I can tell it what I want it to do, iterate on the idea and see how it changes. After I did that a few times I was able to tinker myself on the project and again ask ChatGPT how to improve the code.
Perhaps it’s just my learning style, but it has been really refreshing experience to learn with something(someone).
Feels like a huge burden lifted off my shoulders
When I'm studying a new subject I need guidance rather than a fire-hose of links to resources, but that may just be me.
I program as a hobby without any formal training. Despite doing it on and off for a few decades, I still consider myself an 'advanced beginner' for those reasons. So, having a lay of the land already, a resource like this is helpful when I'm looking for information in a specific area within front-end development.
Somewhat tangential: it's easy to drown in the surfeit of resources out there, to the point of paralysis. I've come to the conclusion that, in the end, you end up teaching yourself. To that end it almost doesn't matter what resource you start with. Just pick one and start asking and answering questions. For those questions you can't answer, the relevant resources will reveal themselves through your search.
+1 to the other commenter's suggestion that GPTs are a huge help in this regard, as long as you set a requirement for yourself not to use any code you don't understand. And GPTs are incredibly helpful with this meta step too.
https://news.ycombinator.com/item?id=39402244
And it comes up in the Show HN-specific rules https://news.ycombinator.com/showhn.html
(One that explains why every step is necessary, and what exactly it does in greater detail, as well as the underlying structures. YT tutorials are a bit outdated and are bad at explaining details.)
Learning after react will give you better idea about complexities and problem statements both the framework solves.
I would try making a small app in React first, and then moving to Svelte to get an idea of why Svelte was made in the first place.