task: Fork Dioxus `native-core` crates about freya HOT 4 OPEN

I think this idea will bring more autonomy, and allow Freya to follow her own path, but it would also add another layer of complexity. So far, using Dioxus allows you not to worry about a series of problems, but as you correctly observed, it makes it difficult to improve performance.
Programming user interface libraries is not easy and having more control over the code is a way to improve the project. I think the Rust community will greatly benefit from a library without dependency hell issues.

from freya.

I think this idea will bring more autonomy, and allow Freya to follow her own path, but it would also add another layer of complexity. So far, using Dioxus allows you not to worry about a series of problems, but as you correctly observed, it makes it difficult to improve performance. Programming user interface libraries is not easy and having more control over the code is a way to improve the project. I think the Rust community will greatly benefit from a library without dependency hell issues.

Instead of "another layer of complexity" I would say "more code to maintain". But yeah, it is true that not having to worry about it would be certainly better when it comes to maintenance, but I think the potential performance optimizations and perhaps removal of abstractions in favor of implicit implementations might be worth doing

from freya.

I agree and it's a good idea

from freya.