alt Hacker NewsI agree with you. Swift should have been a modernized Objective-C. It should have kept the best parts of it, which made it a joy to use, and leave the archaic and the weird things behind.
it would have been a great language, meanwhile they decided to throw everything out and created a language that it is overcomplex, and it is failing to gain any traction outside iOS / Apple's ecosystem. Basically, if you weren't required to use it, nobody would, which makes it a failure.
SwiftUI, should have been a rendering template/library integrated with UIKit. Also, cleaning up some of the UIKit syntanx and framework (simplifying it), and use SwiftUI as a template for UI, would have been ideal.
SwiftUI instead was pitched as replacement, yet it is not near as capable as UIKit, and it is not feature parity even 7 years later.
Just bad leadership by Apple in this case. (probably many of these decissions were 'promo / resume padding driven' as it happens in many large companies).
Replies
> it would have been a great language, meanwhile they decided to throw everything out and created a language that it is overcomplex, and it is failing to gain any traction outside iOS / Apple's ecosystem. Basically, if you weren't required to use it, nobody would, which makes it a failure.
Which means Objective-C was a failure, right?
I wrote something similar a couple years ago too after having a pretty bad time with swift ui and concurrency: https://medium.com/goodones/pareto-optimal-apple-devtools-b4...
> Swift should have been a modernized Objective-C. It should have kept the best parts of it, which made it a joy to use, and leave the archaic and the weird things behind.
That's what Objective-Smalltalk is, or rather, what it started out as.
https://blog.metaobject.com/2019/12/the-4-stages-of-objectiv...
It has now gone further to actually implement Brad Cox's idea of a "Software IC", by taking on ideas from software architecture and metaobject protocols.
https://2024.splashcon.org/details/splash-2024-Onward-papers...
> SwiftUI, should have been a rendering template/library integrated with UIKit.
Do you mean it should have used some sort of XML-ish/HTML-ish templating mechanism? For Interscript, I keep looking at that approach, but so far haven't gone down that road (except for HTMXNative, but there it actually is HTML, so...).
> and leave the archaic and the weird things behind.
The number one archiac/weird thing was message passing instead of methods. Number two was square brackets for message passing.
What would be left of Objective-C once these were left behind?
> and it is failing to gain any traction outside iOS / Apple's ecosystem.
Objective-C was on a path to gain serious traction outside of the Apple ecosystem?
So just like Objective-C?
Rescued by NeXT's acquisition, having failed to gain market adoption otherwise.
While reading this comment, two of Steve Job's line pops up in my head. I think it was the Apple Park opening, and Steve response to insult in WWDC 199x.
> "We need to be true to ourselves, and remember what is important, that is what going to keep Apple Apple, is if we keep us, us."
> "You have got to start with the customers experience and work backwards to the technology.
Right now it feels a lot like Apple without Steve Jobs the first time round. On paper there are a lot of cool tech being worked on. But in the end it was tech from NeXT that really shines.
And a lot of these tech don't actually benefits the customers. If the choice was a language and framework that is slightly harder to developers but brings overall higher quality Apps because the barrier is higher. Compared to a language that wants to replace Assembly to Javascript while being easy to learn like Visual Basic and for everyone to code. I would much rather pick the former.
Edit: I suddenly remember I also submitted a Core Animation introduction video from Steve a while ago on HN [1]. 20 years later we have gone backwards on a lot of things.
[1] https://x.com/lucaslovexoxo/status/2082098579632783441