>If you are wondering how this library is different from other GUI toolkits, check out the FAQ
I see no details at all about how this library is different [or not]. While not an UI expert, I have some low-level experience with gtk, qt, appkit and uikit, vcl, immediate ui, and full manual drawing / event processing. When considering new ui framework, these questions are of high priority: how does it integrate with, or supports for, existing custom run loop; what layout system is used and which problems it elegantly solves (i18n, baselining, excessive boxing, complex constraint cases); how does it implement repeating cell rendering and caching, if any; what are base classes and how metamechanics like scrolling/resize/animation/state cues are distributed across these; how does ui state interact with controlling entity and vice versa; which non-ui mid-controllers exist to reduce or abstract out basic boilerplating. Without these details, it is yet another "throw edits, buttons and labels to fixed form and call it a day framework" at most. UI is usually hard because of internals and extension, not because it lacks fixed functionality. Web is partly so popular because it feels like everything is possible from the styled box primitive (but has its obvious downsides; like a 100mb complex c++ rendering engine that only few groups in the world can implement correctly and somewhat effectively on latest i7). Without all this and many more, there is nothing to reason about — ui is not a yesterday's demand, and regular cases are covered by virtually any mature library. Drawing static hearts with beziers is not a big deal.
Tbh, I'm highly doubting any of listed topics can be designed in just one version step, because even most decent frameworks have so many hidden flaws that require insane efforts to overcome under specific requirements.
It would be great to see how it is actually better, e.g. by comparing side-by-side examples of solving normal-to-hard ui tasks in different libraries.