In the "Core Data Cloudkit Public Database" session, it is mentioned that you will have a "complete local mirror" of all of the CloudKit data. Does this mean that all of the data in the public database will be taking up space on every user's device, even if they do not use the data? I have an app that allows users to create entries of text that they want to memorize. I use the persistent Core Data container to sync the users data between devices and this works extremely well. If I used Core Data shared persistent containers to allow publishing of texts, would there be a way for users to only download specific data? for example, I would like to allow students to download a group of texts published by their teacher

When making a Multiplatform app with SwiftUI (supporting both iOS and macOS), is there a particular reason to use multiple targets over Catalyst (or vice versa)? From what I understand, one big difference is that if you need an "escape hatch" to do something SwiftUI doesn't support, you will only be able to do UIKit if you follow the Catalyst approach. This makes me think Catalyst would be a better fit for an iOS developer like me. In addition, you can access AppKit from Catalyst, which seems to make that approach more flexible (If I need to do something only available in AppKit). On the other hand, creating a new cross-platform project in Xcode creates two targets which makes me wonder if that's the officially recommended approach. Maybe because this would allow more flexibility if you want different build settings for each platform? Another, related question: If I use Catalyst for 1.0 of an app, and release an app to the App Store, can I switch to the multiple-target approach later?