Lightnews — Scholar-powered news

Alexander Lichter

@thealexlichter.com

5.9K followers 690 following 890 posts

💻 DevRel @voidzero.dev 🎬 Video Content https://lichter.link/yt 💚 @nuxt.com team 🎙️ Podcast Host @dejavue.fm

lichter.link

Posts Media Videos Starter Packs

Pinned

Alexander Lichter @thealexlichter.com · Mar 19

🎉 Excited to announce that I'm joining VoidZero as DevRel!

As I was following the project from early on, I couldn't say no to the chance when @evanyou.me reached out.

I can't wait to share all the incredible things lined up for @rolldown.rs, OXC, @vite.dev, and @vitest.dev 👀

34 11 310

Alexander Lichter @thealexlichter.com · 2h

It never stops 👀

Rolldown @rolldown.rs · 2h

In the next version of Rolldown, bundling will be ~10% faster 🎉

This was achieved by optimizing the semantic scoping process

Internal plugins like define and inject now flag when they haven't changed the AST, allowing to skip redundant scope recreation

github.com/rolldown/rol...

fix(rolldown): sync scoping properly in pre_process_ecma_ast by Boshen · Pull Request #6537 · rolldown/rolldown

We need to recreate scoping after each transformation step because none of the steps syncs scoping at present, which is a huge waste that we should fix later in the future. But in the interim, defi...

github.com

Alexander Lichter @thealexlichter.com · 4h

I'm confident that adoption can be even incentivized with a generous free tier 🙌

Alexander Lichter @thealexlichter.com · 9h

github.com/oxc-project/...
github.com/oxc-project/...

Yes, they are supported!

feat(transformer): support for transforming legacy decorator by Dunqing · Pull Request #8614 · oxc-project/oxc

github.com

Alexander Lichter @thealexlichter.com · 10h

Faster.

Alexander Lichter @thealexlichter.com · 10h

github.com/vitest-dev/v... might help in the meantime

Issue with TypeORM decorators when running vitest in TypeScript project · vitest-dev vitest · Discussion #3320

I'm working with TypeORM and I noticed that my code is not working when I don't explicitly set the column type, for example @column({ type: 'varchar' }). I've already added "experimentalDecorators"...

github.com

Alexander Lichter @thealexlichter.com · 10h

Heads up that Oxc doesn’t support transforming stage 3 decorators yet! We will when the proposal hits stage 4

github.com/oxc-project/...

transformer: ecma decorators · Issue #9170 · oxc-project/oxc

https://github.com/tc39/proposal-decorators?tab=readme-ov-file#how-should-i-use-decorators-in-transpilers-today esbuild: evanw/esbuild#3754 babel: https://babel.dev/docs/babel-plugin-proposal-decor...

github.com

Alexander Lichter @thealexlichter.com · 10h

You can do by overriding the Vite dep with Rolldown-Vite. It likely won’t change too much for now though

Any issues you’ve hit?

1 1

Reposted by Alexander Lichter

VoidZero @voidzero.dev · 14h

Watch @antfu.me talk about the new @vite.dev Devtools and what Vite DevTools Plugin means.

Full video in comment 👇

2 8 45

Alexander Lichter @thealexlichter.com · 1d

Luckily a lot of users won't have to pay as we will have a free tier covering OSS, non-commercial and smaller companies.

Instead of gatekeeping, the goal is to make it usable without paying upfront when not making money yet or not planning to make any.

Alexander Lichter @thealexlichter.com · 1d

It is about providing a unified toolchain instead of duct-taping everything together.
A seamless out of the box experience with tools using the same underlying tech (Rolldown & Oxc) for performance & consistency.

More in the newest announcement post.

voidzero.dev/posts/announ...

1 1

Alexander Lichter @thealexlichter.com · 1d

voidzero.dev/posts/announ... should help with this!

Announcing Vite+

Introducing Vite+, a unified toolchain for JavaScript.

voidzero.dev

1 1

Alexander Lichter @thealexlichter.com · 1d

Then you'll be happy to hear that most of the underlying parts that Vite+ uses are FOSS and will stay, including @vite.dev, @rolldown.rs, @vitest.dev, and Oxc.

Ideally, Vite+ can funnel back to OSS and contribute to the Sustainability of these Open Source projects!

Reposted by Alexander Lichter

Rob Palmer @robpalmer.bsky.social · 1d

A blog post on Vite+ and the future of web tooling 📜

🔶 A new CLI that combines Vite, Vitest, Oxlint, Oxfmt
🔶 Centralized simple config
🔶 Adds easy monorepo tooling
🔶 Free for individuals, OSS, & small companies
🔶 Existing tools remain independent & assertively MIT

It’s important to emphasize that Vite+ is a separate, additive layer built on top of the open source projects we maintain. All existing projects — Vite, Vitest, Rolldown, and Oxc — will remain open source under MIT forever. Since Vite+ relies on these open source projects, improving Vite+ requires improving them as well. You are welcome to hold us accountable to that commitment.

2 24

Reposted by Alexander Lichter

VoidZero @voidzero.dev · 1d

Watch VoidZero’s @overlookmotel.bsky.social explain how oxlint tackles JS-Rust interoperability to maintain high performance for plugins.

Full video in comment 👇

3 7 41

Alexander Lichter @thealexlichter.com · 1d

In what sense?

Reposted by Alexander Lichter

Rob Palmer @robpalmer.bsky.social · 3d

This week's @viteconf.org was a LOT of fun.

It rekindled some of the neutrality/plurality magic from the old JSConfEUs. Like Switzerland. Bridging communities & frameworks. Folk cheering each other on, celebrating collaboration 🎉

TechAtBloomberg folk showed up in force - 7 of us!

Evan You on stage at ViteConf announcing Vite+

1 5 74

Alexander Lichter @thealexlichter.com · 1d

Non-commercial use will be free, yes!

2 1

Alexander Lichter @thealexlichter.com · 1d

Whoops! Updated to jestjs.io
Thanks for reporting 🙏

Jest

jestjs.io

Alexander Lichter @thealexlichter.com · 1d

Non-commercial use is falling under the free tier.

Alexander Lichter @thealexlichter.com · 1d

It depends on the pricing and value proposition eventually. Working on Vite+ also wasn't done "blindly"
Companies are interested in standardization, a toolchain that "just works" without investing the time of gluing together tooling solutions and spending engineering time that could go into features

Alexander Lichter @thealexlichter.com · 1d

Also, the difference between a hosting platform vs. a toolchain is another thing to mention here.

Alexander Lichter @thealexlichter.com · 1d

I understand your concerns there but I don't think that'll happen. Here is why:

* Incentive to do so would only be there for features relevant only to companies & enterprises as there will be a fbig ree tier
* The success of Vite+ is directly tied to the stability and adoption of OSS Vite ecosystem

1 1

Reposted by Alexander Lichter

Working Draft @workingdraft.de · 1d

Revision 684: Ecosystem Performance (@e18e.dev) workingdraft.de/684/ mit @thealexlichter.com und Ulrich-Matthias „Ulima“ Schäfer.

Diese Revision wird gesponsert von @europdf.eu ✨ , eurem SaaS zur Erzeugung hochwertiger, barrierefreier PDFs mit Hilfe modernem HTML/CSS/JS – und mit Sitz in der EU 🇪🇺

Revision 684: Ecosystem Performance (e18e) | Working Draft

Diese Revision plaudern wir mit Alexander Lichter (LinkedIn / Mastodon) und Ulrich-Matthias „Ulima“ Schäfer (LinkedIn) über Ecosystem Performance, kurz „e18e“: Warum es sich lohnt, das JavaScript-Ökos...

workingdraft.de

5 7

Alexander Lichter @thealexlichter.com · 1d

Right? 🙌

Alexander Lichter @thealexlichter.com · 1d

Not sure what you mean by "slow rug pull" exactly?

Also Vite+ is largely based on OSS components, so keeping Rolldown, Vite, Oxc etc. in good shape will benefit everyone.