They already have a Swift LSP, but I think expecting iOS developers to switch over to VS Code is probably going a bit far. By the time you build things like code signing, SwiftUI previews, build settings, etc. into it, you’re basically building a new IDE on top of VS Code rather than just supporting VS Code itself.

Don’t get me wrong, it’s possible. It’s just what you’d get at the end wouldn’t really resemble VS Code. Even though I spend a lot of time writing web app code in VS Code, my hope – wearing my iOS developer hat – is that they have a skunkworks project to replace Xcode with a new native IDE, not that they embrace VS Code.

I was more suggesting Apple follow where developers are going and not try to lead where few want to follow. And if developers want VSCode, so be it.

I remember when the Cocos framework (a 2D and later 3D gaming engine) was found to be a popular binary that iOS apps were linking against, Apple suddenly saw fit to push their own game engine. I am not sure how many developers came aboard though. Cocos was, shhh!, cross platform and that was likely a problem from Apple's point of view; a boon from the developer's point of view.

I don’t hear many iOS developers clamouring for VS Code. They mostly just want Xcode fixed. The only mention of it in those two articles is somebody wanting “AI goodies we see with IDE’s like VS Code and Cursor make its way (natively) into Xcode.” But there are many wanting a better Xcode.

I’m not sure you can point to Cocos as a motivating factor for Apple, and it certainly wasn’t sudden. It got popular very early on and plateaued. I remember building apps with it for iPhoneOS. SpriteKit didn’t come along until years later.

I'm not in the loop then if you say iOS devs just want Xcode fixed. You could be right. You would have to weigh that against devs that are not even testing the iOS waters because they see Xcode as a barrier, are more familiar with their VSCode Cmake ecosystem.

A new framework like SpriteKit requires at least a year (an iOS release cycle) before it sees WWDC ... and then a shipping iOS. My recollection is that this is pretty close to the timeline I described where Cocos starts to show up in abundance in the iOS apps submitted and as a result a team is spun up at Apple to create what ends up being SpriteKit.

You'll have to educate me on the plateauing of Cocos. If it did plateau pre Sprite-Kit, what replaced it?

> A new framework like SpriteKit requires at least a year (an iOS release cycle) before it sees WWDC ... and then a shipping iOS. My recollection is that this is pretty close to the timeline I described

It’s not. Cocos got popular in the iPhoneOS 2.x days. SpriteKit came along in iOS 7. We’re talking about a period of five years, not one or two.

> If it did plateau pre Sprite-Kit, what replaced it?

Plateau means it stayed flat, not that it went away. Nothing replaced it. It just didn’t keep getting more and more popular.