[bardi]

You’re right, you shouldn’t have to but it seems to be a penalty you currently have to pay if you go all-in with SwiftUI at this stage in its lifecycle. Admittedly, it’s still relatively/technically immature compared to its predecessor, and so you expect certain limitations in comparison. But four versions in, some of these things I’m surprised continue to be points of frustration. I imagine if Apple themselves were stress testing/dog fooding it exclusively then these are things that would have been ironed out by now. The irony, right?

And you’re right, SwiftUI “onboarding” is almost deceitfully simple/natural. And, in general, there being no friction in its UX makes it so easy to gradually build up issues in complete ignorance. And when it does become apparent, you’re working with a gorilla, i.e. steps to rectify are expensive.

I look forward to trying out Compose for all the reasons stated. Thanks for sharing !

[bardi]

The one other thing I’d add as being a major source of frustration and anxiety is SwiftUI’s update cycle. It’s acceptable for there to be issues in any technology, whatever its age, but it’s not satisfactory to need to wait 12 months for the possibility — not even certainty — of a fix. This is what attracts me most about Compose: the transparency and speed of innovation.