A ripple of excitement pulsed through the crowd. Paul Ford is blogging regularly again.
The folks behind Dark Sky spun themselves out of Apple...
The folks behind Dark Sky spun themselves out of Apple and have built a new weather app: Acme Weather. “We missed those days as a small scrappy shop. So let’s try this again…”
Reply guy
The latest scourge of Twitter is AI bots that reply to your tweets with generic, banal commentary slop, often accompanied by a question to "drive engagement" and waste as much of your time as possible.
I just found out that the category name for this genre of software is reply guy tools. Amazing.
Tags: ai-ethics, twitter, slop, generative-ai, definitions, ai, llms
Quoting Summer Yue
Nothing humbles you like telling your OpenClaw “confirm before acting” and watching it speedrun deleting your inbox. I couldn’t stop it from my phone. I had to RUN to my Mac mini like I was defusing a bomb.
I said “Check this inbox too and suggest what you would archive or delete, don’t action until I tell you to.” This has been working well for my toy inbox, but my real inbox was too huge and triggered compaction. During the compaction, it lost my original instruction 🤦♀️
Tags: ai-ethics, generative-ai, ai-agents, openclaw, ai, llms
Red/green TDD
The Atlas Obscura Podcast
• SiriusXM Podcasts & Atlas Obscura
Lost Wonder: Minister’s Treehouse (Classic)
The world’s biggest treehouse was inspired by a message from God.
This week, we’re bringing you stories of places that, for one reason you another, you can’t visit.
Plus: We want to hear from you! What’s a place in your life that YOU can’t visit? Give us a call at 315-992-7902 and leave a message, or send an email to hello@atlasobscura.com.
Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
The Claude C Compiler: What It Reveals About the Future of Software
London Stock Exchange: Raspberry Pi Holdings plc
21.02.2026
20.02.2026
Sentry
My thanks to Sentry for sponsoring last week at DF. Sentry is running a hands-on workshop: “Crash Reporting, Tracing, and Logs for iOS in Sentry”. You can watch it on demand. You’ll learn how to connect the dots between slowdowns, crashes, and the user experience in your iOS app. It’ll show you how to:
- Set up Sentry to surface high-priority mobile issues without alert fatigue.
- Use Logs and Breadcrumbs to reconstruct what happened with a crash.
- Find what’s behind a performance bottleneck using Tracing.
- Monitor and reduce the size of your iOS app using Size Analysis.
Link: sentry.io/resources/ios-workshop-jan-2026/?utm_source…
How I think about Codex
🎬 Slingshot

This review may contain spoilers. Tap or click to view.
Good space movie. Has great atmosphere overall (apart from a few soundtrack choices maybe). Liked the themes of the movie. The acting was great for the most part. And it has a good twist (or two?)! The very end is a bit unnecessary, when they could've just ended the movie right with the countdown in the airlock.
👍 Recommended!
The Atlas Obscura Podcast
• SiriusXM Podcasts & Atlas Obscura
Tell Us Your Unusual Spring Break Stories
We’re willing to bet that AO listeners didn’t go on the typical spring break trips. We want to hear your off the beaten track spring break travel stories. Where did you go? What made this trip so memorable? Did everything go as planned? Or did things go wrong? We want to hear it all.
Give us a call at (315) 992-7902 and leave a message telling us your name and story. Or you can record a voice memo and give it to us. At hello@atlasobscura.com.
Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
Quoting Thibault Sottiaux
We’ve made GPT-5.3-Codex-Spark about 30% faster. It is now serving at over 1200 tokens per second.
— Thibault Sottiaux, OpenAI
Tags: openai, llms, ai, generative-ai
Andrej Karpathy talks about "Claws"
Adding TILs, releases, museums, tools and research to my blog
Taalas serves Llama 3.1 8B at 17,000 tokens/second
Taalas serves Llama 3.1 8B at 17,000 tokens/second
This new Canadian hardware startup just announced their first product - a custom hardware implementation of the Llama 3.1 8B model (from July 2024) that can run at a staggering 17,000 tokens/second.I was going to include a video of their demo but it's so fast it would look more like a screenshot. You can try it out at chatjimmy.ai.
They describe their Silicon Llama as “aggressively quantized, combining 3-bit and 6-bit parameters.” Their next generation will use 4-bit - presumably they have quite a long lead time for baking out new models!
Via Hacker News
Tags: ai, generative-ai, llama, llms
Recovering lost code
Reached the stage of parallel agent psychosis where I've lost a whole feature - I know I had it yesterday, but I can't seem to find the branch or worktree or cloud instance or checkout with it in.
... found it! Turns out I'd been hacking on a random prototype in /tmp and then my computer crashed and rebooted and I lost the code... but it's all still there in ~/.claude/projects/ session logs and Claude Code can extract it out and spin up the missing feature again.
Tags: parallel-agents, coding-agents, claude-code, generative-ai, ai, llms