[AlienRobot]

In my view, the problem isn't specifically reactivity but the fact that reactivity isn't actually native of the UI toolkit.

Instead of HTML, think about GTK or Swing.

To add React-style "reactivity" to it, instead of just making a dialog to change the "title" of a document and committing the change when you press OK, you'd need a top-level "App" class that holds all the state, a class for state properties with IDs accessible at runtime which probably would be a variant (accepts any primitive type), a binding class to bind the toolkit's textbox to the App's state "title" property (because you'll probably want to bind a lot of textboxes, so it's easier to separate the code into classes), and then every time the user types something into the textbox, instead of using the toolkit's code that is already written for you which updates the textbox' state directly, you block the state change in an event handler, send the state change to the App class, let the App class figure out the differences between the current state and the new state, and then it calls some callback in the binding class that is responsible for actually changing the text in the textbox to reflect the new App state. You'll probably run into a ton of issues (selections resetting, assistive technologies bugging, etc.) that you'll have to deal with externally somehow. All just to make it do exactly the same thing it would have done anyway.

It's like you have a fully autonomous robot and you want to add marionette strings to make it move.

[nogridbag]

I believe every UI developer that has used frameworks like Swing has reached a point where specific user interfaces, even those that look trivial, become too complex primarily due to things like event handlers. Trying to figure out a simple thing like why a radio box is enabled and is marked dirty may require long debugging sessions where one event handler for component A triggers another event handler for component B, which triggers another event handler, etc. And before you know it, 50 events were triggered just during the initial mounting of the UI. Making sense of it all is maddening.

And then said developer does what they think feels right: "I have my state and I simply want all the fields to reflect my state". So they try and rewrite the component into some sort of mini immediate-mode style component. All of the event handlers get wired up a single "layoutComponent" method that tries to call all the UI component setters based on the current state with all the problems you alluded to. I know I've done this type of things numerous times before going back all the way to my first internship, well before React was even a thing.

I think modern frameworks solve the reactivity issue this well enough, that it really doesn't matter if the underlying framework is not natively reactive. I will say though that I've primarily used Vue.js where most state is relatively local and we don't try to re-render the world on every input. I think part of the problem with modern dev is likely that React has become the default instead of a simpler framework like Vue.

[AlienRobot]

That's true. Events are the WORST thing about GUI programming. They're so convenient and so undebuggable it almost feels like a trap.

Ironically, in most cases events are only used by one object, but you always want to consider the possibility that two objects will want to observe the same event, so now you need an entire event dispatching class, and then you'll want observable properties, and the nail on the coffin is going to be observable lists. When you reach that point, one event triggers another, which changes a property, triggering another event, and so on and so on. You are 5 layers deep into event callbacks. The call tree just has the same "callCallbacks()" method over and over again.

Bugs start happening because of the order in which callbacks are called becomes important, so now you need a way to give some callbacks priority over others, or make them happen after all normal callbacks were called. One callback destroys an object which has callback on the event that destroyed it, so you're going to need a wrapper around your callbacks that gets notified when callback's object is destroyed to change its reference to null in order to avoid executing code on the destroyed object if this happens while iterating the callbacks in the event dispatcher. Sometimes calling callbacks in wrong order is a performance hit, when it doesn't just get stuck into an infinite loop and you run out of stack.

I wonder if there is GUI programming paradigm that solves all of this or that you can call the "best" one. Maybe it's reactivity, maybe not. Who knows.