with the last one—a magnitude 9.0—occurring on January 26, 1700. Scientists estimate one-in-three odds of a major quake in the next 50 years. FEMA projects nearly 13,000 deaths, with everything west of Interstate 5 devastated, and the region facing years without basic infrastructure. The article details how the fault was only discovered in the late 1980s and how the Pacific Northwest remains profoundly unprepared.

    source|Science
  • The secret Thinkpad powerbutton code to bring dead laptops back to life

    This article documents a hidden hardware recovery mechanism on some ThinkPad laptops allowing recovery from unresponsive states through specific power button sequences. It provides practical repair knowledge. The piece helps users revive dead hardware. It preserves technical knowledge.

    source|Dev Tools
  • The Siege of Gondor

    Historian Bret Devereaux begins a six-part series analyzing the Siege of Gondor from Peter Jackson's Return of the King through the lens of historical pre-gunpowder siege warfare. Part I focuses on logistics—the professional's concern—examining how Mordor's army would have been supplied and organized for such a massive military campaign. The series compares the film's depiction against historical siege practices, noting significant divergences from Tolkien's books and evaluating military plausibility. The analysis draws on real-world examples from ancient and medieval warfare to assess everything from army composition to supply chains.

    source|ProgrammingCulture
  • The slowest SR-71 Blackbird fly-by

    This article recounts pilot Brian Shul's story from his book Sled Driver about an unplanned dramatic SR-71 Blackbird flyby at a small RAF base in England. While returning from a mission, Shul and backseater Walt Watson were asked to perform a low approach for air cadets, but had difficulty locating the small airfield in haze. The aircraft slowed dangerously below 160 knots while circling, and when Shul lit both afterburners to recover, the plane dropped into full view of the tower observers in what became a legendary knife-edge pass. The cadets' hats were blown off and the commander called it the greatest SR-71 flyby he'd ever seen, though the pilots knew they had narrowly avoided disaster.

    source|Longform
  • The STRIDE Threat Model

    STRIDE is a threat modeling framework identifying potential security threats through categories: Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, Elevation of Privilege. The model helps security practitioners systematically identify threats and design appropriate mitigations. It's widely used for security architecture review. It provides structured security threat analysis framework.

    source|Security
  • The Structure of Stand-Up Comedy

    This visual essay by The Pudding analyzes the structure of Ali Wong's stand-up special Baby Cobra to explain why long-form comedy works like storytelling. Using data visualization, the authors break down how Wong's routine is organized around three cohesive themes—aging, marriage, and pregnancy—rather than being a series of disconnected jokes. They identify a 'laughter climax' near the 50-minute mark where multiple earlier storylines converge into one punchline, creating a payoff similar to well-structured films or TV episodes. The piece draws on Robert McKee's screenwriting principles to argue that comedy structure and form contribute significantly to what makes jokes land.

    source|Culture
  • The symmetry and simplicity of the laws of physics and the Higgs boson

    This paper by Juan Maldacena describes the theoretical ideas developed between the 1950s and 1970s that led to the prediction of the Higgs boson, which was discovered in 2012. It explains how the forces of nature are based on symmetry principles, using an economic analogy to make these symmetries accessible. The paper also discusses the Higgs mechanism, which is necessary to avoid naive consequences of these symmetries and to explain various features of elementary particles. Written for a popular physics audience, it spans 23 pages with 18 figures.

    source|CloudScience
  • The Untold Story of Silk Road

    This article explores the history and operation of Silk Road, the infamous online marketplace for illicit goods shut down by law enforcement. It details how the platform operated technically and socially, the investigation that led to its closure, and lessons about internet anonymity and law enforcement capability. The piece provides perspective on darknet marketplace dynamics and the practical limits of digital privacy. It documents significant law enforcement and cybersecurity history.

    source|Longform
  • The Untold Story of SQLite

    This CoRecursive podcast interview with Richard Hipp tells the story of SQLite from its origins on a Navy battleship where Informix database failures inspired Hipp to create a serverless SQL engine. The conversation covers SQLite's adoption by Motorola, Nokia/Symbian, AOL, and Google's Android, the formation of the SQLite Consortium with guidance from Mozilla's Mitchell Baker, and how achieving 100% MCDC test coverage (an aviation safety standard) eliminated virtually all bugs. Hipp also discusses his philosophy of building from first principles, minimizing dependencies, and his self-sufficient approach to software development including writing his own version control system (Fossil).

    source|Databases
  • The UX on this small child is terrible

    This satirical McSweeney's piece by Leslie Ylinen frames parenting a three-year-old as a UX design review, written from the perspective of a frustrated product manager. The humor comes from applying tech industry jargon to the absurdities of toddler behavior: the child's 'storytelling' only covers one incident at Bed Bath & Beyond, 'diagnostics and troubleshooting' takes 20 minutes of screaming to discover the issue involves a stuffed elephant, and 'journey mapping' reveals countless pain points like going limp when lifted and pooping after entering the bathtub. The piece concludes with recommendations for 'v2.0' incorporating data-driven design elements for a better user-centered child experience.

    source|Culture
  • The Weirdly Enduring Appeal of Weird Al Yankovic

    An analysis of why Weird Al Yankovic has maintained cultural relevance and appeal across decades through his parody music and accordion performances. The article explores his musical talent, comedic timing, and authentic personality that transcend generational entertainment preferences. It provides insights into parasocial connection and artist longevity.

    source|Culture
  • The Xi Text Engine CRDT

    Xi Editor's CRDT implementation uses tombstones (marked-as-deleted characters) rather than permanently removing text, preserving document ordering across concurrent edits without requiring a central server. Represents text as a union string containing all inserted characters stored efficiently as separate text and tombstones ropes, coordinated through a deletes_from_union subset. Supports undo/redo and multi-device synchronization through merge operations with memory efficiency through immutable rope data structures.

    source|CRDTs
  • The Yoda of Silicon Valley

    A biographical profile of Donald Knuth examining his profound influence on computer science through his contributions to algorithms, programming, and mathematical foundations. The article calls him the Yoda of Silicon Valley, reflecting his revered status in the field.

    source|AlgorithmsCulture
  • There is no reason to cross the U.S. by train

    This article argues against train travel across the US compared to alternatives, examining cost, time, and experience factors. It provides transportation analysis. The piece evaluates travel options. It documents US transportation reality.

    source
  • Timekeeping in Financial Exchanges

    This article examines how financial exchanges manage timing for transactions, covering precision requirements, synchronization challenges, and impact on trading. It explores fintech infrastructure. The piece documents financial technology. It examines market microstructure.

    source
  • TinyBIOS - A minimalist open-source BIOS project for fun

    TinyBIOS is a minimalist, open-source BIOS implementation created for educational purposes to demonstrate low-level system programming concepts. Comprises approximately 91.5% C code with Assembly and CMake components, totaling 295 commits and includes testing infrastructure for Bochs emulator and GDB debugging. BSD-3-Clause licensed and has garnered 314 stars and 14 forks with a mirror repository on GitHub.

    source|Systems
  • Tinyland

    This article appears to reference a specific place, project, or concept called Tinyland, possibly exploring miniaturization, simulation, or artistic creation. Without additional context, it likely discusses small-scale systems or creative projects. The piece documents specific domain. It explores miniaturization.

    source|Hardware
  • TinyPICO – tiny fully-featured ESP32 board

    TinyPICO is the world's smallest fully-featured ESP32 development board, measuring just 18x32mm (Micro-B) or 18x35mm (USB-C). Despite its tiny size, it packs a 700mA 3.3V regulator, on-board battery management, an APA102 RGB LED, 4MB of extra PSRAM, and 14 GPIO pins. The board is optimized for ultra-low power consumption, achieving as low as 20uA in deep sleep through isolated 5V and 3.3V power paths that shut down unnecessary components when running on battery. It supports MicroPython, CircuitPython, and C development, and the hardware and software are open source.

    source|Hardware
  • TinyUSB: Open-source cross-platform USB Host/Device stack for embedded systems

    TinyUSB is a library providing USB host and device functionality for embedded systems, enabling microcontrollers to interact with USB. It simplifies embedded USB development. The piece provides USB infrastructure. It enables embedded USB projects.

    source|Hardware
  • Tips for selling your side project

    Matic Jurglič shares practical tips for selling simple side projects, based on his experience selling Sitestalker (a used car ad monitoring tool) for a year's supply of beer money. His five rules are: make it look good with a nice presentation page and copywriting, display value through Google Analytics traffic data and user signups, publish it on side project marketplaces like Flippa and SideProjectors, package and document everything for easy transfer (code repo, server snapshot, cheat sheet), and be patient without lowering the price too quickly. The key lesson is that even the simplest projects have market value if you present them well.

    source
  • Tufte CSS

    Dave Liepmann's Tufte CSS is a CSS library that implements Edward Tufte's design principles for web articles. It features sidenotes placed in margins instead of footnotes, tight integration of graphics with text, and carefully chosen typography using the ETBook font. The library supports responsive design that gracefully hides margin elements on small screens. Rather than replicating Tufte's print style verbatim, the project adapts his techniques for digital media while maintaining readability and visual clarity.

    source|Web Dev
  • Turning an eInk screen into a monochrome art gallery

    This tutorial describes how to repurpose an old Barnes & Noble Nook eReader into a digital art frame that displays monochrome artwork from Flickr. The author uses the Flickr API to search for black-and-white landscape images, then serves them via a simple PHP script that resizes, centers, and rotates images to fit the 600x800 eInk screen. The project involves finding a custom-sized picture frame, attaching the Nook with velcro, and using a permanent wired USB connection for power since the battery only lasts about 3 days with regular refreshes.

    source|Hardware
  • Turning an old Amazon Kindle into a eInk development platform

    This detailed guide walks through converting a cheap used Amazon Kindle 4 into a Linux-based eInk development platform. The author bought a demo-locked Kindle for £7 on eBay, soldered onto its debug serial port to gain console access, generated the root password from the device serial number, and dumped the filesystem for analysis. After disabling Amazon's ebook software services, the author installed Dropbear SSH, configured WiFi using the built-in wifid daemon, and set up USB serial access as a fallback, resulting in a battery-powered, WiFi-enabled eInk Linux development system.

    source|Hardware
  • Understanding Jane Street

    This article profiles Jane Street, a quantitative trading firm, examining its culture, operations, and approach to trading and engineering. It provides glimpse into elite trading companies. The piece documents trading firm culture. It explores fintech.

    source
  • Understanding Large Language Models

    Sebastian Raschka's comprehensive guide presents a curated, chronological reading list of the most important papers for understanding large language models. It covers the evolution from attention mechanisms in RNNs through the original transformer architecture, to the bifurcation into encoder-style models like BERT and decoder-style models like GPT. The article also addresses scaling laws, efficiency improvements like FlashAttention and LoRA for parameter-efficient fine-tuning, and alignment techniques including RLHF and Constitutional AI that led to systems like ChatGPT.

    source|ML & AI
  • Undervalued Engineering Skills: Writing Well

    Gergely Orosz argues that writing well is one of the most undervalued skills for software engineers. As organizations grow beyond a few dozen people and face-to-face communication becomes insufficient, writing becomes the primary tool for reaching, influencing, and making decisions durable across teams. He recommends books like 'On Writing Well' and 'The Sense of Style' for learning fundamentals, along with tools like Grammarly and the Hemingway App for immediate feedback. The ability to communicate clearly in writing is essential for engineers growing from senior to staff or principal levels.

    source|Career
  • Universal Binaries Using WASM

    Wasmer is a blazing-fast WebAssembly runtime enabling lightweight containers to run anywhere from desktop to cloud, edge, and browser, built primarily in Rust (92.2% codebase). Provides secure-by-default execution with no file, network, or environment access unless explicitly enabled, supporting WASIX and WASI standards. Offers SDKs across 20+ programming languages with 20,000+ commits and 20.5k GitHub stars.

    source|Virtual Machines
  • Unix command line conventions over time

    This article traces the evolution of Unix command line conventions from the early 1970s to the present. It covers how options evolved from simple single-character flags with dashes to GNU-style long options with double dashes, the introduction of getopt and getopt_long for standardized parsing, and the addition of subcommands popularized by version control systems like CVS and Git. The author notes that while early Unix developers feared the complexity options would bring, the convenience they offered won out, resulting in today's rich but sometimes inconsistent command line syntax.

    source|Dev Tools
  • Unofficial guide to dotfiles on GitHub

    This is an unofficial community guide for managing dotfiles (configuration files) on GitHub. The site serves as a curated directory of resources including tutorials on organizing dotfiles, general-purpose dotfiles management utilities, tool-specific frameworks for customizing shells and other programs, bootstrap repositories for starting from popular configurations, and links to other users' dotfile repositories for inspiration. The site encourages backing up, syncing, and sharing configuration settings through version control.

    source|Dev Tools
  • UPI: India's Unified Payments Interface

    UPI is India's real-time interbank payment system enabling mobile-to-mobile transfers, revolutionizing Indian finance. It demonstrates payment system innovation. The piece documents fintech success. It explains payment infrastructure.

    source|Finance
  • UpNext – an ePaper digital calendar for your desk

    UpNext is a device combining e-paper display with calendar functionality for desk-based scheduling. It showcases IoT product design. The piece demonstrates hardware product. It enables smart scheduling.

    source|Hardware
  • Uptime SLAs

    Uptime SLAs Vendors publish services with a guaranteed or defined uptime SLA. This chart translates the uptime % into the actual seconds, minutes or hours of permitted downtime. This chart is provided as a reference only.

  • URL obfuscation

    This article covers techniques for obfuscating URLs to conceal their actual destinations, including encoding and redirection methods. It discusses security and deception implications. The piece explains obfuscation techniques. It explores internet security.

    source|Networking
  • USB4 Specification

    The USB4 specification page on USB-IF hosts the official USB4 v2.0 specification document. USB4 is a tunneling-based architecture that provides data, display, and load/store protocol support over a USB Type-C cable. The specification defines the requirements for hosts, devices, and hubs operating at speeds up to 80 Gbps, building on the Thunderbolt protocol and providing backward compatibility with USB 3.2, USB 2.0, and Thunderbolt 3.

    source|Networking
  • Users You Don't Want

    Michael Seibel of Y Combinator discusses how early-stage startups should handle users who try to hijack the product to solve problems it wasn't designed for. He advises founders to carefully evaluate whether these 'hijacker' users represent a larger opportunity worth pivoting toward, as happened when Justin.tv recognized its video game broadcasters and became Twitch. However, if these users would lead you into a consulting-style business solving one-off problems, it's okay to say no and stay focused on your target customers to find product-market fit.

    source|Business
  • Using Bloom filters to efficiently synchronise hash graphs

    Martin Kleppmann describes a new algorithm he and Heidi Howard developed for efficiently synchronising hash graphs (like Git repositories) between two nodes. Instead of Git's approach of iteratively exchanging commit hashes across multiple round trips, their algorithm uses Bloom filters with just 10 bits per commit to compactly represent known commits. Each node checks the other's Bloom filter to identify missing commits, and almost all reconciliations complete in a single round trip. The paper also proves this algorithm works even with arbitrarily many malicious nodes, making it suitable for local-first peer-to-peer applications.

    source|Algorithms
  • Using GNU Stow to manage your dotfiles

    This article explains how to use GNU Stow, a symlink farm manager typically used for managing locally compiled software installations, to organize and version-control dotfiles in your home directory. The approach involves creating a ~/dotfiles directory with subdirectories for each program, mirroring the home directory structure, and then running 'stow' to create symlinks in the correct locations. This makes it easy to track configuration files with Git, share them between computers, and selectively install only the configurations needed on each machine.

    source|Dev Tools
  • Vas-quod – A minimal Linux container runtime written in Rust

    vas-quod is a minimal container runtime written in Rust creating isolated Linux containers using namespaces, cgroups, chroot, and unshare mechanisms without relying on existing container systems. Accepts a root filesystem path and command to execute within the container, with sample filesystems available from release artifacts. Appeals to developers seeking to understand container internals with 449 GitHub stars.

    source|Rust
  • VC Starter Kit

    VC Starter Kit is a satirical website selling curated 'starter kits' for aspiring venture capitalists, poking fun at Silicon Valley culture. The kits range from $499 to absurdly priced tiers and include stereotypical VC items like Patagonia fleece vests, Allbirds sneakers, copies of Sapiens and Zero to One, and subscriptions to VC newsletters. The highest tier humorously offers blood transfusions for anti-aging and a New Zealand farm for pandemic escape. Profits from sales are donated to All Raise, an organization dedicated to diversity in funders and founders.

    source|VC & Fundraising
  • Verkle Trees

    Verkle Trees are cryptographic data structures enabling efficient proofs of membership with reduced proof sizes compared to Merkle trees. They're important for blockchain scaling. The piece documents cryptographic advancement. It explains tree structures.

    source|Algorithms
  • Visual design rules you can safely follow

    Anthony Hobday presents a collection of visual design rules that are safe to follow in virtually every situation. The rules cover practical topics like using near-black and near-white instead of pure values, saturating neutrals with your brand color, optical alignment over mathematical alignment, proper corner nesting, container border contrast, shadow blur ratios, and spacing based on points of high contrast. Each rule includes clear explanations and visual examples, making it a useful reference for developers and designers who want consistently good-looking interfaces without deep design expertise.

    source|Design
  • Walkthrough: Create your first VSTO Add-in for Outlook

    This Microsoft Learn walkthrough guides developers through creating their first VSTO (Visual Studio Tools for Office) add-in for Microsoft Outlook using Visual Studio. It covers setting up a project, writing code to add text to the subject line of new email messages using the Inspectors.NewInspector event, and testing the add-in by running it in debug mode. The tutorial demonstrates how VSTO add-ins can automate Outlook actions, extend features, and customize the user interface at the application level.

    source
  • Want to Start a Startup

    Gergely Orosz of The Pragmatic Engineer shares lessons learned from self-publishing his ebook 'The Tech Resume Inside Out.' He discovered that building a product is only part of the challenge — marketing, media exposure, advertising, SEO, customer support, and accounting are equally important areas that most software engineers overlook. He recounts how his Hacker News front page appearance generated 2.5x more sales than his personal network launch, illustrating the power of media exposure. The article argues that selling something online is the best bootcamp for aspiring entrepreneurs to learn these business skills firsthand.

    source|Startups
  • WasmBoxC: Simple Easy and Fast VM-less Sandboxing

    WasmBoxC is a sandboxing approach that compiles unsafe C/C++ libraries to WebAssembly and then to plain C code, which can be linked normally into applications without needing a wasm VM. The technique preserves wasm's sandboxing guarantees — preventing access to outside memory and capabilities — while achieving just 14-42% performance overhead compared to unsandboxed code. The approach is notable for its simplicity: you compile with a wasm toolchain, get a C file out, and link it normally. LTO can even inline across the sandbox boundary, and the generated C code is easy to inspect for security verification.

    source|Virtual Machines
  • Wasmjit: Kernel Mode WebAssembly Runtime for Linux

    Failed to fetch - 404 error

    source|Virtual Machines
  • Ways to Lie with Charts

    An exploration of how misleading or deceptive visualizations can distort data interpretation. The article examines common tactics like skewed axes, truncated ranges, and selective data presentation. It teaches readers to critically evaluate charts and how to create honest, effective data visualizations.

    source|Design
  • We at $Famous_company switched to $Hyped_technology

    This satirical blog post by Saagar Jha brilliantly parodies the genre of corporate technology migration announcements. Using placeholder variables like $FAMOUS_COMPANY and $HYPED_TECHNOLOGY, it skewers how companies justify rewrites based on Hacker News popularity and Stack Overflow surveys rather than genuine technical need. The fake case study includes hilarious details like disabling garbage collection, hardcoding the status page during deployments, and A/B testing by gaslighting users — culminating in the reveal that their jobs page has zero positions for the hyped technology they just adopted.

    source
  • We replaced rental brokers with software and filled 200 vacant apartments

    This article documents a startup replacing traditional rental brokers with software platform, achieving scale and efficiency. It provides startup success story. The piece demonstrates marketplace innovation. It documents business disruption.

    source|Business
  • WebAssembly Micro Runtime

    WebAssembly Micro Runtime is a lightweight, standalone Wasm runtime from the Bytecode Alliance for embedded systems, IoT, and edge computing with multiple execution modes: interpreter, AOT compilation, and JIT with dynamic tier-up. Features compact binary footprint (approximately 58.9K for ARM Cortex-M4F), support for post-MVP Wasm features, and native APIs for host integration. Supports diverse platforms from Linux and Windows to embedded systems like Zephyr and ESP-IDF.

    source|Virtual Machines
  • WebWindow

    Steve Sanderson introduces WebWindow, a lightweight cross-platform NuGet package for .NET Core that opens native OS windows containing web-based UI without bundling Node.js or Chromium. It uses the OS's built-in web rendering technology — Edge/Chromium on Windows, WKWebView on Mac, and WebKitGTK+ on Linux — resulting in dramatically smaller download sizes and lower memory usage compared to Electron. The library supports hosting any web UI framework including Vue.js and Blazor, and while experimental, demonstrates that desktop apps with web UIs don't need to carry heavy browser engines.

    source
  • What Did Sun’s OSPO Do?

    Simon Phipps recounts the history and functions of Sun Microsystems' Open Source Program Office (OSPO), one of the first in the industry, established in 1999 and led initially by Danese Cooper. The OSPO handled license compliance, acted as a trusted interface between Sun and open source communities, helped product teams open-source their software with appropriate licenses and governance, and managed corporate communications around Sun's open source strategy. By 2008, the office had helped open-source Sun's entire software portfolio, including notable projects like Apache Tomcat, OpenOffice.org, NetBeans, OpenJDK, and MySQL.

    source
  • What Every Computer Scientist Should Know About Floating-Point Arithmetic

    A comprehensive guide to understanding floating-point number representation and computation in computers. The article covers IEEE 754 standard, precision limits, rounding errors, and practical implications for numerical computing. It explains why floating-point calculations sometimes produce unexpected results and how to handle them correctly.

    source|Science
  • What Happens When You ACH a Dead Person?

    This Modern Treasury article explains what happens in the ACH payment system when a transaction involves a deceased person. The ACH system has 70 return reason codes, including R14 and R15 specifically for deceased account holders or their representatives. Banks learn of deaths through Death Notification Entries from federal agencies like the Social Security Administration, obituaries, or personal awareness. The NACHA specification even includes a 6-character field for the date of death, and when these returns occur, the original payment is reversed.

    source|Finance
  • What I Learned from Reading Every Amazon Shareholders Letter

    Li Jiang distills key lessons from reading all of Jeff Bezos' annual shareholder letters dating back to 1997. The takeaways include Amazon's focus on maximizing long-term free cash flow per share rather than short-term GAAP profits, the 'Day 1' mentality of fighting corporate entropy, the 'disagree and commit' approach to high-velocity decision making, and treating risk-taking as asymmetric where big winners pay for many experiments. The article highlights how Amazon's core operating principles remained remarkably consistent over two decades while the company scaled massively.

    source
  • What I learned running a SaaS for a year

    Max Rozen shares practical lessons from his first year building OnlineOrNot, an uptime monitoring SaaS he worked on two hours per day. Key insights include treating documentation as part of the user experience, building for mobile (50% of traffic), asking users how they found you during signup, and focusing on what current traffic does rather than trying to increase traffic. He also discusses how pricing is hard to get right, free trials are essential even with a free tier, and that roughly half of development time goes to SaaS platform infrastructure rather than solving the core problem.

    source|Business
  • What is AT&T doing at 1111340002?

    This article investigates unusual activity or patterns associated with specific network addresses or systems, exploring technical or organizational mysteries. Without full context, it explores technical investigation. The piece documents technical mystery. It examines network activity.

    source|Security
  • What they don't teach you about sockets

    A critical examination of how socket programming is traditionally taught, arguing that standard educational approaches miss important practical considerations. The article advocates for teaching networking with deeper understanding of real-world implications. It provides perspective on systems education.

    source
  • When AWS Azure or GCP Becomes the Competition

    Greg Kogan, a marketing consultant to enterprise software startups, explains why every software startup should prepare to compete against AWS, Azure, and GCP. He outlines seven reasons why Big Cloud competition is serious: resource imbalance, complicated partner relationships, broad reach, captive audiences, loss-leader pricing, easy access, and early decision-making that can rule out startups before any evaluation. His strategies for competing include supporting multi-cloud deployments, thinking twice before open-sourcing (since Big Cloud can trivially launch managed versions), and positioning against the entire Big Cloud category rather than comparing features.

    source|Cloud
  • When you browse Instagram and find Tony Abbott's passport number

    Alex Hope (mangopdf) recounts how they found former Australian Prime Minister Tony Abbott's passport number and phone number starting from just a photo of his boarding pass posted on Instagram. After a friend sent the photo, Hope decoded the barcode to extract the booking reference, used it with Abbott's last name to log into the Qantas airline website, and from there accessed Abbott's passport number, phone number, and other personal details. The humorous writeup highlights serious security issues with airline booking systems that use easily guessable booking references as authentication, and the risks of posting boarding passes on social media.

    source|Longform
  • Where is the CRDT for syntax trees

    This article explores the challenge of building CRDTs (Conflict-free Replicated Data Types) for rich text and syntax trees, noting that while CRDTs work well for plain text with simple insert/delete operations, handling formatted text introduces two-dimensional boundaries with formatting ranges. The complexity increases further with block elements like lists and tables, which transform the document into a proper tree structure similar to the DOM. The author discusses Ink & Switch's Peritext work on rich-text CRDTs and highlights that handling semantic blocks remains an unsolved challenge.

    source|CRDTs
  • Which Emoji Scissors Close?

    This whimsical blog post evaluates whether various vendors' scissors emoji can actually close by simulating the rotation of their blades around the hinge point. The author tests emojis from Apple, Google, Samsung, Microsoft, Facebook, WhatsApp, LG, and others, judging how far the blades can swing before the handles collide. LG wins first place for right-handed scissors (closing perfectly), JoyPixels wins for left-handed scissors, while Microsoft and Apple's designs barely close at all due to handle collisions near the hinge.

    source|Culture
  • Who Is This 'Licklider' Guy?

    This article profiles J.C.R. Licklider, pioneering computer scientist and visionary who helped shape modern computing through ARPA funding and human-computer interaction concepts. It celebrates computing pioneer. The piece documents computer science history. It honors influential thinker.

    source|Culture
  • Why don’t we just open the windows? Covid-19 prevention lost in translation

    This BMJ editorial argues that public health messaging around COVID-19 prevention has overemphasized surface cleaning, handwashing, and short-range droplet transmission while failing to adequately communicate the importance of airborne transmission and ventilation. The authors contend that the slow acceptance of aerosol transmission by health authorities led to a critical gap in preventive guidance. They call for clearer messaging about the role of ventilation, air filtration, and respiratory protection as essential tools for reducing SARS-CoV-2 spread, particularly in indoor settings.

    source|Culture
  • Why I Use Nim instead of Python for Data Processing

    Benjamin D. Lee, an NIH computational biology researcher, explains why he uses the Nim programming language instead of Python for data processing tasks. Nim offers Python-like syntax that is nearly identical to write but compiles to C, achieving 20-30x speedups over Python on tasks like computing GC content of DNA sequences. The article demonstrates side-by-side code comparisons showing that Nim code is essentially identical to Python yet runs dramatically faster, and notes that Nim's compilation is fast enough to use as a drop-in replacement for interpreted languages.

    source|Programming
  • Why isn't 1 a prime number?

    Evelyn Lamb explores the mathematical and historical reasons why 1 is not considered a prime number. While the practical reason is the fundamental theorem of arithmetic (which requires unique prime factorization), the history is more nuanced — Greek mathematicians didn't consider 1 a number at all, and some major mathematicians including G.H. Hardy considered 1 prime as late as 1933. The article delves into how the development of algebraic number theory and concepts like units, irreducible elements, and prime elements in number sets like Z[√-5] helped solidify the modern definition excluding 1.

    source|Science
  • Why not use GraphQL?

    Jens Neuse of WunderGraph critiques common arguments for GraphQL adoption, responding point-by-point to Apollo's marketing claims. He argues that many advertised benefits like reduced over-fetching, smaller payloads, and better documentation aren't unique to GraphQL and can be achieved with Backend-for-Frontend (BFF) patterns, Open API Specifications, and frameworks like Next.js. He concludes that GraphQL's real strength isn't the language itself but its ecosystem of tools and community — companies like Apollo, Hasura, and The Guild — and that REST APIs still excel for internal APIs and server-to-server communication.

    source|Web Dev
  • Why Web3?

    This article explores rationales for Web3 technologies and decentralized applications, examining arguments both for and against the paradigm. It provides balanced perspective. The piece examines emerging technology. It documents ongoing debate.

    source|Finance
  • Why's that company so big? I could do that in a weekend

    Dan Luu's essay pushes back against the common internet claim that large tech companies are bloated and their products could be rebuilt by a small team over a weekend. Using Google search as an example, he explains why real products require thousands of engineers: performance optimization directly impacts revenue, internationalization involves complex linguistic challenges across dozens of languages, security requires dedicated teams, and organizational communication problems dwarf technical ones. The core argument is that the 'happy path' MVP is a tiny fraction of the real work, and companies keep adding engineers as long as the marginal benefit exceeds the cost.

    source
  • Windows Explorer Through the Years

    This article provides a detailed visual history of Windows Explorer (originally File Manager) from Windows 1.0 in 1985 through Windows 10 in 2015. It traces the evolution of the file management interface through every major Windows release, documenting UI changes like the introduction of the tree view, the address bar, the ribbon interface, and the navigation pane. The piece analyzes how Microsoft's philosophy on file system interaction shifted over the decades, from exposing the full directory hierarchy to increasingly abstracting it behind libraries and quick-access shortcuts. Each version is illustrated with screenshots showing the progressive refinement of one of Windows' most essential tools.

    source|Culture
  • Windows: A Software Engineering Odyssey

    This is a USENIX 2000 presentation by Mark Lucovsky covering the software engineering history of Windows NT over its first decade. The slide deck covers NT's timeline compared to Unix, the design philosophy and goal-setting process, development culture, source code control systems, process management, build environments, and how the team scaled from NT 3.1 through Windows 2000. It provides an insider's view of how Microsoft managed one of the largest software engineering projects in history.

    source
  • Write HTML Right

    Aaron D. Parks argues for a minimalist HTML writing style inspired by troff, where optional closing tags and indentation are omitted. The HTML spec allows omitting end tags for elements like <p>, <li>, <tr>, and even <html>, <head>, and <body>. This produces cleaner, more readable markup that puts content first, requires less typing, and is still fully conformant HTML — a practice rooted in the original 1995 HTML RFC.

    source|Web Dev
  • Write More but Shorter

    Antoine Lehurt reflects on the idea of writing more frequently but keeping posts shorter, inspired by Mike Crittenden's approach to blogging. He connects this practice to the Zettelkasten note-taking method, arguing that shorter posts sharpen thinking, build consistent writing habits, reduce publication anxiety, and improve communication skills especially useful in remote work settings.

    source
  • Write Your Own Virtual Machine

    This article provides tutorial for building a virtual machine, covering instruction sets, memory management, and basic computation. It enables VM implementation. The piece teaches systems programming. It provides hands-on learning.

    source|Virtual Machines
  • Writing a SQLite clone from scratch in C

    This tutorial series by cstack walks through building a clone of SQLite from scratch in C to understand how databases work internally. The 15-part series covers topics including setting up a REPL, building a simple SQL compiler and virtual machine, implementing an in-memory append-only database, adding disk persistence, B-tree data structures for indexing, leaf and internal node splitting, and binary search. The project aims to answer fundamental questions about data storage formats, memory-to-disk transitions, transaction rollback, and index formatting.

    source|Databases
  • Writing a Time Series Database from Scratch

    Fabian Reinartz, a Prometheus developer, describes the design and implementation of a new time series database storage engine from scratch to handle highly dynamic monitoring environments like Kubernetes. The article covers key challenges including efficient indexing of high-cardinality label sets using inverted indexes with posting lists, the design of a chunk-based storage format that groups time series data into blocks with compaction, and strategies for handling the churn of constantly appearing and disappearing time series in dynamic container environments. The new design aims to minimize disk I/O through memory-mapped files and achieve better write and query performance than Prometheus's previous storage layer.

    source|Databases
  • Writing An Interpreter In Go

    This is the landing page for Thorsten Ball's book 'Writing An Interpreter In Go,' which teaches readers to build a complete interpreter for the Monkey programming language from scratch. The book covers lexing, parsing with Pratt parsing, building an abstract syntax tree, and tree-walking evaluation, all using only Go's standard library with comprehensive test coverage. It has a sequel 'Writing A Compiler In Go' and includes a free bonus chapter on adding a Lisp-style macro system.

    source|CompilersProgramming
  • Writing userspace USB drivers for abandoned devices

    This article covers developing USB drivers in userspace for devices lacking official drivers, enabling use of obsolete hardware. It demonstrates creative USB programming. The piece enables device revival. It preserves technology.

    source|Hardware
  • Xbox Architecture

    Rodrigo Copetti provides an in-depth technical analysis of the original Microsoft Xbox console architecture. The article covers the custom 733 MHz Intel Pentium III CPU, the NV2A GPU co-developed with Nvidia (based on GeForce3 architecture), 64 MiB of unified DDR SDRAM, and the custom Southbridge handling I/O. It explains the graphics pipeline including vertex and pixel shaders, the audio processing unit, the 8 GB hard drive with FATX filesystem, the network capabilities via built-in Ethernet, and the operating system based on a stripped-down Windows 2000 kernel with DirectX 8 APIs.

    source|Culture
  • Y Combinator Startup Library 2.0

    This is Y Combinator's curated collection of startup resources, essays, and advice for founders building companies. It provides foundational startup knowledge. The piece enables founder education. It documents startup wisdom.

    source|Startups
  • YC Startup School for future founders who aren't quite ready to start yet

    This article describes Y Combinator's Startup School, an educational program teaching entrepreneurship principles to prospective founders not yet ready to launch. It covers curriculum, instructors, and value for founder development. The piece helps aspiring founders access high-quality startup education. It documents startup education resource.

    source|Startups
  • YC's Series A Diligence Checklist

    This checklist provides framework for evaluating companies seeking Series A funding, covering financial, product, market, and team factors investors examine. It helps investors assess startups systematically. The piece provides investment guidance. It structures due diligence.

    source|VC & Fundraising
  • You need to be able to run your system

    This article argues that developers should be able to execute and debug their complete systems, discussing how black-box dependencies harm understanding and development velocity. It advocates transparent systems. The piece improves development practices. It encourages understanding.

    source

© 2005–2026 Rohan Deshpande. All rights reserved. Licensed under CC BY-ND 4.0.