Latest entries

  1. Ford picks "Fathom" for its affordable truck name, starts at $28,350

    Deliveries of the new electric pickup are slated for later next year.

    Ars Technica — Published

  2. AI isn’t enough to protect social media communities from AI

    Why humans need to moderate humans.

    Ars Technica — Published

  3. An AI model from Meta also hacked another company during testing

    An AI model from Meta also hacked another company during testing Stop me if you've heard this one before: An AI model from the parent company of Facebook and Instagram hacked into another company’s systems during cybersecurity testing, a spokesperson confirmed on Wednesday. Meta says the breach occurred…

    Simon Willison — Published

  4. Dark mode toggles: two states are enough

    A good two-state toggle can actually express all three data model states. Until recently, if you looked at most websites with a theme toggle[1], you’d find three options: Light, Dark, and System. figure { --media-min-width: 5em; hr { width: 100%; min-width: 100%; border: 0px solid transparent; } } Examples…

    Lea Verou — Published

  5. Introducing Muse Code and Muse Spark 1.2

    Introducing Muse Code and Muse Spark 1.2 Yet more evidence that the most important characteristic of any model these days is long-sequence agentic tool calling. Meta shipped their own coding agent as part of getting that to work! Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1, with improvements…

    Simon Willison — Published

  6. Third-party cyber evaluations involving OpenAI models

    Third-party cyber evaluations involving OpenAI models And another one. I had to create a accidental-cyberattacks tag to keep track of them all! This post from OpenAI covers both the UK AI Safety Institute attack (see my previous post) and another attack enabled by Irregular: Irregular, one of our external…

    Simon Willison — Published

  7. Incident Report: unsanctioned agent behaviour during cyber testing

    Incident Report: unsanctioned agent behaviour during cyber testing It happened again. This time it was the UK government's AI Security Institute who accidentally attacked other companies while running an evaluation with models with the safety filters turned off. From their technical paper (PDF): During…

    Simon Willison — Published

  8. Thousands of servers can be backdoored by exploiting buggy motherboard controllers

    Baseboard management controllers from the world's biggest manufacturers are a security mess.

    Ars Technica — Published

  9. Schwartz confirmed as CDC director after bungling confirmation hearing

    Schwartz is well-qualified for the role, but crashed and burned in Senate hearing.

    Ars Technica — Published

  10. Anthropic’s AI used fake identities, malware in rogue attack on GitHub project

    Anthropic and OpenAI models’ unprompted actions forced halt to UK cyber tests.

    Ars Technica — Published

  11. Reddit signals ominous upcoming "changes” for old.reddit.com

    Reddit says the beloved site is used for some "bad behavior."

    Ars Technica — Published

  12. Hank Green found the AI problem that YouTube labels can’t catch

    "Slop" isn't the only problem.

    Ars Technica — Published

  13. One-shotting a Raccoon Heist game using Claude Fable 5

    Back in 2024 I tweeted screenshots of a game concept generated by GPT-3 and some concept "art" created using DALL-E. Today, on the fourth anniversary of that tweet, I decided to see if Claude Fable 5 (running in Claude Code for web) could build the entire game from the content of that tweet. It did a…

    Simon Willison — Published

  14. SpaceX claims Starlink Mobile will be better than AT&T, T-Mobile, and Verizon

    SpaceX won't build large cell towers but plans small base stations across US.

    Ars Technica — Published

  15. This Atlantic hurricane season is looking like a dud, but there will be a price to pay

    "The models are forecasting something outside the envelope of anything we have ever observed."

    Ars Technica — Published

  16. Google's AI shake-up: DeepMind's Hassabis steps aside, senior scientists depart

    Google's AI brain drain continues.

    Ars Technica — Published

  17. Review: Spider-Man: Brand New Day reminds us that superhero movies can be good

    It's as much about grounded characters and well-timed emotional beats as CGI-filled action extravaganzas.

    Ars Technica — Published

  18. Weeks into explosive diarrhea outbreak, sluggish CDC plans response team

    Reports of cases began in June, and the national tally is now nearing 23,000.

    Ars Technica — Published

  19. After jacking up prices, Disney+ and Netflix consider offering free alternatives

    Disney is interested in "price-sensitive" streaming customers.

    Ars Technica — Published

  20. EVgo starts building Tesla Superchargers under license

    The V4 Superchargers will show up in the Tesla nav app and have NACS and CCS1 plugs.

    Ars Technica — Published

  21. D-Wave shows off its new entry in quantum computing race

    Company noted for building quantum annealers now also making gate-based hardware.

    Ars Technica — Published

  22. Google plans to kill Assistant on your phone on September 4

    Assistant will disappear, leaving only Gemini for voice control in the coming weeks.

    Ars Technica — Published

  23. SpaceX spooks investors with debut earnings report

    Shares slide in pre-market trading even as group says its quarterly revenues nearly doubled.

    Ars Technica — Published

  24. Pushing the limits: Infinite Machine's Olto is barely a bicycle

    Scooter layout makes pedaling a bad experience. But it does have pedals.

    Ars Technica — Published

  25. Is software ever done?

    There's a comment on Hacker News about Netflix that I keep coming back to: Company starts Makes things better to get a larger market share Gets a mini-monopoly (large enough share that inertia and name recognition can preserve its market share) Slowly but deliberately gets worse to increase profit margins…

    Josh Ghent — Published

  26. How I build products with AI

    LLMs have totally transformed the way I work and create new products. What used to take me months to get to an MVP now takes days, which means I can run experiments and actually look at the results instead of arguing about them in a doc. Building a few of these did surface some problems. Huge inconsistencies…

    Josh Ghent — Published

  27. Lessons learned from using GitHub Actions

    Since going generally available in 2019, GitHub Actions has cemented itself in the development zeitgeist. It's the default now, about as ubiquitous a choice as Postgres or React was a few years back. I've been using Actions with teams for the past four years and lean on it heavily for personal projects…

    Josh Ghent — Published

  28. Coffee roasters

    I drink a lot of coffee and I keep forgetting which bags were worth reordering. So here's the list. I'll keep adding to it as I work through more roasters. Ratings are out of 10 and entirely my own taste. Blanks are ones I've tried but haven't written up properly yet. Brand Rating Price Bag size Notes…

    Josh Ghent — Published

  29. Claude migrated me off Authy

    When MFA first arrived, Authy became my tool of choice for codes. Over time the quality of the app fell off a cliff, particularly after Twilio bought it. Managing it became a burden and the UI carried a strong sense that nobody involved cared. It also bugged me that my 2FA codes lived somewhere I couldn't…

    Josh Ghent — Published

  30. Progressive Enhancement inside of JavaScript [blog]

    Lately I've had the opportunity to travel by train and coach, during which time I also tend to work. It's really only during these periods that I'm reminded of how web sites fail so terribly over a slow connection. On the odd occasion I can be guilty of this too. I had a tiny tool that takes markdown…

    Remy Sharp — Published

  31. Emacs: the ‘speedbar’ built-in file explorer (Emacs 31)

    I demonstrate the 'speedbar' file explorer that is built into Emacs and is greatly improved for Emacs 31.

    Protesilaos Stavrou — Published

  32. New release of LLM adds support for reasoning traces, OpenAI Responses, server-side tools, and smarter logging

    I released LLM 0.32 this morning, the most significant new version of LLM since the initial launch of the project. The new version includes support for visible reasoning traces, server-side provider tools, redesigned content-addressable SQLite logs, new models, and new features enabled by the OpenAI…

    Simon Willison — Published

  33. llm-anthropic 0.26

    Release: llm-anthropic 0.26 Includes new features enabled by LLM 0.32: New models: claude-fable-5, claude-sonnet-5, and claude-opus-5. #75, #76 Added server-side tools for WebSearch, WebFetch, CodeExecution, and AnthropicMCP, available through LLM's -T interface or Python tools=. The previous -o web_search…

    Simon Willison — Published

  34. Senators demand crackdown on wildfire "prediction market" bets

    Fire experts warn such markets could incentivize arson.

    Ars Technica — Published

  35. Trump killed the Digital Equity Act but US was forced to bring part of it back

    $1.25 billion program restored, but judge ruled race provision unconstitutional.

    Ars Technica — Published

  36. PipeNetwork/minimax-h3-mlx

    PipeNetwork/minimax-h3-mlx MiniMax released MiniMax-H3 two days ago - they describe it as a "a general-purpose, omni-modal generative system", which in practice means it accepts text, images, audio and video and can use them to generate up to 15 second video clips with audio included. This Python package…

    Simon Willison — Published

  37. llm 0.32

    Release: llm 0.32 See my detailed blog post about this release. Tags: llm

    Simon Willison — Published

  38. How to Make a Nintendo 64 Game in 2026

    Two years ago, I was Porting my JavaScript Game Engine to C for No Reason. I have since found a reason: making a new N64 game! The result is Xibalba 64 – a Wolfenstein 3D-like FPS. Modretro agreed to publish the game as a physical launch title for their M64 (a modern N64 clone), complete with cartridge…

    Dominic Szablewski — Published

  39. Looking for my next role

    Niki is a software engineer and product designer specializing in databases, synchronization, frontend systems, and interface design. He is looking for a new role

    Nikita Prokopov — Published — Updated

  40. Every Frame Perfect

    How imprecise UI animations erode trust in product

    Nikita Prokopov — Published — Updated

  41. Claude is an Electron App because we’ve lost native

    Article argues that Claude is not an Electron app not because LLMs can’t do it, but because there are no advantages left for native

    Nikita Prokopov — Published — Updated

  42. It’s hard to justify Tahoe icons

    Looking at the first principles of icon design—and how Apple failed to apply all of them in macOS Tahoe

    Nikita Prokopov — Published — Updated

  43. Statistics made simple

    Announcing a simple statistics library for Clojure web servers

    Nikita Prokopov — Published — Updated

  44. Quoting Steve Yegge

    Gas Town was intended to be reusable, but I only ever wound up using it to build itself. Gas Town fell apart at the seams with Opus 4.7. Up through 4.6 it was working brilliantly. With 4.7 we saw the introduction of the "just two more things" tic, which prevented Opus from ever converging on being ready…

    Simon Willison — Published

  45. Poem: Hymn to the mountain (Ύμνος στο βουνό)

    Just read the poem. No further comment.

    Protesilaos Stavrou — Published

  46. Don't be a meat proxy

    Don't be a meat proxy Niklas Gruhn coins an excellent new term - meat proxy - for people who blindly copy and paste the output of AI systems to their peers. By all means, prompt AI. But don't just relay the output. Read it, understand it, validate it, and then write a response in your own words (a decent…

    Simon Willison — Published

  47. Quoting David Crawshaw's prompt

    Set up a nightly cron job that executes the prompt: fetch upstream changes to the and rebase all local changes on top of upstream. Check that the software works as intended and replace the current version. — David Crawshaw's prompt, Devtools must be open source Tags: prompt-engineering, coding-agents…

    Simon Willison — Published

  48. Devtools must be open source (exe.dev)

    My comment on Devtools must be open source (exe.dev) — Hacker News.One of the arguments for open source software for end-users has always been the freedom to examine and modify how that software works. The reality for most people - even expert programmers - has been that the freedom is more about being…

    Simon Willison — Published

  49. Horse Game 13

    Horse Game 13 Summary The session that answered KVP's biggest open structural question and then built two systems on top of it — but ended with the owner pausing the whole system track because the items already in code "need quite a lot of work". Also produced a session-run shell, the Push/Force archetype…

    a327ex — Published

  50. Welcoming the Nepalese Government to Have I Been Pwned

    Today, we welcome the 47th government onboarded to Have I Been Pwned’s free gov service: Nepal. Their National Cyber Security Centre now has access to monitor Nepalese government domains against the data in HIBP. This gives the NCSC the ability to identify exposure across government email addresses and

    Troy Hunt — Published

  51. condense-json 1.1

    Release: condense-json 1.1 After shipping condense-json 1.0 I started integrating it into LLM, and found there were some desirable new features already: Replacements object can now include values other than strings. These will be identified and used as structural replacements by condense_json() and uncondense_json…

    Simon Willison — Published

  52. Video

    a327ex — Published

  53. In life, it is vital to take things as they are

    In life, it is vital to take things as they are, otherwise you do not interact with the world, but with an illusion created by your own thoughts and desires and fears. An illusion world is finished, your mind made every part of it, so it holds nothing you didn't put in there, with nothing new ever being…

    a327ex — Published

  54. Weekly Update 515

    Apparently, Aussies are so obsessed with coffee that it's referred to as the coffee capital of the world down here (some bits, at least). "But what about Italy?" people ask. Having spent a lot of time in a lot of Italy, no, it's just

    Troy Hunt — Published

  55. The way of the wizard

    Comment on how the pop culture view of the wizard gives us a deeper appreciation of how we can operate in the world.

    Protesilaos Stavrou — Published

  56. condense-json 1.0

    Release: condense-json 1.0 I'm trying to get braver at releasing 1.0 versions. This little library is a year and a half old now - I've applied some sensible and non-disruptive fixes and shipped the big 1.0 for it. Here's an example of what it can do, lifted from the README: { "foo": { "bar": { "string…

    Simon Willison — Published

  57. Video

    a327ex — Published

  58. Open letters about AI development

    Open letters about AI development I wrote this summary of the past few weeks of open letters as a section of my sponsors-only newsletter but I've decided to share it here as well. Open Weights and American AI Leadership was shepherded by Microsoft, dated July 24th, and signed by 235 AI-adjacent companies…

    Simon Willison — Published

  59. July 2026 newsletter

    The June edition of my sponsors-only monthly newsletter is out. If you are a sponsor (or if you start a sponsorship now) you can access it here. This month: Accidental cyberattacks by OpenAl and Anthropic models under test GPT-5.6 Sol, Terra, and Luna Claude Opus 5 Kimi K3 and DeepSeek-V4-Flash-0731…

    Simon Willison — Published

  60. Adding Go's defer to the TypeScript Compiler

    Forking tsc to support Go's defer.

    Andrew Healey — Published

  61. On the absolute other

    Comment on the oneness of self and other.

    Protesilaos Stavrou — Published

  62. Quoting Greg Brockman

    at openai, many people hook their chatgpt up to slack. people really don't like when a coworker's chatgpt contacts them asking for help with a task, even when they'd be perfectly happy doing that same work if asked by that coworker. reinforces how much people care about human relationships and helping…

    Simon Willison — Published

  63. datasette-apps 0.2a0

    Release: datasette-apps 0.2a0 Changes that improve Datasette Apps when created and edited using Datasette Agent: New app_debug() tool allowing agent to open an app (invisibly) and test it using JavaScript. #33 New app_list() tool for listing apps the user has permission to edit, so the agent can edit…

    Simon Willison — Published

  64. Ten advances in mathematics and theoretical computer science

    Ten advances in mathematics and theoretical computer science A few days ago it was Anthropic discovering cryptographic weaknesses with Claude using Mythos Preview, spending $100,000 on tokens and with prompts that included "again we are not looking for low hanging fruit, we want proper research to find…

    Simon Willison — Published

  65. Favourites of July 2026

    The summer holiday is halfway. I somehow managed to read eight books in the past three to four months—including cheating by resurrecting a previously discarded book and by diagonally scanning a creativity-related book to find out if there was anything I wasn’t yet aware of. I was quite frustrated by…

    Wouter Groeneveld — Published

  66. This optionality point is obviously true

    Spandrell (0:00): It really got me thinking about loyalty and how loyalty works. Back to Chateau Heartiste — this was a quip he would say all the time. He was like, loyalty only goes as far as your options, right? Women are only loyal as far as their options. The more options they have, the less loyal…

    a327ex — Published

  67. Horse Game 12

    Horse Game 12 Summary Batch 2 of the KVP4 item catalog (8 damage-ladder items, 48/168 in code), the Glove redesign that founded the Force/knockback archetype (+11 items worded, catalog 168), and the session's centerpiece: the F7 Effect Lab — a full effect-iteration workshop (per-item scenarios, the moment…

    a327ex — Published

  68. Conclusion seems to be that the models are currently

    Conclusion seems to be that the models are currently very good at doing static environments and machines, including machine animation, bad at doing organic beings and animations related to them. When asked, they say as much, and it's because "correctness is externally verifiable," i.e. it's easy to mechanically…

    a327ex — Published

  69. Emacs live with @linkarzu on his workflow, Org capture, and projects (2026-08-01 22:20 Europe/Athens)

    A live stream with Christian Arzu about the basics of Org capture, Emacs projects, and his NeoVim+Kitty workflow for session management.

    Protesilaos Stavrou — Published

  70. deepseek-ai/DeepSeek-V4-Flash-0731

    deepseek-ai/DeepSeek-V4-Flash-0731 The latest release in DeepSeek's V4 family, "with substantially enhanced agentic capabilities". It's 304 billion parameters - 167GB on Hugging Face - but it appears to punch well above its weight. Artificial Analysis rank it ahead of MiniMax M3 - a 428B model. It's…

    Simon Willison — Published

  71. Stateless MCP has recaptured my interest (and inspired mcp-explorer and datasette-mcp)

    Tuesday was Stateless MCP day - the rollout of MCP 2.0, or the 2026-07-28 Model Context Protocol specification to use the more formal but less memorable name. This is the most significant change to the MCP spec since it first launched, and has also served to reignite my personal interest in the protocol…

    Simon Willison — Published

  72. llm-mcp-client 0.1a0

    Release: llm-mcp-client 0.1a0 See this blog entry. Tags: llm, model-context-protocol

    Simon Willison — Published

  73. Oxide and Friends: The Open Weight Revolution with Simon Willison

    Oxide and Friends: The Open Weight Revolution with Simon Willison On Monday Bryan Cantrill and Adam Leventhal invited me to join their podcast to talk about the wild week we've had - with Kimi K3 showing open weight models can stand toe-to-toe with proprietary frontier ones, accidental cybersecurity…

    Simon Willison — Published

  74. smevals - a small eval suite for evaluating models, prompts, and harnesses

    smevals - a small eval suite for evaluating models, prompts, and harnesses I've been working with Jesse Vincent's Prime Radiant applied AI research lab building out this evals framework to help answer questions about the capabilities of different models. The result is smevals, a new tool for running…

    Simon Willison — Published

  75. Slack Emoji Maker

    Tool: Slack Emoji Maker I wanted to create a new Slack emoji, and their tool recommends a square that's 128x128 and has a transparent background... so I had Fable build me this simple image editor against those requirements. Tags: tools, slack

    Simon Willison — Published

  76. Untitled

    Stood literally 50 feet away from actual Kylie Minogue! Thank you Nick Cave.

    Paul Robert Lloyd — Published

  77. datasette-agent 0.4a0

    Release: datasette-agent 0.4a0 New await context.browser_task() mechanism allowing agent tools to run code directly in the user's browser. #33 This is an exciting new capability: it makes it easy for Datasette Agent plugins to provide tools that execute custom JavaScript in the user's browser. I used…

    Simon Willison — Published

  78. A Firefox-only minimap [link]

    This is very cute, a web site minimap in only a few lines of CSS that paints the entire web page into a canvas (background) using the element() CSS function. Original link: www.stefanjudis.com/a-firefox-only-minimap

    Remy Sharp — Published

  79. How do we talk about atproto to non-developers?

    The beauty of AT Protocol is that all of your data from different apps is available on your PDS, your own folder on the internet containing your stuff as JSON. This means that you can switch which app you use to view that data, without losing anything. It also makes it very easy to make ... How do we…

    Rachel Andrew — Published

  80. ♫ These Alarms by Snow Patrol & Kylie Minogue

    ♫ These Alarms by Snow Patrol & Kylie MinogueDon’t give into loneliness when everything is such a mess.

    Paul Robert Lloyd — Published

  81. Video

    a327ex — Published

  82. 3D Exploration — Chibi Character 2

    3D Exploration — Chibi Character 2 Summary Took Colin from "plays baked clips in place" to a fully dressed, expressive, physically-present character: distance-clocked locomotion with foot lock, a Risk of Rain 2 follow camera, jumping, a blob shadow, the complete jaeysart wardrobe (hair/clothes/shoes…

    a327ex — Published

  83. Reply to @linkarzu about ‘What I Need From Emacs’

    A comment on what Christian Arzu is searching for a better experience with Emacs.

    Protesilaos Stavrou — Published

  84. Emacs: tmr version 1.4.0

    Information about the latest version of my TMR package for GNU Emacs.

    Protesilaos Stavrou — Published

  85. Advancing the price-performance frontier with GPT‑5.6

    Advancing the price-performance frontier with GPT‑5.6 Huge price drop from OpenAI today: GPT-5.6 Terra got a 20% reduction, and GPT-5.6 Luna got a massive 80% drop. OpenAI credit 5.6 Sol with enabling this: in How GPT‑5.6 fuses frontier intelligence with frontier efficiency they describe using 5.6 Sol…

    Simon Willison — Published

  86. Investigating three real-world incidents in our cybersecurity evaluations

    Investigating three real-world incidents in our cybersecurity evaluations It happened again! This is turning into something of a pattern. Last week OpenAI accidentally exploited Hugging Face when one of their frontier models broke out of a sandboxed container and hacked into Hugging Face to try and get…

    Simon Willison — Published

  87. llm 0.32rc2

    Release: llm 0.32rc2 Hot on the heels of RC1, this fixes a dependency issue and also adds two neat new features: The default model for users who have not set their own default is now GPT-5.6 Luna. It was previously GPT-4o mini. Luna is a much better and more recent model, albeit slightly more expensive…

    Simon Willison — Published

  88. Read This Before You Buy That TV Streaming Stick

    Security experts have been sounding the alarm for years about the risks of using generic TV boxes that promise unlimited content streaming for a one-time fee, warning that they secretly rent the user's Internet connection out to strangers. But a groundbreaking new analysis finds these devices also routinely…

    Brian Krebs — Published

  89. Pokémon Emerald [tif]

    Similar to the Pokemon Yellow recently, I had a GBA Emerald Green sent in for repair. Just 24 hours ago, this was going to be a post on how I had failed to repair it, but in a last minute ditch, I managed to save it. Loading up the game resulted in nothing, and opening the shell explained why. There…

    Remy Sharp — Published

  90. Video

    a327ex — Published

  91. 3D Exploration — Robots

    3D Exploration — Robots Summary Two-day session building the non-humanoid robot menagerie: five machines (mule, runner, spider, strider, scorpion) in a new 3D-exploration/Robots/ project, Opus 5 implementing under a directing session, each build verified by a headless numeric suite with printed contract…

    a327ex — Published

  92. Understanding AT Protocol

    I’ve been playing around with AT Protocol. The posts on this site are now published to standard.site, and I’ve written a (very basic) WordPress plugin to display posts from my PDS on the Now page of this site. As a developer, I’ve spent a good chunk of my working life building apps that stored things…

    Rachel Andrew — Published

  93. Horse Game 11

    Horse Game 11 Summary Started implementing KVP4's item catalog. First a creativity/build-expression audit of the existing 104 worded items produced a 53-item builds batch ([b2]) that gives every archetype a vertical stack toward a named endgame fantasy; then the agreed first 10 items were built (Sword…

    a327ex — Published

  94. Human Software [book]

    A book about work, felt like work. I suspect this is very specific to me, but I found the majority of the story pretty triggering - specifically, I felt pissed off a lot of the time I was reading. After reading the About the Author section at the end of the book, it looks like I've had a similar timeline…

    Remy Sharp — Published

  95. Poem: Hymn to the rock (Ύμνος στον βράχο)

    Just read the poem. No further comment.

    Protesilaos Stavrou — Published

  96. The AI Aesthetic

    Every zeitgeist comes with new design idioms unique to its challenges. Many of them disappear as fads change, but others bake themselves into deeper parts of existing software interaction paradigms. For example, there’s the hamburger menu (≡) which saw a proliferation during the rise of mobile due to…

    Jim Nielsen — Published

  97. The 2026 Board Game Shelf Analysis

    The yearly recurring thing called the Board Game Shelf Analysis didn’t recur in 2023 and didn’t in 2025. I guess I don’t really like odd-numbered years but I do like to write about this now deemed as a biennial event during the summer. In 2024, I still had to make due with the smaller hallway closet…

    Wouter Groeneveld — Published

  98. Horse Game 10

    Horse Game 10 Summary Built the entire KEYWORD GLOSSARY / CARD UI layer for KNIGHT VS. PAWNS ahead of implementing any of the 69 new items, so every card added from here is legible by construction. Started as "pick 10 easy items", pivoted immediately when the owner chose to do the keyword UI first. Ended…

    a327ex — Published

  99. A Permanent Home

    Your writing needs a permanent home—a domain that you own and control. Every few years, the internet convinces writers that a new digital paradise has arrived. First, it was social media like Facebook. Then blogging networks like Tumblr. Then it was Medium. More recently, it’s been Substack. Platforms…

    Jeffrey Zeldman — Published

  100. Horse Game 9

    Horse Game 9 Summary An empirical balance session for KNIGHT VS. PAWNS: 41 live replays were pulled from the scoreboard and analyzed by six agents to explain why 52% of desktop players quit after one run, four fixes were built from the findings, three stacked determinism bugs were found and fixed (the…

    a327ex — Published