We think this belongs where the conversation with the agent is happening, not in Slack. With DeltaDB you can jump into a teammate's thread and work alongside them and their agent.
Sign up for early access: https://t.co/JH0QKHXLiA
@Neolectron That's a fair point! Honestly, it took me years to set it up, AI whipped it in a single session. If anything ever drifts or breaks, I'll just point AI at it again. So not much to worry
I was always way too lazy to get my dotfiles properly set, I postponed it for years, AI just got it done in 30min, just like I'd expect
https://t.co/vJQEelrM2M
New @elmlang release today!
🎉 0.19.2 🥳
Incremental builds of 850k lines of Elm in <350ms. Gearing up for a batch of smaller releases as we head towards Elm 1.0.
https://t.co/RNd5snG2Ia
Nice! And good call about being the fastest. I explicitly explain it in the docs too
"the speed comes from mutating in place, and that costs you things:
No time-travel or persistence. Mutating in place means there’s no history to replay and no snapshot to serialize. Spawned actors and nested states are out too. If you’re modeling full application state (undo/redo, save-and-restore, deep actor trees), reach for XState; that’s exactly what it’s built for, and it’s excellent at it.
Dunky is for UI behavior, at scale."
Took XState v6 alpha for a spin.
Benchmarked it vs v5 and Dunky, the same little state machine in each engine, and fired the same event over and over in a tight loop.
Looks like it does a bit more work under the hood, but it's early alpha, so just a peek 🔬
Digging into 12 libraries before building Dunky's first components (Radix, shadcn, React Aria, Base UI, Chakra, Mantine, Carbon, Fluent, Polaris, Ant Design, Spectrum, Base Web).
The goal was to figure out what to build first by looking at what everyone already agrees on. The components common to all of them were Button, Dialog, Select, Tabs, Tooltip, Checkbox, Radio, Switch, and TextField. That's the first set.
I also tried to rank them by real usage, but most "most-used component" npm stats are junk. Nearly every lib shows a flat % across all components because the umbrella package re-exports all packages. Adobe Spectrum was the only exception, the only one where components are installed independently, and the numbers mean something.
Now the fun part starts, Dunky's state machine will be used as the guardrail to drive agents to do specs → test → code. Let's see what comes out. 🚀
🚀 Introducing Astryx from Meta – an open source design system that’s AI fluent and fully customizable without dependencies. 150+ accessible components, brand-level theming, ship-ready templates, and a CLI. Built at Meta over 8 years, now yours. Astryx Design → https://t.co/rBHFBYS76h and GitHub → https://t.co/OyHbxy4Jrl
Last week I introduced the first Dunky lib.
Why? Shipping design systems across different platforms meant always rebuilding the components. Endless rewiring.
The full vision is here
👉 https://t.co/mFhSqX28jN
screaming here to the void (since I have almost no followers 🫠)
Im adding Vue, Solid and Svelte connectors to Dunky state-machine (guided by AI)
Im no expert on those, any help is welcome
🧵