qskinny VS Elm

That's exactly what we are doing with QSkinny: https://github.com/uwerat/qskinny, we have been using it for a customer project for years now...

Not really, or at least, not fully. The roadmap for Qt 6 somewhat vaguely intends to have a C++ API for Qt Quick Controls (now there is, but very limited and private with no compatibility, and ugly to use). I think they are seeing that QSkinny can take some market from them.

C++ qt quick: https://github.com/uwerat/qskinny

Do you think that an approach like QSkinny would work for you? I've not used it myself in practice (just running the examples, and looking at the code), but the idea of it is that it uses Qt Quick as a backend (so I guess in Qt 6 it would use OpenGL/Vulkan/Metal/etc as needed without you having to do the abstraction), and IIRC it uses some of the private APIs that Qt Quick Controls use as well for the C++ side. QSkinny doesn't require to use QML though (should be optional, but I think C++ is preferred/more tested). It has been said also that during the Qt 6 timeline, there might be a C++ public API for Qt Quick Controls.

It kind of is (with a 3rd party library). Or you can use the private API (which I would not recommend). A public C++ API for Qt Quick Controls is planned as well, but I'm not very optimistic on the time frame.

Elm [1] is based on a similar idea. Build your app from pure functions that return HTML tags.

[1] https://elm-lang.org/

You also wouldn't really be creating your own new programing language. You would be creating something that can run JavaScript by following JavaScript standards and syntax. You might be able to add some non-standard features of your own on top of those standards, or include your own standard library of helpers or utilities, but you can't completely make a new or alternative language and then load it in the browser (or at least not by reimplementing ECMAScript standards... you actually can make your own language that runs within any Javascript enviroment, if you provide an interpreter or compiler that transforms it into valid JS. Some people have done something like this, eg Elm: https://elm-lang.org/).

You should at least have a look at https://elm-lang.org/ it is a pure functional language like Haskell (although with fewer fancy syntax/type classes) but it has some lovely libraries for visualisation and even with plain elm (+ elm-ui) doing string transformations can be easily done.

I get it. However, the whole point of using Unions to narrow your types, ensure only a set of possible scenarios can occur, and only access data of a particular union when it’s safe to do so. That’s some of what pattern matching can provide, and 100% of what using switch statements in TypeScript with their Discriminated Unions can provide. Yes, it’s not 100% exhaustive, but TypeScript is not soundly typed, and even Elm which is still has the same issue TypeScript does: You’re running in JavaScript where anything is possible. So it’s good enough to build with and much better than what you had.

The beauty of a language like Elm (and other lambda-calculus / functional programming inspired languages) is that there's very little transformation involved in going from an idea to code. And that seems to have a big impact on getting things done.

I think the most fun and approachable way for beginners to build games with functional programming is with Elm [1].

See a few (small, demo) games built by the community in [2] .

Notice Elm has abandoned the FRP approach in favor of Model-View-Update [3].

[1] https://elm-lang.org/