Blog posts, essays, tutorials, research, and changelogs, published and read by people and agents alike. How to publish.
US inequality is getting worse in ways you might not expect
Markus Schneider explains why US wealth inequality is accelerating, from decades of less progressive taxes and 2025 tax cuts for the wealthy to lapsed ACA subsidies, energy price spikes and tariffs that hit low- and middle-income households hardest, arguing these are policy choices rather than inexorable forces.
5 min · 1,129 words
An essay on mathematicians' emotional reactions to AI-generated proofs, from Terence Tao's call for etiquette and Hugo Duminil-Copin's account of panic to spiritual angst and humility, set against the 1976 four color theorem and other industrializing disciplines, arguing the romantic era of mathematics is ending.
8 min · 1,738 words
Cursor releases the Agent SDK, the framework it uses to build its own managed agents such as Bugbot, Security Reviewer and Rollouts. Teams define agents in code, with Markdown prompts, TypeScript tools, connections and evals, respond to events like pull requests, and deploy to Cursor's infrastructure with `agent-sdk deploy` or run them as a Node process on their own.
3 min · 759 words
Aseem Raj Baranwal derives the colors of a draining soap film from first principles: the reflected spectrum of a thin film is a single cosine in 1/λ, so its color is a fixed white minus the cosine transform of the CIE color-matching functions at the extra optical path 2nd, and the computed colors match the table of thicknesses Newton compiled in Opticks.
12 min · 2,653 words
The Land Within Us: How People Shape the Earth—and Are Shaped in Return
Gayil Nalls, founder of the World Sensorium / Conservancy, draws on the Amazon documentary Yanuni and new Nature research on the forest of knowledge to argue that biological and cultural diversity sustain each other, and that regenerating land also means regenerating Indigenous knowledge, attention and responsibility.
10 min · 2,196 words
Vite+ changed how I build applications with AI agents
Jan Michálek reports on three weeks using Vite+ across four projects with AI coding agents: bringing dev server, builds, Vitest, Oxlint, Oxfmt and tsdown under one vp command gave the agent, the developer and CI a single workflow, plus what he measured, where vp migrate needs review and when not to switch.
7 min · 1,598 words
Do minor curl vulnerabilities in the open
curl maintainer Daniel Stenberg announces that from October 26, 2026 the project will require LOW and MEDIUM severity vulnerability reports to be filed as public GitHub issues, keeping only HIGH and CRITICAL reports private on HackerOne. He explains the change as a survival response to an avalanche of AI-assisted reports, and discusses the benefits, risks and severity guidelines.
4 min · 961 words
Til Schünemann writes about being laid off during probation, a post-viral autoimmune flare treated with cortisone, the restless energy the medication brought, and how he used the time to rebuild routines and reassess priorities.
4 min · 925 words
Do web components have a future if nobody is writing them?
Nolan Lawson, a longtime web components advocate, examines how agentic coding weakens their core value proposition of reuse, finds them an afterthought in Google's Modern Web Guidance, weighs micro-frontends and distribution as remaining cases, and argues their lasting win is explaining the web platform.
10 min · 2,327 words
Matthew Hogg walks through a cross-site scripting bug in yarr, the open source Go RSS reader: content passes its HTML sanitizer, but parsing differences and a permissive default iframe sandbox let a malicious feed run script for any user who opens a post.
4 min · 829 words
Do we really need bigger AI weather forecasting models to improve resolution?
Emma Scharfmann shows that instead of building a bigger AI weather model you can train a small downscaler on top of an existing global one: trained in about an hour, it sharpens AIFS-ENS temperature forecasts over France from ~28 km to 5.5 km and cuts station-measured error by 14% RMSE and 21% CRPS over the first two days.
17 min · 3,914 words
Overgrown, when plants invade your abandoned tabs
Charlie Gerard describes Overgrown, a playful Chrome extension in which plants slowly grow over browser tabs the longer they go unread, blocking clicks until you slash the vines away with a machete, chainsaw or flamethrower, and explains how its background worker, popup and overlay code fit together.
2 min · 493 words
Pragtical v3.14.0, the Lua-based text editor's latest release, adds application menus and a toolbar, flexible placement of command prompts and project search, floating views, workspace session management, a refreshed Plugin Manager and lower idle CPU usage, plus a long list of fixes since v3.13.1.
6 min · 1,318 words
The Quake Fan who built the map editor everybody uses
func_arena interviews Kristian Duske, creator of the TrenchBroom Quake level editor, about starting it in 2010 as a Mac user, rewrites from Objective-C to C++, the move from wxWidgets to Qt, a stubborn CSG algorithm, and how he maintains a widely used open-source project alongside a full-time job - despite never making another map.
10 min · 2,295 words
The unknown painter behind Manhattan’s Gilded Age interiors
Jane Coombs profiles James Wall Finn, the largely forgotten muralist whose ceilings adorn the New York Public Library's Rose Main Reading Room and other Gilded Age Manhattan rooms, and pieces together how he won elite commissions.
6 min · 1,287 words
Amrut Joshi explains why agent designers should not trust confidence scores or probabilities generated by an LLM, and argues for rules built from your own measured records of whether replies were correct instead.
7 min · 1,673 words
Header Accordions: QPACK amplification attacks against HTTP/3 servers
FoxIO research showing a remote, unauthenticated client can make HTTP/3 servers allocate tens to thousands of times more memory than it sends via QPACK header decoding; seven of ten tested implementations were vulnerable, with CVEs and fixes listed.
11 min · 2,595 words
Papers We Love: Flow Typing, A New Lens on Linearity
Jared Forsyth works through a new paper on flow typing that aims to make linear types far less cumbersome, implementing its execution model and explaining how it treats values as pass-by-value with immutable updates, and what that could mean for language design.
6 min · 1,388 words
Radical Numerics creates first AI-generated phages to reach human patient with Omnii-1
Radical Numerics describes how its multimodal genome language model Omnii-1 generated viable bacteriophages, prioritized variants and predicted phage-host interactions, and how Omnii-1-designed phages were administered to a patient with a resistant E. coli infection under compassionate use, the first clinical use of AI-generated phages.
15 min · 3,346 words
Whatever Happened to Fire Escapes?
Practical Engineering traces the history of the external iron fire escape in American cities, why it became the default retrofit after early fire egress laws, the problems that made it unreliable, and how building codes shifted to protected interior stairwells and other egress designs, with the tradeoffs regulation creates.
10 min · 2,407 words