Simon Willison’s Weblog

Subscribe

December 2022

Dec. 1, 2022

People are complex, and they get energy in complex ways. Some managers get energy from writing some software. That’s great, particularly if you avoid writing software with strict dependencies. Some managers get energy from coaching others. That’s great. Some get energy from doing exploratory work. Others get energy from optimizing existing systems. That’s great, too. Some get energy from speaking at conferences. Great. Some get energy from cleaning up internal wiki’s. You get the idea: that’s great. All these things are great, not because managers should or shouldn’t program/speak at conferences/clean up wiki’s/etc, but because folks will accomplish more if you let them do some energizing work, even if that work itself isn’t very important.

Will Larson

# 6:35 pm / management, will-larson

three.js examples: webgl_postprocessing_pixel (via) Neat new example for three.js that uses a pixel-shader postprocessor to apply an isometric pixel-art feel to a 3D scene.

# 9:57 pm / 3d, javascript, pixelart

Dec. 2, 2022

Datasette’s new JSON write API: The first alpha of Datasette 1.0

Visit Datasette's new JSON write API: The first alpha of Datasette 1.0

This week I published the first alpha release of Datasette 1.0, with a significant new feature: Datasette core now includes a JSON API for creating and dropping tables and inserting, updating and deleting data.

[... 2,817 words]

Dec. 4, 2022

A new AI game: Give me ideas for crimes to do

Visit A new AI game: Give me ideas for crimes to do

Less than a week ago OpenAI unleashed ChatGPT on the world, and it kicked off what feels like a seismic shift in many people’s understand of the capabilities of large language models.

[... 1,069 words]

Dec. 5, 2022

Building A Virtual Machine inside ChatGPT (via) Jonas Degrave presents a remarkable example of a creative use of ChatGPT: he prompts it to behave as a if it was a Linux shell, then runs increasingly complex sequences of commands against it and gets back surprisingly realistic results. By the end of the article he’s getting it to hallucinate responses to curl API requests run against imagined API versions of itself.

# 1:43 am / ai, gpt3, openai, generative-ai, chatgpt, llms

AI assisted learning: Learning Rust with ChatGPT, Copilot and Advent of Code

Visit AI assisted learning: Learning Rust with ChatGPT, Copilot and Advent of Code

I’m using this year’s Advent of Code to learn Rust—with the assistance of GitHub Copilot and OpenAI’s new ChatGPT.

[... 2,661 words]

Weeknotes: datasette-ephemeral-tables, datasette-export

Most of what I’ve been working on for the past week and a half is already documented:

[... 603 words]

Dec. 6, 2022

The primary problem is that while the answers which ChatGPT produces have a high rate of being incorrect, they typically look like they might be good and the answers are very easy to produce. There are also many people trying out ChatGPT to create answers, without the expertise or willingness to verify that the answer is correct prior to posting. Because such answers are so easy to produce, a large number of people are posting a lot of answers. The volume of these answers (thousands) and the fact that the answers often require a detailed read by someone with at least some subject matter expertise in order to determine that the answer is actually bad has effectively swamped our volunteer-based quality curation infrastructure.

StackOverflow Temporary policy: ChatGPT is banned

# 12:16 am / gpt3, generative-ai, openai, chatgpt, stackoverflow, ai, llms

Understanding a Protocol. Andrew’s latest notes on how ActivityPub and Mastodon work under the hood, based on his extensive development work building out Takahē.

# 12:50 am / andrew-godwin, mastodon, activitypub

I Taught ChatGPT to Invent a Language (via) Dylan Black talks ChatGPT through the process of inventing a new language, with its own grammar. Really fun example of what happens when someone with a deep understanding of both the capabilities of language models and some other field (in this case linguistics) can achieve with an extended prompting session.

# 7:30 pm / linguistics, gpt3, openai, prompt-engineering, generative-ai, chatgpt, llms

Dec. 7, 2022

talk.wasm (via) “Talk with an Artificial Intelligence in your browser”. Absolutely stunning demo which loads the Whisper speech recognition model (75MB) and a GPT-2 model (240MB) and executes them both in your browser via WebAssembly, then uses the Web Speech API to talk back to you. The result is a full speak-with-an-AI interface running entirely client-side. GPT-2 sadly mostly generates gibberish but the fact that this works at all is pretty astonishing.

# 10:52 pm / ai, webassembly, gpt3, openai, generative-ai, whisper

Introducing sqlite-loadable-rs: A framework for building SQLite Extensions in Rust. Alex Garcia has built a new Rust library for creating SQLite extensions—initially supporting custom scalar functions, virtual tables and table functions and with more types of extension coming soon. This looks very easy to use, partly because the documentation and examples are already delightfully thorough, especially for an initial release.

# 11:08 pm / sqlite, rust, alex-garcia

Dec. 9, 2022

Data-driven performance optimization with Rust and Miri (via) Useful guide to some Rust performance optimization tools. Miri can be used to dump out a detailed JSON profile of a program which can then be opened and explored using the Chrome browser’s performance tool.

# 5:19 pm / chrome, performance, rust

Dec. 10, 2022

Playing with ActivityPub (via) Tom MacWright describes his attempts to build the simplest possible ActivityPub publication—for a static site powered by Jekyll, where he used Netlify functions to handle incoming subscriptions (storing them in PlanetScale via their Deno API library) and wrote a script which loops through and notifies all of his subscriptions every time he publishes something new.

# 12:58 am / deno, tom-macwright, mastodon, activitypub

Dec. 11, 2022

Over-engineering Secret Santa with Python cryptography and Datasette

We’re doing a family Secret Santa this year, and we needed a way to randomly assign people to each other without anyone knowing who was assigned to who.

[... 2,044 words]

Dec. 15, 2022

Datasette 1.0a2: Upserts and finely grained permissions

Visit Datasette 1.0a2: Upserts and finely grained permissions

I’ve released the third alpha of Datasette 1.0. The 1.0a2 release introduces upsert support to the new JSON API and makes some major improvements to the Datasette permissions system.

[... 2,844 words]

Dec. 20, 2022

TL;DR: To serve users at the 75th percentile (P75) of devices and networks, we can now afford ~150KiB of HTML/CSS/fonts and ~300-350KiB of JavaScript (gzipped). This is a slight improvement on last year's budgets, thanks to device and network improvements. [... This is] what we should be aiming to send over the wire per page in 2023 to reach interactivity in less than 5 seconds on first load

Alex Russell

# 9:54 am / web-performance, alex-russell

Weeknotes: Datasette 0.63.3, datasette-ripgrep

Visit Weeknotes: Datasette 0.63.3, datasette-ripgrep

We’re back in the UK to see family over Christmas (our first trip back since 2019). Here are a few notes from the past couple of weeks.

[... 801 words]

Boring Python: code quality. James Bennett provides an opinionated guide to setting up Python tools for linting, code formatting and and other code quality concerns. Of particular interest to me is his section on packaging checks, which introduces a whole bunch of new-to-me tools that can help avoid accidentally shipping broken packages to PyPI.

# 7:55 pm / james-bennett, packaging, python

Dec. 22, 2022

However, with millions of new active users rushing into Mastodon, I’m forced to reevaluate that. I think I may have become too focused on what I saw of as the limits of a federated setup (putting yourself into someone else’s fiefdom), without recognizing that if it started to take off (as it has), it would become easier and easier for people to set up their own instances, allowing those who are concerned about setting up in someone else’s garden the freedom to set up their own plot of land.

Mike Masnick

# 10:56 am / mastodon

A 4.2GiB file isn’t a heist of every single artwork on the Internet, and those who think it is are the ones undervaluing their own contributions and creativity. It’s an amazing summary of what we know about art, and everyone should be able to use it to learn, grow, and create.

Danny O'Brien

# 9:47 pm / danny-obrien, stable-diffusion, generative-ai

Speech-to-text with Whisper: How I Use It & Why. Sumana Harihareswara’s in-depth review of Whisper, the shockingly effective open source text-to-speech transcription model release by OpenAI a few months ago. Includes an extremely thoughtful section considering the ethics of using this model—some of the most insightful short-form writing I’ve seen on AI model ethics generally.

# 9:49 pm / ethics, ai, openai, whisper

Dec. 24, 2022

Detailed comment on HN describing how Second Life works these days. “There are about 27,500 live regions today, each with its own simulator program, always on even if nobody is using it. Each simulator program takes about one CPU and under 4GB on a server.”

# 11:57 pm / secondlife

Dec. 28, 2022

Reverse Prompt Engineering for Fun and (no) Profit (via) swyx pulls off some impressive prompt leak attacks to reverse engineer the new AI features that just got added to Notion. He concludes that “Prompts are like clientside JavaScript. They are shipped as part of the product, but can be reverse engineered easily, and the meaningful security attack surface area is exactly the same.”

# 8:56 pm / gpt3, prompt-engineering, prompt-injection, swyx, generative-ai, llms

Dec. 31, 2022

Draw SVG rope using JavaScript (via) Delightful interactive tutorial by Stanko Tadić showing how to render an illustration of a rope using SVG, starting with a path. The way the tutorial is presented is outstanding.

# 5:31 pm / graphics, javascript, svg, explorables

2022 in projects and blogging

In lieu of my regular weeknotes (I took two weeks off for the holidays) here’s a look back at 2022, mainly in terms of projects and things I’ve written about.

2022 » December

MTWTFSS
   1234
567891011
12131415161718
19202122232425
262728293031