Ryan Carniato
@ryansolid.bsky.social
6.8K followers 78 following 420 posts
got signals? @solidjs.com @sentry.io previously: @netlify.com @markojs.com
Posts Media Videos Starter Packs
ryansolid.bsky.social
Received this in the mail today for my son. I see the @solidjs.com mascot is alive and well.

Thanks @davedbase.com!
plush duck in the shape of the solidjs logo
Reposted by Ryan Carniato
tkdodo.eu
It turns children into siblings 🤯. @ryansolid.bsky.social blowing every React dev's mind with nested data fetching and fine-grained reactivity 🔥

www.youtube.com/live/fnyK-xX...
JetBrains JavaScript Day 2025
YouTube video by JetBrains
www.youtube.com
ryansolid.bsky.social
Although to be fair my biggest interest here is nested statements, not native control flow. That is just a bonus. More fine-grained solutions don't need state to be owned by components and can nest it at will. Which is powerful because it removes a lot duplication around control flow logic in state.
ryansolid.bsky.social
JSX hasn't proven this isn't needed. You might argue React's model has never been able to leverage this so JSX is perfectly adequate. I think your response is inline with what I'd expect on the React side of the spectrum. This is different enough to feel uncomfortable.
ryansolid.bsky.social
The idea isn't making this Solid specific. I'm just pointing out at some point JSX has to evolve if it is going to keep up with other innovations in templating in the space. Those others don't have certain characteristics that JSX has and can't preserve backwards compat. We'd still support JSX.
ryansolid.bsky.social
Nested declarations of state need statements in JSX which is basically the problem. With fine-grained components don't matter, so putting createSignal under and if actually can simplify logic and make it more portable. Native control flow was just a bonus. Could help with type narrowing.
ryansolid.bsky.social
Yeah you wouldn't most likely. Would just need to build it out like any other tool chain with its own extension.
ryansolid.bsky.social
The key is the extension isn't JS. This is a different language that happens to be a superset of JS. It introduces new type of block. Could just stop there and not have a `component` keyword and just use the `@{}` on functions too.
ryansolid.bsky.social
The best part of the proposal is people can. Each feature is separately adoptable.
ryansolid.bsky.social
Do expressions look like they are never happening. But the templating power is a real thing. Especially for fine-grained systems that could locate state anywhere in the template.
ryansolid.bsky.social
JSX are templates? Obviously I want to use templates.
ryansolid.bsky.social
There could be a reference transform for Babel, which then could be used in Vite etc as a baseline. But yes this is a whole other thing so it requires language server, code completion etc..
Reposted by Ryan Carniato
danielroe.dev
this is worth saying:

nuxt is an independent open source project and not directed or owned by any company.

check out github.com/nuxt/governa... for specifics.

you can see the companies which sponsor the project financially on OpenCollective and GitHub Sponsors.
GitHub - nuxt/governance: Nuxt Project Governance
Nuxt Project Governance. Contribute to nuxt/governance development by creating an account on GitHub.
github.com
ryansolid.bsky.social
I can definitely respect the desire to be principled here. The first thing I noticed is not having all the facts in terms of which projects would be impacted. And ultimately if people are going to arbitrary about it, I'd like to point them at a more sensible line.
ryansolid.bsky.social
My interest is obviously in preserving science and innovation. OSS support is power in the same way as philanthropy is. These projects run independently and do important work. Work that is far more important than who currently sponsors that work.
ryansolid.bsky.social
Sorry I don't mean about specifically the Guillermo issue. His last post was about social responsibility, and was being critical of people justifying posting on X because that is where their followers are.
ryansolid.bsky.social
It doesn't even directly impact me but there is so much wrong with that. Obviously nothing in the face of the genocide. But it doesn't make it anymore right.
ryansolid.bsky.social
Alright I suppose I'm done then. It just really really irks me how everything gets lumped in with that asshat as you called him. The dedication and passion that goes into keeping OSS takes a big enough toll without this sort of garbage. Then making these projects targets.
ryansolid.bsky.social
These jobs aren't normal jobs. They get to work on projects for the betterment of everyone without contributing to the bottomline of the companies that employ them. It isn't about finding 7 positions. It's about finding 7 positions worth of pure charity. In this economy too..
ryansolid.bsky.social
When this can have real impact on projects that do important work I will say something. It's hard enough to make a difference.

I'm not saying don't hold Guillermo and Vercel accountable.
ryansolid.bsky.social
So in a sense while it isn't impossible to find new Patrons, it takes some doing. The real challenge is this allowed teams to grow in a way they haven't historically. Backtracking on that is basically spells out the end.
ryansolid.bsky.social
The alternatives are like Next, where you are ran by the company and part of their roadmap. You can start your own company but then you need to come up with an angle to be profitable. Hard when your software is innately free. And multi-sponsorship is unstable at best. Look what happened to Babel.
ryansolid.bsky.social
We can get into the funding models for open source. Maybe I will do a stream about it. The problem is the "patronage model" is probably the best we have right now in terms of balance. It just works a lot better for researchers than it does for production tools, since so dependent on single entity.