It’s a simple question on paper, but I wanted to ask a wider community on their thoughts and struggles with the current UI libraries. I’m finding out whats hard, annoying, or could use a bit of improvement. All of this feedback will be taken into account and reciprocated into a new library of my own.
Speed, syntax that feels like it fits in with the language (e.g. jsx), and NOT being React. Reactive programming is great, but don’t be another React clone.
My favorite UI library DSL: https://luau.org/
Works all the time 101%
No need for silly “addons”
Jokes aside, UI libraries are just stupid.
It’s a problem of a programmer all the time and not the language.
Be more creative. After all, bro, humanity has literally evolved for thousands of years to be creative, and you, like a normie, are going to slap react? BRO WHAT? 
local new = Instance.new
local Frame = new "Frame" do
Frame.BackgroundColor3 = Color3.new(1,1,1)
end
i agree there really is no need for a ui library, i don’t plan on making this into a large scale open-source driven project. as a programmer, its always nice to bake your own cake for once, instead of using pre-made ingredients. i feel package managers like npm have ruined what it means to be called a “programmer”. im just a hobbyist who wants to make a nice proof of concept.
I disagree. If I want a good XML parser, for instance, I’m not going to write the parser myself. I might do that as its own project, but if I just need one right now and immediately, I should have the ability to say packagemanager install xml-parser
I agree. With you.
Honestly, I think a UI library should have Good documentation and examples. That’s one of the biggest things that makes or breaks a library for me. I don’t use libraries anymore nowadays, but back in the days when I did, clear examples made a huge difference. It’s just so much easier to get started when everything’s explained well instead of having to guess how things work.
your missing the point, in this scenario, its great!
think about it like this, if i want an xml parser, quick and dirty, i will rely on a library. but if i’m using this a bit more than i thought, i would want to also explore programming that avenue as well. this might be my curious nature, but imo this has helped a lot when learning.
yes, there are libraries even I use, but i find it important to also branch out, and test whats uncomfortable, because ultimately as a programmer, its your job to problem solve.
the issue lies when people overuse a library and become dependent on its existence.
Well, if it performs the same or better as the library, I’ll use mine. Most of the time it won’t though, so I’ll just use the library. I feel like reinventing the wheel does help you learn, but you shouldn’t always use your version of the wheel.
i feel this is short stunting a little, i would take the oppurtunity for growth and to learn.
i could ask myself questions like how does A outperform B?
i can use those to better myself in the future and make a new wheel.
this approach might not be the best for everyone tbh but ive been getting consistent results for a while