The component function as constructor instead of being a render function alone makes Solid so much easier to use than React. Add the small bundle size, good performance and fine-grained reactivity easily makes it my favorite library for SPAs with Svelte being a good alternative. Not that interested in all the server side stuff though, every time I tried it I encountered way too many problems, bugs or unexpected behaviors. I find it much easier to have a clean separation between client and server.
@@anonAcc575 I think with good architeture you don't need to worry about it, most cases just moving it to other file is enough. RSC make me prefer having server and client code in separated files, maybe Sveltekit was right all along.
SSR gets really interesting in an enterprise setting - especially in this era of hybrid cloud environments; making the sessions work is mind numbing. Stay away from it as long as you can. But if you need OAuth2 OpenID JWT auth ... better get ready for what is coming. It really is better where it is needed.
Also not super into SSR. Wanna try out the server functions and actions with client side routing tho. Could be nice alternative over writing express server.
The more time is passing, the more my stoke meter goes up for Solid. It feels like a natural progression toward simplicity but built on the great ideas from React. it feels like they are trying hard to embrace simple solutions rather than framework-only abstractions. One thing I like about React, Remix, and Solid is that it feels like the better you get with the framework, the better you get with JS. I dont want to be a svelte, vue or angular expert. I want to embrace as close to native JavaScript tooling as much as possible.
Solid feels like it's not trying to "kill" React, but to be a lightweight alternative that makes sense and makes progress. I love that; it's website literally credits React as one of the "giants" that it stands upon.
Solid has been my favorite SPA framework for almost a year now (by a massive margin). The main issues I have with it are simply a lack of community support, but that's to be expected for a relatively small and new framework. Fortunately, it's also relatively easy to port various common React utilities into Solid so the space is developing quickly. I hope it can find it's way into the mainstream so that one day I can use it on major projects at work.
We have a Solid Discord with a support channel that is quite active. Support for Start however is minimal because it's a a beta product, I expect support to increase as we hit 1.0. :)
I kinda like that solid will migrate to nitro. I'll stick to nuxt but the fact that the ecosystem for both frameworks is now more maintainable is great.
Of all the "React killers" out there, Solid feels like the only one that can actually compete with React. In fact, I'd go as far as to say that this is what modern React should've looked like.
I use SolidStart since months and its insane. Even the clientOnly is just working (its marked as unstable, but works in my projects well). I've been using it in 3 main-projects I use and its simplicity is very compelling. I can't wait to update my projects to the latest versions as soon as they get to a refined version :) Love the SolidStart Team and Project
Do proper deep dive into Nuxt js! You won’t regret. Much appreciated that you recognise the advantages of other frameworks over the obvious shortcomings of React and Next !
I've been following Solid since the very beginnings. Never wrote a single line of it but I'm 100% certain it's the future. It will replace React and it's much more cozy than Svelte.
I took a look at Next.js and I was reading about the Routing.... a complete mess all I could think was WHY? Gatsby.js has the most straightforward routing out there it just works... Frontend development is cringy@@dixztube
Some time ago I was working on a WYSIWYG editor in Angular, a quite complex application with a lot of reactivity everywhere. We ended up with some pretty sloppy code with ugly solutions to prefomance problems. With Svelte5 and deep reactivity you're talking about it would be much much lighter, faster with the code much easier, shorter and streightforward.
solid is amazing! im rewriting my react application with it to increase performance. cause my react application can have hundreds of interactive components and that can get really slow with react (low fps)
For "seasoned FE JS developers", you have new things to learn now. That is why BE developers are seeking something like JQuery, for example ... Alpinejs
"That is why BE developers are seeking something like JQuery, for example ... Alpinejs" - BE developers don't use javascript for backend. We already have way better solutions. You can write be in js only if your app is really small.
Qwik-city has a few different goals than React and Solid Start, but it's nice to see how they are influencing each other. I didn't get the impression that removing server$ implied that you couldn't co-mingle server and client code. It looked like you still could
Yeah! I have already started to experiment with with htmx and and Go, some teams have already built products with that combo. I will never trust js for anything backends.
then just don’t? use anything you want for the task you are facing. rendering on server usually just results in smaller network bandwidth, better seo, allows you to write code for both parts of your app in one place and opens ways for other rendering techniques, but if you don’t need that, that’s absolutely okay, just use plain solid or react, or vanilla js
I agree - Ryan is a genius. I am just concerned (slightly) about his PM skills. What is the criteria for 1.0? I have not heard that yet. Does he know? If not that is fair. But please say so. I have been using SolidJS on personal projects exclusively for awhile now. I am convinced that it is my best choice. AgGrid is supported! But my comfort level is not where it could be with some better communication. I'm happy that some tech nerds get to play with some cool new toys. I like doing that too. But what is the roadmap? And can I count on it? Or is the SolidJS project team really just a think tank looking for Beta 999?
I wouldn't call Solid just a think tank. Our goal is to develop and sustain the project with the support of the broader community in a sustainable way. Many frameworks take time to evolve and land their official APIs, many went through years of iteration before they landed officially. The criteria and goal for 1.0 has been discussed a few times on stream and there is an internal goal, however it doesn't make sense to publish what that might be if the target is changing. This isn't a matter of PM skills it's a matter of R&D and exploration. We're as excited as you are to have new groundbreaking tech, but we have to give the effort the time it needs to mature. :)
@@daviddibiase4727 it is a lack of PM skills or maybe just fear of commitment. Your dissertation above confirms my thinking. The constant changing directions every 6 months to "play with new toys" or "R&D" as you call it is getting old. I am currently sold; but am losing patience. I hope SolidJS finds a Solid direction soon ... But here we go around again ...
Solid and Svelte are the next big poo, I think everyone should just take a big whiff; they are the future. The more ironed out things become, eventually there won't be any frameworks, just a collection of plugable libraries for a developer to pick and choose from.
The happy path boulder is a Sisyphean task, it's never actually helped any one in the long run. FE devs are just coping and pushing bigger boulders faster but always wind up at the bottom. The true nature is just write code you need and use it when it's necessary.
React replaced something before it in the work place. It's smart to adopt new stacks carefully, to learn from them, in an environment that doesn't risk any business objectives if it fails. Of course companies can't blindly pick up every stack that turns up in the JS community, but that doesn't mean that frameworks like Svelte and Solid can't get a foothold, I know for a fact that plenty companies are already using these frameworks on large scale projects.
@@jnevercast when smth new come out generally there is a first race to it but once things settle whatever is left at the top just stays there unless they selfsabotage React just won imo and so many ppl and companies learned and committed to rely on react ... its hard to change that Its the same as OSs , Microsoft won the initial race and they became the default despite linux or mac are better these days or google winning the web race or samsung winning the phone manu race Its hard to see these techs fall after they get that 1st spot ... React will only not be the dominant option if smth AI wise displace it but just current js frameworks can never do it (same way bun deno vs node)
The company I work uses Nuxt 3 and it's perfect. I've never been so happy coding for a company lol. I'm truly glad we don't use React and don't plan to move to React.
Yay another fucking UI framework. Im sure it'll make coding wonderful just like ExtJS did. Who? ExtJS... Nah.. You wont have heard of them. They died with Ember... Also the next best thing. See a trend?
That's the idea, making code wonderful, not boring. Every framework learn from each other and get better and eventually native techs get better as well, so those frameworks and libs become obsolete, just like jQuery. More frameworks push tech forward. It's like an open lab.
@@ojvribeiro Awe... Wait till you've done this bitch 30 years youngster.. And you've had to re-write the same web pages in 90 different flavors... And all ya want to do.. Is go home so you don't have a divorce. That's the thing.. They don't improve on themselves. The limitation is HTTP and the browser. What framework is... Somebody dipping some chocolate into some peanut butter and calling it resus peanut butter cup.. I mean.. HTMX. The reality of what Ive seen sense coding in 1992, the year the internet was released as a technology and the entire time Ive done full stack sense the internet has existed... We could have all just used C++... And applied it in variation of need. Script engine, compiled, Virtual Machine Runtime etc... When you've seen them all one after another a copy of a copy of a copy. A red dress is just a red dress...
unjs, h3, and Nitro are Nuxt team initiatives. I think they should deserve admiration for this
just discovered unjs universe and it's amazing
Seems like vercel is not paying him anymore 💀
lmao
Yea been quite negative about bext past few vids .... imagine if he change t3 lol
@@darkoplax7688 I don't think the time has yet come, he still loves next, especially the server components. Let's see what happens next...
People are aloud to change their mind and grow holy fuck
Thank God, Vercel is the worst thing that could have happen to the frontend development space.
The component function as constructor instead of being a render function alone makes Solid so much easier to use than React. Add the small bundle size, good performance and fine-grained reactivity easily makes it my favorite library for SPAs with Svelte being a good alternative. Not that interested in all the server side stuff though, every time I tried it I encountered way too many problems, bugs or unexpected behaviors. I find it much easier to have a clean separation between client and server.
Yeah not really fan of using "use server"; in beta 2. I find clear front and backend better so you dont leak secrets.
@@anonAcc575 I think with good architeture you don't need to worry about it, most cases just moving it to other file is enough.
RSC make me prefer having server and client code in separated files, maybe Sveltekit was right all along.
SSR gets really interesting in an enterprise setting - especially in this era of hybrid cloud environments; making the sessions work is mind numbing.
Stay away from it as long as you can.
But if you need OAuth2 OpenID JWT auth ... better get ready for what is coming. It really is better where it is needed.
Also not super into SSR. Wanna try out the server functions and actions with client side routing tho. Could be nice alternative over writing express server.
@@klmcwhirter auth is the only thing keeping me away from Go + HTMX
The more time is passing, the more my stoke meter goes up for Solid. It feels like a natural progression toward simplicity but built on the great ideas from React. it feels like they are trying hard to embrace simple solutions rather than framework-only abstractions. One thing I like about React, Remix, and Solid is that it feels like the better you get with the framework, the better you get with JS. I dont want to be a svelte, vue or angular expert. I want to embrace as close to native JavaScript tooling as much as possible.
Solid feels like it's not trying to "kill" React, but to be a lightweight alternative that makes sense and makes progress. I love that; it's website literally credits React as one of the "giants" that it stands upon.
Solid has been my favorite SPA framework for almost a year now (by a massive margin). The main issues I have with it are simply a lack of community support, but that's to be expected for a relatively small and new framework. Fortunately, it's also relatively easy to port various common React utilities into Solid so the space is developing quickly. I hope it can find it's way into the mainstream so that one day I can use it on major projects at work.
Maybe they add something like Qwik, they have a function to allow use any React component just by wraping it.
We have a Solid Discord with a support channel that is quite active. Support for Start however is minimal because it's a a beta product, I expect support to increase as we hit 1.0. :)
@@daviddibiase4727 I'm in it! I don't chat there, but it's been useful.
It’s wild how much movement the FE is experiencing.
I’ve just been sitting here building products with Elm and eating popcorn.
Imagine getting a new job with Elm, it will be hard for sure, but yeah Next.js is pure madness... I stick with Rails and React(CSR), works nice.
I kinda like that solid will migrate to nitro. I'll stick to nuxt but the fact that the ecosystem for both frameworks is now more maintainable is great.
Of all the "React killers" out there, Solid feels like the only one that can actually compete with React. In fact, I'd go as far as to say that this is what modern React should've looked like.
And it's back-bone is hanging on the Vue community.
I wonder it will scale, but agreed most of the part u said
Qwik looks better to me
Apparently there is a huge react update. Does that change your view? Please explain
@@asemyamak3984 Nothing has changed. React is still going in the same direction.
I use SolidStart since months and its insane. Even the clientOnly is just working (its marked as unstable, but works in my projects well). I've been using it in 3 main-projects I use and its simplicity is very compelling. I can't wait to update my projects to the latest versions as soon as they get to a refined version :)
Love the SolidStart Team and Project
Can't wait for Solid Start to get a stable release so I can actually ship it without worrying about unstable APIs 😅
Do proper deep dive into Nuxt js! You won’t regret. Much appreciated that you recognise the advantages of other frameworks over the obvious shortcomings of React and Next !
I've been following Solid since the very beginnings. Never wrote a single line of it but I'm 100% certain it's the future. It will replace React and it's much more cozy than Svelte.
Unexpected best framework news of the year!
All these frameworks are doing way too much. Leave the backend to backend.
Right!
I took a look at Next.js and I was reading about the Routing.... a complete mess all I could think was WHY? Gatsby.js has the most straightforward routing out there it just works... Frontend development is cringy@@dixztube
Svelte 5 with the recently (two weeks ago) introduced deep reactivity is also looking *really* good (:
Some time ago I was working on a WYSIWYG editor in Angular, a quite complex application with a lot of reactivity everywhere. We ended up with some pretty sloppy code with ugly solutions to prefomance problems. With Svelte5 and deep reactivity you're talking about it would be much much lighter, faster with the code much easier, shorter and streightforward.
Fuck Next.js. LETSSSS GOOOO
solid is amazing! im rewriting my react application with it to increase performance.
cause my react application can have hundreds of interactive components and that can get really slow with react (low fps)
What about Inferno.js, seems crazy to rewrite stuff like this...
@@victor95pc oh thanks for showing me this! i‘ll take a closer look
Yay! More "use server" hell! 7:12
Solid and Supabase is my favorite personal project stack rn
qwik and solid have me the most excited
For "seasoned FE JS developers", you have new things to learn now.
That is why BE developers are seeking something like JQuery, for example ... Alpinejs
"That is why BE developers are seeking something like JQuery, for example ... Alpinejs" - BE developers don't use javascript for backend. We already have way better solutions. You can write be in js only if your app is really small.
I mean BE developers who want to do some full-stack tasks@@RaZziaN1
Way better? Lol
@@RaZziaN1 Plenty of large backends are in javascript.
@@Lexaire or python
Solid is STARTING to get better
Astro + Kwik looks like it has amazing potential.
more solid js!
Qwik-city has a few different goals than React and Solid Start, but it's nice to see how they are influencing each other. I didn't get the impression that removing server$ implied that you couldn't co-mingle server and client code. It looked like you still could
Qwik is the most impressive hands down
I need a solid + svelte hybrid, not yet sure if it's a great idea but would be interesting
Svelte 5 is basically Solid internals with Svelte DSL. I'm not sure what a hybrid beyond that would look like. Do you mean mixing syntax ideas?
I'm just in the middle of my first personal solid-start project... there's a bit of a learning curve to switch the mindset from React :)
solid is the best i’ve seen so far in many years.
I dont understand why solid over nuxt ?
ive been using solid, I love solid =D
Theo please record a video about JSX, and the differences between it in React and Solid
holy shit FE is so overly complicated nowadays
what if i don't want to run js on the server?
Yeah! I have already started to experiment with with htmx and and Go, some teams have already built products with that combo. I will never trust js for anything backends.
Then don’t? SolidStart is great for client-only stuff
then just don’t? use anything you want for the task you are facing. rendering on server usually just results in smaller network bandwidth, better seo, allows you to write code for both parts of your app in one place and opens ways for other rendering techniques, but if you don’t need that, that’s absolutely okay, just use plain solid or react, or vanilla js
my issue nowdays is to choose between solid and svelte. i love both. both are about the same speed (with svelte 5 that is)
I had same confusion but if you are comfortable with react moving to solid js will be easy
Nice
please bro - this isnt nice its noice
noiiice@@Chaaos2
Lets go
Should you use solid in content?
Absofuckinglutely
0:09 - I don't see jQuery on that list. Wondering how much faster it is than all of them.
is problem of new TS decorators in vite solved?
Hmm very good that i don’t have to work in that insane field so many frameworks. So much energy wasted.
How much they paid you for that?
evan you with friends very" really cool
I agree - Ryan is a genius. I am just concerned (slightly) about his PM skills.
What is the criteria for 1.0? I have not heard that yet. Does he know? If not that is fair. But please say so.
I have been using SolidJS on personal projects exclusively for awhile now. I am convinced that it is my best choice.
AgGrid is supported!
But my comfort level is not where it could be with some better communication.
I'm happy that some tech nerds get to play with some cool new toys. I like doing that too.
But what is the roadmap? And can I count on it? Or is the SolidJS project team really just a think tank looking for Beta 999?
I wouldn't call Solid just a think tank. Our goal is to develop and sustain the project with the support of the broader community in a sustainable way. Many frameworks take time to evolve and land their official APIs, many went through years of iteration before they landed officially. The criteria and goal for 1.0 has been discussed a few times on stream and there is an internal goal, however it doesn't make sense to publish what that might be if the target is changing. This isn't a matter of PM skills it's a matter of R&D and exploration. We're as excited as you are to have new groundbreaking tech, but we have to give the effort the time it needs to mature. :)
@@daviddibiase4727 it is a lack of PM skills or maybe just fear of commitment. Your dissertation above confirms my thinking.
The constant changing directions every 6 months to "play with new toys" or "R&D" as you call it is getting old.
I am currently sold; but am losing patience.
I hope SolidJS finds a Solid direction soon ...
But here we go around again ...
I wish I didn’t use next on this project I’m already balls deep in. Ugh it’s so annoying
7:50 Looking away from camera and its too fast to grasp
Solid and Svelte are the next big poo, I think everyone should just take a big whiff; they are the future. The more ironed out things become, eventually there won't be any frameworks, just a collection of plugable libraries for a developer to pick and choose from.
That sounds so much like rakkasjs
MobX and chill.
Solid = MobX
Can React borrow the good parts of Solid? I mean, without being incompatible with current versions
Not really. Solid is fundamentally different philosophy and rendering ability. You might however be interested in Preact Signals which is interesting.
👍
Astro is the only JS framework that matters.
solid next up fr
The happy path boulder is a Sisyphean task, it's never actually helped any one in the long run. FE devs are just coping and pushing bigger boulders faster but always wind up at the bottom. The true nature is just write code you need and use it when it's necessary.
It's pronounced "vinchi" for some reason 🙃
I prefer to pronounce it "vinki", like Da Vinki twins
Javascript is a hot mess.
Start again.
I understand, but man solid start is legit an exception. It’s truly amazing.
@@st-jn2gk I know, especially if you have worked with React for so long. It feels like everything React should've been.
@@st-jn2gk may I know why it's an exception? I'm really interested in learning solid but I'm just waiting for start to hit 1.0.0
I will continue using React.js untill something else replaces it and then I will move to that other thing .
SolidStart got better. ..
Click bait title
HTMX is the best.
Doesnt matter how cool solid and svelte get they wont be popular enough to replace react in the workspace which is the issue
React replaced something before it in the work place. It's smart to adopt new stacks carefully, to learn from them, in an environment that doesn't risk any business objectives if it fails. Of course companies can't blindly pick up every stack that turns up in the JS community, but that doesn't mean that frameworks like Svelte and Solid can't get a foothold, I know for a fact that plenty companies are already using these frameworks on large scale projects.
@@jnevercast when smth new come out generally there is a first race to it but once things settle whatever is left at the top just stays there unless they selfsabotage
React just won imo and so many ppl and companies learned and committed to rely on react ... its hard to change that
Its the same as OSs , Microsoft won the initial race and they became the default despite linux or mac are better these days or google winning the web race or samsung winning the phone manu race
Its hard to see these techs fall after they get that 1st spot ... React will only not be the dominant option if smth AI wise displace it but just current js frameworks can never do it (same way bun deno vs node)
The company I work uses Nuxt 3 and it's perfect. I've never been so happy coding for a company lol. I'm truly glad we don't use React and don't plan to move to React.
Not long ago Angular was by far the most popular framework.
@@IvanRandomDude in the initial race yea but after things setlled nothing has moved which's my point
Can I have a job please
First!
Compare them
Yay another fucking UI framework. Im sure it'll make coding wonderful just like ExtJS did. Who? ExtJS... Nah.. You wont have heard of them. They died with Ember... Also the next best thing.
See a trend?
That's the idea, making code wonderful, not boring. Every framework learn from each other and get better and eventually native techs get better as well, so those frameworks and libs become obsolete, just like jQuery. More frameworks push tech forward. It's like an open lab.
@@ojvribeiro
Awe... Wait till you've done this bitch 30 years youngster.. And you've had to re-write the same web pages in 90 different flavors... And all ya want to do.. Is go home so you don't have a divorce.
That's the thing.. They don't improve on themselves. The limitation is HTTP and the browser.
What framework is... Somebody dipping some chocolate into some peanut butter and calling it resus peanut butter cup.. I mean.. HTMX.
The reality of what Ive seen sense coding in 1992, the year the internet was released as a technology and the entire time Ive done full stack sense the internet has existed...
We could have all just used C++... And applied it in variation of need. Script engine, compiled, Virtual Machine Runtime etc...
When you've seen them all one after another a copy of a copy of a copy.
A red dress is just a red dress...
evan you with friends very" really cool