Prisma
@prisma.io
about 2 months ago
💬 0
♻️ 0
❤️ 0
A modern browser for the npm registry that nudges the ecosystem towards best practices and standards perfectly aligns with our mission: to make the next generation of JS developers more productive than ever before.
We're happy to support our friends at @npmx.dev! In case of @rolldown.rs, the bundler handles tree shaking at the module level, then hands off to the Oxc minifier for statement-level DCE. Both are part of the same toolchain and built to work together.
Learn more in our in-depth Rolldown DCE chapter
rolldown.rs/in-depth/dea... In practice, both work together in a pipeline: ◆ Your bundler tree-shakes unused exports across modules ◆ Then, your minifier (e.g. Oxc) then runs DCE to clean up what's left inside each module Technically, tree shaking is really a type of DCE that happens at the module/export level.
Dead code elimination (DCE) removes dead code within a module. Think unused variables, or unreachable code.
```ts
function process(x) {
if (false) { log('') } // removed
const answer = 42 // removed
return x * 2
}
console.log(process(4))
``` Tree shaking removes unused exports across modules.
```ts
// math.ts
export function add(a, b) { return a + b }
export function mult(a, b) { return a * b }
// app.ts
import { add } from './math.ts'
```
You import `add` but not `mult`? The bundler drops unused exports This is called tree shaking. Tree shaking and Dead Code Elimination are often used interchangeably but they are not the same! While both work together to make your bundle smaller, there is a difference...
Built on top of @vite.dev 8 and @rolldown.rs.
RSC support via Vite's RSC plugin.
Linting via Oxlint. adding npm package names to my mental list of things that sneakily have secondary markets - domain names - 1-800 phone numbers - instagram/twitter/et al handles - license plates - ...
Feb 25, 10 AM ET | 3 PM UK
www.linkedin.com/events/74248... When the future isn’t settled yet, teams still have to ship. Tomorrow’s MACH Forward LinkedIn Live, hosted by Jasmin Guthmann, features Dana Lawson on protocol choice, interoperability pressure, and how teams make standardization decisions while still moving forward without premature lock-in.
the second highest paid actor was the other guy in the big bang theory
did you know that the highest paid actor in hollywood from 2015 to 2018 was jim parsons, the lead actor in 'the big bang theory'
Join us tonight alongside the @convex and @tryprofound teams for a laid-back happy hour with other builders, founders, and engineers.
🍸 Private bartender and curated drinks
🍽️ Great food
🤝 Amazing conversations
6pm at the Convex office. RSVP here:
luma.com/7aky03zl 📄 Docs if you want to go deeper:
AI Gateway overview: docs.netlify.com/build/ai-gat... 📺 Video walkthrough: youtu.be/u6aeRtYj3lI
🔖 Changelog: www.netlify.com/changelog/20... Choose 'Sync with production' and Netlify starts a new agent runner session that intelligently reapplies your previous output(s) on top of the latest production deploy. Less drift. Less lost work. More confidence.
The new flow: If prod changed, you’ll get a 'Sync with production' prompt. Skip it and you’ll see a warning that you could overwrite other changes.
The problem this fixes: Publish site → run Task 1 → publish → run Task 2 → publish → surprise, Task 1 changes vanish. This showed up most for manual deploy and non Git workflows.
Running multiple agent tasks should not wipe out previously published changes. Now, when production has changed since your agent run started, Netlify prompts you to 'Sync with production' so you don’t accidentally overwrite work.
Coding agents didn't try to make the editor work remotely. They made the editor optional. Once you accept that separation, the next logical step is for agent platforms to become git forges. They already own the compute.
https://pepicrft.me/blog... Coming in tomorrow's stable release... 🧜♀️
what, seo for ai? there are like 50 companies all trying to do this, and in the process making the web immeasurably worse