Overall 6.3 ended up mostly about integration: stdlib, c/c++ (often driven by swift-java interop needs), and builds.
SPM (swift package manager) is slowly reaching out to take over xcode build features (e.g., resources most recently), and they have been inching and hacking their way to a new swift-build engine (now experimental) and "prebuilt" modules for the compiler/IDE macro support. Things Go Wrong all the time when trying to replace parts while driving, and there's very little visibility or control over the interactions between SPM and Xcode.
Conversely to integrations, endogenous language progress seems relatively quiet because there's a lot of behind-the-scenes work extending the initial blush of lifetime controls to iterators, stdlib types, etc. (never mind parameter packs). They're in year 4+ of building out the function-coloring of concurrency and isolation domains (6.2 had an "easy" mode that caused lots of problems), so adding year 2+ dataflow coloring of lifetimes is doubly complicating.
Naturally, diverse support for exciting new features results in quite a spread in the third-party library ecosystem of support for various language features. And now with AI entrepreneurs rewriting and porting any project that can score hits and kudos, the usual open-source authorities are losing their steam.
Finally, there's the baseline platform complexity of multiple OS and devices, physical and simulator debugging support, cloud CI...
So: that makes four dimensions of complexity. As a Swift developer you'd have to be very careful to stay on a working path as things shift underneath you.
good to see incredible stuff being shipped in Swift. Haven't used it since v3 though.
around 2015-17 - Swift could have easily dethroned Python.
it was simple enough - very fast - could plug into the C/C++ ecosystem. Hence all the numeric stuff people were doing in Python powered by C++ libraries could've been done with Swift.
the server ecosystem was starting to come to life, even supported by IBM.
I think the letdown was on the Apple side - they didn't bring in the community fast enough whether on marketing, or messaging - unfortunately Swift has remained largely an Apple ecosystem thing - with complexity now chasing C++.
> the server ecosystem was starting to come to life, even supported by IBM.
I was in college at the time and doing some odd freelance jobs to make some money. Unbeknownst to my clients I was writing their website backends in swift, using build packs on heroku to get them hosted.
It was a fun time for me and I love swift but I will admit last year I went ahead and rewrote an entire one of those sites in good ol typescript. I love swift but anything outside of the Apple ecosystem with it just seems like it hasn’t hit critical mass yet.
Even today, with the fancy Swift 6.3, the experience of using Swift for anything other than apps for Apple platforms is very painful. There is also the question of trust - I don't think anyone would voluntarily introduce Apple "The Gatekeeper" in parts of their stack unless they're forced to do it.
What, of course Apple uses Swift on the server, that's the only reason they're investing in any of this. Many of the foundational Swift on the server libraries were written at Apple and later opened, like SwiftNIO.
Exactly true - they've created all these "working groups" of open source / volunteers to care for Android / Server / Wasm / ... all while being constraint "as an Apple product". Of course the end result is crappy
I'm sorry, that's absolutely bullshit. In fact, I wish we had left everyone who complained behind—the python community would have been happier and healthier for it. Absolute crybabies who wanted to be catered to without caring for how intractable the problems with python2 were—e.g. dealing with unicode was a royal pain in the ass, and the bytes/string divide completely fixed it. IMO, it was the best-executed breaking change I've ever witnessed in a language.
In comparison, e.g. Scala 2 -> Scala 3 was an absolute nightmare—it just didn't have the same vocal wailing from maintainers in the community (or, I suppose, a fraction of Python's popularity to begin with).
Being to aggressive in breaking stuff gets you a shitshow like Node.js or Ruby. Long-term source code compatibility is a very useful feature for open source and a sign of a mature eco system. Feel free to add stuff, but once it's part of a stable release it has to be maintained long after a "better" way to do it comes along.
nodejs itself doesn't have very many breakages; i have plenty of code that is unchanged from 0.12 to 24. npm is a whole other kettle of fish but I don't think you can blame the core project for the sins of everyone that publishes to the package manager. Python2 -> Python3 on the other hand had a lot of breakage in "standard" code.
I was enthusiastic about early TensorFlow in Swift efforts, sorry when the effort ended. My interest then flowed into early Mojo development for a while.
I wrote an eBook on Swift several ago but rarely update that book anymore. Count me as one of the many developers who for a while thought Swift would take over the world. At least Swift is a fun language to use, and now with LLM coding tools writing macOS/iOS/iPadOS apps is fairly easy.
funnily enough, I talked recently to someone working on the swift compiler (not an Apple employee) to make Swift functions differentiable. So its not all dead yet
Python's interactive interpreter makes it pretty useful as a shell, for iterative development, and crucially useful in a Jupyter notebook. I've also found CircuitPython's interpreter to be bonkers useful in prototyping embedded projects. (This, on top of the nice datascience, ML, and NN libraries).
Swift just wasn't doing the same things. And even if it did, Swift would compete with other languages that were understood as "a better Python", like Julia. Even then, Swift only came to Linux in 2016, Windows in 2020, and FreeBSD less than a year ago with WWDC 2025.
I think it doesn't help that the mid 2010s saw a burst of Cool and New languages announced or go mainstream. Go, Julia, Rust, TypeScript, Solidity, etc. along with Swift. I think most of us only have space to pick up one or two of these cool-and-new languages every few years.
> could plug into the C/C++ ecosystem. Hence all the numeric stuff people were doing in Python powered by C++ libraries could've been done with Swift.
In 2015-2017 you could interop with C, C++ support wasn't added until very recently.
I do agree with you though and I am not sure what the exact reasoning is, but Swift is definitely an Apple ecosystem language despite the random efforts to gain traction elsewhere.
> around 2015-17 - Swift could have easily dethroned Python.
Why could it?
> it was simple enough - very fast - could plug into the C/C++ ecosystem. Hence all the numeric stuff people were doing in Python powered by C++ libraries could've been done with Swift.
Half a dozen languages fit this description.
> the server ecosystem was starting to come to life, even supported by IBM.
No, not at all. Kitura, Vapor (a fitting name) were just a toys that no serious player ever touched.
After that, and IBM losing interest, Apple did hire a few competent people (including contributors to Netty and Akka) to build the Swift Server Workgroup.
But I don't know why I'd pick Swift on the server when Rust is better in almost every dimension, with a thriving and more community-driven ecosystem.
I think it's not about that but about dogfooding Swift on the server. Apple uses Go, Java etc for a lot of its server components and refused to invest in hiring people that would extend the ecosystem for server Swift.
It certainly doesn't help, but among big tech, Apple is not the only company where teams are siloed and independent. Microsoft has people writing Java or Go instead of C# too.
I assume the server side usage is not zero, but not enough to reach a critical mass, you're probably right there.
It remains to be seen how much Mojo has learnt from that effort.
NVidia, AMD and Intel now have doubled now into giving Python GPU JITs, and Julia, the same capabilities as their CUDA, ROCm, and SYSCL offerings with C++.
With Julia and Python having their 1.0 long behind them.
Lattner probably left because Apple didn't give the team any breathing room to properly implement the language. It was "we must have this feature yesterday". A lot of Swift is the equivalent of Javascrip's "we have 10 days to implement and ship it":
Swift has turned into a gigantic super complicated bag of special cases, special syntax, special stuff...
We had a ton of users, it had a ton of iternal technical debt... the whole team was behind, and instead of fixing the core, what the team did is they started adding all these special cases.
For this language to become default at Apple they had to be doing a massive amount of internal promotion - in other words they knew where it was going.
And then if that's the case, how were they not ready to solve the many problems that a big organization would run into? And all the schedule constraints that come with it?
> Swift has turned into a gigantic super complicated bag of special cases, special syntax, special stuff...
That's true, but only partly true. It already was a gigantic super complicated bag of special cases right from the start.
Rob Rix noted the following 10 years ago:
Swift is a crescendo of special cases stopping just short of the general; the result is complexity in the semantics, complexity in the behaviour (i.e. bugs), and complexity in use (i.e. workarounds).
Apple's new Swift language has taken a page from the C++ and Java playbooks and made initialization a special case. Well, lots of special cases actually. The Swift book has 30 pages on initialization, and they aren't just illustration and explanation, they are dense with rules and special cases
Of course, that doesn't mean that it didn't get worse. It got lot worse. For example (me again, 2020):
I was really surprised to learn that Swift recently adopted Smalltalk keyword syntax ... Of course, Swift wouldn't be Swift if this weren't a special case of a special case, specifically the case of multiple trailing closures, which is a special case of trailing closures, which are weird and special-casey enough by themselves.
A prediction I made was that these rules, despite or more likely because of their complexity, would not be sufficient. And that turned out to be correct, as predicted, people turned to workarounds, just like they did with C++ and Java constructors.
So it is true that it is now bad and that it has gotten worse. It's just not the case that it was ever simple to start with. And the further explosion of complexity was not some accidental thing that happened to what was otherwise a good beginning. That very explosion was already pretty much predetermined in the language as it existed from inception and in the values that were visible.
From my exchange with Chris regarding initializers:
"Chris Lattner said...
Marcel, I totally agree with your simplicity goal, but this isn't practical unless you are willing to sacrifice non-default initializable types (e.g. non-nullable pointers) or memory safety."
Part of my response:
"Let me turn it around: Chris, I totally agree with your goal of initializable types, but it is just not practical unless you are willing to sacrifice simplicity, parsimony and power (and ignore the fact that it doesn't actually work)."
Simplicity is not the easy option. Simplicity is hard. Swift took the easy route.
[...] when you first attack a problem it seems really simple because you don't understand it. Then when you start to really understand it, you come up with these very complicated solutions because it's really hairy. Most people stop there. But a few people keep burning the midnight oil and finally understand the underlying principles of the problem and come up with an elegantly simple solution for it. But very few people go the distance to get there.
-- Steve Jobs (borrowed and adapted from Heinelein)
> Even if the languages started tiny, complexity eventually grows on them.
And then of course the case that proves the opposite, Clojure. Sure, new ideas appear, but core language is more or less unchanged since introduced, rock solid and decades old projects still run just fine, although usually a bit faster.
That's a pretty far cry from "complexity is unavoidable". Reading that to me implies that the complexity is inherent in programming language design, whereas this follow-up argument seems to say that complexity is the result of tacking on new features.
The latter is a bit tautological, since the size of the language grammar is itself a measure of complexity.
The current market share shows how far you can go with just being a better Java.
If (or when? I haven't checked recently) a decent and well-thought-out LLVM backend emerges for it, ideally with some new underlying complexity seeping through, the market share might expand overnight.
And as for C++, while some complexity is certainly unavoidable, a rigorous complexity control is desperately needed. Ideally, the same way Bell Labs folks did when they initially conceived Go from Algol68 and C and similar (before or after joining Google; I couldn't tell), and Rich Hickey did when he initially designed Clojure. Some people are managing the complexity using style guides and clang-tidy checks. Which is great in that doing so doesn't need lengthy language committee decisions. But that approach hasn't been enough to make code _sufficiently_ safe; every now and then an enterprising engineer or team finds a way to abuse a feature in a way that produces unsafe or unpredictable results. Rust is a bit better and solves a few of the common problems, but sadly the list of potential issues (of using Rust in a codebase at scale; Engineers' faults, not Rust's) is long and growing. My verdict is we need both complex and simple LLVM languages, ideally co-designed to have no interop problems by design, while allowing expressing some logic in the simple parts and some logic in the complex parts. Or better, a 3 tier design would be nearly perfect: expressive config language, glue and research language, and core building blocks language. I think a clojure-style language can be designed to achieve all three.
Yes, that's one approach to avoiding ever growing complexity, maybe the other languages should try it sometime ;)
With that said, everything around Clojure keeps improving and getting better. While the language doesn't have static types, clojure.spec offers something that is even better than static typing (imo), and doesn't even require any changes to the core language. Something else other mainstream languages could learn too.
Just IMO, but... no. To me a "could have easily" requires n-1 things to have happened, and 1 thing not happening. Like, we "could have easily" had a nuclear exchange with the USSR, were it not for the ONE Russian guy who decided to wait for more evidence. https://en.wikipedia.org/wiki/1983_Soviet_nuclear_false_alar...
But even in '15-'17, there were too many people doing too many things with Python (the big shift to data orientation started in the mid/late 90's which paved the way to ML and massive python usage) by then.
The 'n' was large, and not nearly of the 'n' things were in Swift's favor then.
Swift was feeling pretty exciting around ~v3. It was small and easy to learn, felt modern, and had solid interop with ObjC/C++.
...but then absolutely exploded in complexity. New features and syntax thrown in make it feel like C++. 10 ways of doing the same thing. I wish they'd kept the language simple and lean, and wrapped additional complexity as optional packages. It just feels like such a small amount of what the Swift language does actually needs to be part of the language.
I get this feeling with C#. I have been here since its release. I looked at Swift and then they moved very quickly at the beginning, so the book I had to teach me was out of date moments after it was printed. With all the complexity being thrown in, I stuck with C++ because at least it was only 1 language I had to keep track of (barely)!
I've been using C# since the first release in 2003/4 timeline?
Aside from a few high profile language features like LINQ, generics, `async/await`, the syntax has grown, but the key additions have made the language simpler to use and more terse. Tuples and destructuring for example. Spread operators for collections. Switch expressions and pattern matching. These are mostly syntactic affordances.
You don't have to use any of them; you can write C# exactly as you wrote it in 2003...if you want to. But I'm not sure why one would forgo the improved terseness of modern C#.
Next big language addition will be discriminated unions and even that is really "opt-in" if you want to use it.
The C# impl is still early and I think what will end up happening is that a lot of the boilerplate will end up being owned by source generators in the long term. C# team has a habit of "make it work, make it better". Whatever v1 gets released is some base capability that v2+ will end up making more terse. I'm happy and OK with that; I'd rather have ugly unions than no unions (yes, I already use OneOf)
Ah Source Generators, after all these years still badly documented, when searching you most likely will find the original implemenation meanwhile deprecated, have poor tooling with string concatenation, and only have a few great blog posts from .NET MVPs to rely on.
:shrug: we're using them very effectively and there are plenty of resources at this point.
Very useful for reducing boilerplate and we can do some interesting things with it. One use case: we generate strongly typed "LLM command" classes from prompt strings.
I would remove result builders and all other uses of @attributes that change the semantics of the code (e.g property wrappers).
I would remove the distinction between value types and reference types at the type level. This has caused so many bugs in my code. This distinction should be made where the types are used not where they are defined.
I would remove everything related to concurrency from the language itself. The idea to let code execute on random threads without any explicit hint at the call site is ridiculous. It's far too complicated and error prone, which is why Swift designers had to radically change the defaults between Swift 6.0 and 6.2 and it's still a mess.
I would remove properties that are really functions (and of course property wrappers). I want to see at the call site whether I'm calling a function or accessing a variable.
I would probably remove async/await as well, but this is a broader debate beyond Swift.
And yes you absolutely do have to know and use all features that a language has, especially if it's a corporate language where features are introduced in order to support platform APIs.
I agree with you about result builders, silly feature that only exists for SwiftUI.
But a lot of what you said, except for the concurrency and property wrapper stuff, largely exists for Obj-C interop. The generated interface is more readable, and swift structs act like const C structs. It’s nice.
I'm not a Swift user, but I can tell you from C++ experience that this logic doesn't mitigate a complex programming language.
* If you're in a team (or reading code in a third-party repo) then you need to know whatever features are used in that code, even if they're not in "your" subset of the language.
* Different codebases using different subsets of the language can feel quite different, which is annoying even if you know all the features used in them.
* Even if you're writing code entirely on your own, you still end up needing to learn about more language features than you need to for your code in order that you can make an informed decision about what goes in "your" subset.
The absence of guard in Kotlin is one of those things that regularly trips me up when bouncing between it and Swift. Rather than Swift losing guard I’d prefer if Kotlin gained it.
Focusing on the keywords rather than the macros, I think the rest of them have legitimate use cases, though they're often misused, especially fileprivate.
You can take this approach in personal projects - with teams you need to decide on this and then on-board people into your use of the language. This does not work.
That number is unfairly exaggerated. The list includes ~40 internal keywords used only by language developers, plus dozens of tokens that would be called preprocessor directives, attributes, or annotations in other languages (e.g. `canImport` as in `#if canImport(...) #endif`; `available` and `deprecated` as in `@available(*, deprecated) func`).
Ah makes sense, personally I wouldn't consider reserved but unused words as keywords in the sense that you don't need to know them to read the language (even though they're keywords in some other technical sense). I was curious because I just tried counting number of keywords by language and it seemed surprisingly ambiguous/subjective/up to the language to say what's a "keyword" vs some type of core module. So my attempt (https://correctarity.com/keywords) probably has mistakes...
>"around 2015-17 - Swift could have easily dethroned Python."
NumPy, SciPy, Pandas, and Pytorch are what drove the mass adoption of Python over the last few years. No language feature could touch those libraries. I now know how the C++/Java people felt when JS started taking over. It's a nightmare to watch a joke language (literally; Python being named for Monty Python) become the default simply because of platform limitations.
Eh, I don't think Swift would ever have dethroned Python. What pain point would it practically solve? I don't use Python often but I don't hear folks complaining about it much.
I do, though, think Swift had/has(?) a chance to dethrone Rust in the non-garbage collected space. Rust is incredibly powerful but sometimes you don't really need that complexity, you just need something that can compile cross-platform and maintain great performance. Before now I've written Rust projects that heavily use Rc<> just so I don't have to spend forever thinking about lifetimes, when I do that I think "I wish I could just use Swift for this" sometimes.
You're right, though, that Swift remains Apple's language and they don't have a lot of interest in non-Apple uses of it (e.g. Swift SDK for Android was only released late last year). They're much happier to bend the language in weird ways to create things like SwiftUI.
No way something that compiles as slowly as Swift dethrones Python.
Edit: Plus Swift goes directly against the Zen of Python
> Explicit is better than implicit.
> Namespaces are one honking great idea -- let's do more of those!
coupled with shitty LSP support (even to this day) makes code even harder to understand than when you `import *` in Python.
Edit 2: To expand a little on how shitty the LSP support is for those who don't work with Swift: any trivial iOS or macOS project that builds fine in Xcode can have a bunch of SourceKit-LSP (the official Swift LSP) errors because it fails to resolve frameworks/libraries. The only sane way to work with Swift in VS Code or derivatives I've found is to turn off SourceKit diagnostics altogether and only keep swiftc diagnostics. And I have the swift-lsp plugin in Claude Code, there's a routine baseline of SourceKit errors ignored. So you have symbols without explicit namespaces, and the LSP simply can't resolve lots of them, so no lookup for you. Good luck.
That's funny. To me magic is implicit by definition and Python strikes me as a very magical language compared to something like Java that is way more explicit.
The same thing you'd do yoursef if you wanted to assign it to a namesake local variable even if it was in a dict to begin with: you'd make the dash an underscore.
It would be extremely unlikely that you would replicate the name as a local variable if it was in a dict, but regardless a dict doesn't have that limitation. The namespace thing is atrocious and bad design -- no straightforward way to iterate over them, merging/updating them is awful, collides with keyword methods (keys, items, etc.), and so on; thankfully more modern argument parsing libraries didn't repeat this mistake. It's just a shame this ended up in the standard library, but then Python standard library has never really been any good, e.g. logging and urllib1234567.
I spent last week (with Opus, of course) porting the xv6-riscv teaching operating system to a bunch of different languages. Zig, Nim, LISP, and Swift.
The improvements in embedded Swift have definitely made it one of the most enjoyable/productive languages to work on the OS. I feel like I can build useful abstractions that wrap raw memory access and make the userland code feel very neat.
On the other hand, the compilation times are SO bad, that I'm really focusing on the Nim port anyway.
It's been a long time since I came across Nim. I thought it was really interesting about 12 years ago. What made you land on Nim instead of any of the more obvious alternatives?
I was looking for something that allows easy access to direct memory, with a syntax thats a little easier to explain than C. Frankly, zig was not actually a real viable option based on that syntax requirement but I still wanted to explore it.
Nim is really incredible. The only things I cannot get over is the fact that it goes the inheritance route in a way I find to be hacky and fragile (no more than one level, really?) and traits are not a core feature. If Nim's primary approach was composition + Rust-style traits (that work at compiletime and runtime), I'd have a hard time wanting to use anything else.
Yeah, for a language that claims to be a better modern alternative to C, zig verbose syntax is really an eyesore to look at compared the very same codebase written in C...
I _can_ do the same with Rust, doesn't mean it's "the language I reach for" for making e.g. a website. Because the tooling, ergonomics, hireability factor, etc. are still very harshly against it.
Same with Swift, but I'd call that more of a wasted opportunity because Apple, unlike Rust Foundation, has a mountain of money to make it happen, and yet they don't seem to care.
I don’t believe that’s true. Things are moving constantly, and in the right direction. Then again it would help if you cited particular grievances, because being a regular (cross-platform/cross-target) Swift user I am not sure what you are talking about…
I did not choose ClearSurgery’s example randomly. I was at a conference recently where the CTO was here, and he explicitly told us they were moving fast thanks to the Swift ecosystem. (I am not working there personally, nor am I affiliated.)
if they really want me to use this lang for everything, they'd have to 1. massively improve compilation speed, 2. get the ecosystem going (what's the correct way to spin up an http server like with express?) and 3. get rid of roughly 150 of the 200 keywords there are
especially w.r.t. the last one, of course everyone frets at huge breaking changes like this, so it won't happen, so people won't use it
> 3. get rid of roughly 150 of the 200 keywords there are
I don't understand this point. Could you explain?
The new keywords enable new language features (ex: async/await, any, actor), and these features are opt-in. If you don't want to use them, you don't have to.
What are they keywords you think should be removed?
> these features are opt-in. If you don't want to use them, you don't have to.
Using a language is more than just writing it with a pre-established knowledge of what subset of features you think is worth the tradeoffs. More keywords/features means when you try to figure out how to do something new, there may be 15 different ways and you need to analyze and figure out which is the best one for this scenario, which ones are nonstarters, etc.
That's was more or less the whole design goal of Go. It was made by C++ programmers who were fed up with how many features were in the language, so they kept the feature set limited. Even the formatting is decided by the language. You may not agree with every decision, but what matters is decisions were made and they're standardized, so everyone is on the same page. You can read anyone else's code, and you know exactly what's going on.
Not just uber slow to compile, because as a Rust dev I could take that. But it rejects correct programs without telling you why! The compiler will just time out and ask you to refactor so it has a better shot. I understand that kind of pathological behavior is present in many compilers but I hit it way too often in Swift on seemingly benign code.
I don't know why anyone would want to use Apple tools if they are not developing for Apple platforms. Apple barely maintains compatibility for their own platforms, using Swift on a non-Apple platform is setting yourself up for doubule pain.
Why are you interpreting this comment as "never receives updates"? It takes great effort to maintain API compatibility, some things aren't improved or are implicitly deprecated.
In a way it still is true. Swift works on Windows and Linux until it doesn't. It's taken until a couple years ago for other build systems to get swift support (which I suppose is the fault of said build system, but Swift taking so long to be cross-platform contributed to that), and even now it (still) doesn't quite work right. C interop is a mess requiring hacks to generate clang modules to actually get Swift to see them (and CMake for example provides no easy way of doing this, or last time I checked it didn't). Oh and Swift tends to take over the linker and compilation pipelines when you enable it, at least with CMake, because... Reasons? I honestly don't know why. It causes very weird errors when I integrated Swift code into my C++ project that were a pain to actually diagnose. I eventually got it working, but still, it wasn't simple or seamless.
No mention of compilation speed improvements? Very unfortunate. Compilation times slower than rust really hampers the devx of this otherwise decent language.
I tried Swift a few months ago in a project that made use of a bunch of bigger dependencies and I was instantly shocked by the compilation times. It's quite unimaginable to me using Swift for everyday work because of that. Especially when coming from the fast compile times of Go. But it's really unfortunate because I really enjoyed writing Swift because it feels like a very well made language. But iterating on some code and trying to get quick feedback is pure pain.
> Swift 6.3 introduces the @c attribute, which lets you expose Swift functions and enums to C code in your project. Annotating a function or enum with @c prompts Swift to include a corresponding declaration in the generated C header that you can include in your C/C++ files
Why did this take so long to be added? Such strange priorities. Adding an entire C++ compiler for C++ interoperability before adding... C exports. Bizarre.
C++ interop got attention because it helps Apple absorb low-level codebases that already moved past pure C. Exporting Swift to plain C mostly means more DIY FFI spaghetti.
Once enums, ownership rules, and nullability cross that boundary, the generated header stops looking like a neat bridge and starts looking like one more place for ABI bugs to hide. Closures make it weirder fast, because now your error handling and calling conventions can drift just enough to produce the kind of bug that wastes a whole afernoon.
I remember building dylibs in Swift for use in C programs, had to use @cdecl annotation iirc to achieve that, which was experimental. Good to see it's finally official
It would be really nice if instead we could just do one style of development and then ship a set of libraries as used to work for OpenSTEP (which was why it had "OPEN" in the name).
Interestingly, Kotlin has a pretty solid cross-platform story.
I'd pick it over Swift if targeting Android since it can build and run in the JVM as well as natively -- and has Swift/ObjC interop. Its also very usable on the server if you wanted to, since you can use it in place of Java and tap into the very mature JVM ecosystem. If that's what you're into.
And I have a lot more faith in JetBrains being good stewards of the language rather than Apple, who have a weird collection of priorities.
Kotlin is practically a no-brainer when you have JVM at your finger tips, versus something like Swift which is comparatively young.
I tried to use Vapor with Swift recently and struggled to get something working because the documentation looked comprehensive, but had a lot of gaps. I ended up throwing it out because I didn't have the time to dig through the source to understand how to do something, when I could use a mature framework in any other language instead.
The promise is there but I'm just not ready to invest. My youthful days of unbounded curiosity are coming to an end and these days I just want to get something done without much faff.
Mind you, Kotlin/Native (which is what gets used when you're compiling for iOS) doesn't have access to the JVM.
However, the Kotlin community is fundamentally all about open source, whereas Apple & iOS Devs have an allergy to it. The quality and quantity is already miles above the vast majority of what's in the Swift ecosystem. https://klibs.io has all the native compatible libs. And if you're targeting a platform where the JVM is available then yeah, it's massive. Compose makes UI tolerable compared to JWT too. Even large projects like Spring are Kotlin first nowadays.
JetBrains has monetary interest in promoting Kotlin beyond Android, there’s zero incentive to promote Swift as the language outside of iOS and Mac. They don’t need to capture minds of devs for them to develop for Apple devices.
The language doesn’t really matter. The underlying SDK/framework is where the action is at.
However, I suspect that we may not be too far off, from LLMs being the true cross-platform system. You feed the same requirements, with different targets, and it generates full native apps.
Fully agree. I have zero Swift knowledge and currently use LLM to write a native app. I'm well aware of the SDKs and concepts in iOS development, so even if something's wrong I got intuition where to look and how to make the LLM fix it.
This is going to be used much more than Swift for servers. Swift is a primarily client-side mobile language. It makes sense that you tap into reusing the logic.
The noncopyable types improvements are the most underrated part of this release. Finally makes it practical to model unique ownership in Swift without fighting the compiler.
How is the toolchain? Does Swift Lint and Swift Format support the newest version. Honestly, the modern program language should have the built-in formatter and recommend lint rules. It is not just shipping a program language, it is a while ecosystem.
i worked with SwiftUI for about two years, and i think it's a really nice language. the compiler is very slow though.
but i think it's too coupled to Apple still. when i tried getting anything running on non-Apple, i had so much trouble i decided then to not even bother.
Whats the stdlib situation for swift in comparison to newish languages like go or rust. I know its not batteries included lke python - and doesnt have a massive dev ecosystem of helper libs seeming to be mostly tied to macOS/iOS operating system API/ABI.
There are still challenges with basics like compression, which tends to involve trawling Github for the least dubious toy project. Even Apple's Compression framework is missing important algorithms like ZSTD.
Another problem is the Apache Software Foundation don't seem to have any Swift maintainers, which means there really aren't any good pure Swift libraries for Arrow or Parquet.
There are some really good open-source libraries from Apple like Swift Collections or Swift Binary Parsing.
As of very recently, the entire stdlib (i.e. "Foundation") is open source and available on all platforms Swift targets. For a while, the Linux builds had a much smaller/limited version of Foundation, but it's fully supported now.
did they ever add #define? bridging constants from the build system to swift with static NSString * const kConstValue = @XSTR(CONST_VALUE); is soooo annoying.
Re: module name selectors, wasn't this already possible, e.g. ModuleA.getValue()? Though I suppose this disambiguates if you also have a type called ModuleA.
Anyone else think the weather in the screen shot at the top of the page is a bit off? Snow in Lisbon (apparently it snowed there once in 2006), rain in Reykjavik at -1°. AI slop?
If you look at the code above, the temperature and weather are selected independently randomly. That alone is not indicative of AI-generated code; a human could write something similar for demo/learning purposes.
Swift truly is one of the languages of all time. I started a mobile app with the UI built in Swift and the core in Rust. The amount of implicit and hidden behaviour, magical fields being generated on objects because of certain annotations, the massive amount of @decorators...it's too much. I'm going to have an LLM generate the SwiftUI and touch as little of it as possible.
It's also strange because before I learned Swift, I heard about how beautiful it is but I find it much noisier than Rust.
Idk, I don't swift is objectively noisy but there is something about reading it that just hurts my brain. Stuff like
ForEach(vm.sorted(by: \.name)) { user in
row(for: user)
}
\. looks like a syntax error to me, wtf. Also, "in" instead of -> ???
.padding(.vertical, 4)
I also really dislike how you can leave off the type name, meaning you never actually know what that Type is unless you literally inspect the signature of the caller.
I remember jumping into Swift from Obj-C in 2014 and using words like "beautiful" and "expressive" to describe syntax, and saying things like "you can tell what language someone is coming from by how they code in Swift". Now it's grown as it has - sometimes feels like season 4 of Lost. Makes me feel old too.
While the Flash guys had to use a native development environment and compile their stuff, I could just edit JavaScript in a plain text file and hit reload.
20 years later, and some of the same friends now swear by Swift. And have to use a native development environment and compile their stuff. While I still prefer to just edit JavaScript in a plain text file and hit reload.
Beyond the marketing view... here are the (incomplete) changes:
Here are the evolution proposals that landed in 6.3: Overall 6.3 ended up mostly about integration: stdlib, c/c++ (often driven by swift-java interop needs), and builds.SPM (swift package manager) is slowly reaching out to take over xcode build features (e.g., resources most recently), and they have been inching and hacking their way to a new swift-build engine (now experimental) and "prebuilt" modules for the compiler/IDE macro support. Things Go Wrong all the time when trying to replace parts while driving, and there's very little visibility or control over the interactions between SPM and Xcode.
Conversely to integrations, endogenous language progress seems relatively quiet because there's a lot of behind-the-scenes work extending the initial blush of lifetime controls to iterators, stdlib types, etc. (never mind parameter packs). They're in year 4+ of building out the function-coloring of concurrency and isolation domains (6.2 had an "easy" mode that caused lots of problems), so adding year 2+ dataflow coloring of lifetimes is doubly complicating.
Naturally, diverse support for exciting new features results in quite a spread in the third-party library ecosystem of support for various language features. And now with AI entrepreneurs rewriting and porting any project that can score hits and kudos, the usual open-source authorities are losing their steam.
Finally, there's the baseline platform complexity of multiple OS and devices, physical and simulator debugging support, cloud CI...
So: that makes four dimensions of complexity. As a Swift developer you'd have to be very careful to stay on a working path as things shift underneath you.
All for the love of it?
good to see incredible stuff being shipped in Swift. Haven't used it since v3 though.
around 2015-17 - Swift could have easily dethroned Python.
it was simple enough - very fast - could plug into the C/C++ ecosystem. Hence all the numeric stuff people were doing in Python powered by C++ libraries could've been done with Swift.
the server ecosystem was starting to come to life, even supported by IBM.
I think the letdown was on the Apple side - they didn't bring in the community fast enough whether on marketing, or messaging - unfortunately Swift has remained largely an Apple ecosystem thing - with complexity now chasing C++.
> the server ecosystem was starting to come to life, even supported by IBM.
I was in college at the time and doing some odd freelance jobs to make some money. Unbeknownst to my clients I was writing their website backends in swift, using build packs on heroku to get them hosted.
It was a fun time for me and I love swift but I will admit last year I went ahead and rewrote an entire one of those sites in good ol typescript. I love swift but anything outside of the Apple ecosystem with it just seems like it hasn’t hit critical mass yet.
> Swift has remained largely an Apple ecosystem
Even today, with the fancy Swift 6.3, the experience of using Swift for anything other than apps for Apple platforms is very painful. There is also the question of trust - I don't think anyone would voluntarily introduce Apple "The Gatekeeper" in parts of their stack unless they're forced to do it.
You can use swift on the server but what for? You have a gigantic ecosystems in languages X,Y,Z.
Even Apple does not use Swift on the server (AFAIK) so why would you?
What, of course Apple uses Swift on the server, that's the only reason they're investing in any of this. Many of the foundational Swift on the server libraries were written at Apple and later opened, like SwiftNIO.
That's outright false:
https://www.swift.org/blog/swift-at-apple-migrating-the-pass...
You could have easily fact-checked before forming an opinion, but at least the buffoon down there agreeing with you is worse
Man you gotta touch some grass instead of just insulting people.
Smoking grass can also help with this
> Even Apple does not use Swift
Exactly true - they've created all these "working groups" of open source / volunteers to care for Android / Server / Wasm / ... all while being constraint "as an Apple product". Of course the end result is crappy
Yea there is no incentive. Why use Swift on the server or in k8s when you have gazillion other languages that are performant and have the ecosystems.
Python 3 barely managed to dethrone Python.
I'm sorry, that's absolutely bullshit. In fact, I wish we had left everyone who complained behind—the python community would have been happier and healthier for it. Absolute crybabies who wanted to be catered to without caring for how intractable the problems with python2 were—e.g. dealing with unicode was a royal pain in the ass, and the bytes/string divide completely fixed it. IMO, it was the best-executed breaking change I've ever witnessed in a language.
In comparison, e.g. Scala 2 -> Scala 3 was an absolute nightmare—it just didn't have the same vocal wailing from maintainers in the community (or, I suppose, a fraction of Python's popularity to begin with).
Being to aggressive in breaking stuff gets you a shitshow like Node.js or Ruby. Long-term source code compatibility is a very useful feature for open source and a sign of a mature eco system. Feel free to add stuff, but once it's part of a stable release it has to be maintained long after a "better" way to do it comes along.
nodejs itself doesn't have very many breakages; i have plenty of code that is unchanged from 0.12 to 24. npm is a whole other kettle of fish but I don't think you can blame the core project for the sins of everyone that publishes to the package manager. Python2 -> Python3 on the other hand had a lot of breakage in "standard" code.
True. Google was even thinking of switching TensorFlow from Python to Swift.
https://github.com/tensorflow/swift
That’s really because Chris Lattner was at Google Brain at the time. Don’t think it ever took off in meaningful ways
I was enthusiastic about early TensorFlow in Swift efforts, sorry when the effort ended. My interest then flowed into early Mojo development for a while.
I wrote an eBook on Swift several ago but rarely update that book anymore. Count me as one of the many developers who for a while thought Swift would take over the world. At least Swift is a fun language to use, and now with LLM coding tools writing macOS/iOS/iPadOS apps is fairly easy.
funnily enough, I talked recently to someone working on the swift compiler (not an Apple employee) to make Swift functions differentiable. So its not all dead yet
Python's interactive interpreter makes it pretty useful as a shell, for iterative development, and crucially useful in a Jupyter notebook. I've also found CircuitPython's interpreter to be bonkers useful in prototyping embedded projects. (This, on top of the nice datascience, ML, and NN libraries).
Swift just wasn't doing the same things. And even if it did, Swift would compete with other languages that were understood as "a better Python", like Julia. Even then, Swift only came to Linux in 2016, Windows in 2020, and FreeBSD less than a year ago with WWDC 2025.
I think it doesn't help that the mid 2010s saw a burst of Cool and New languages announced or go mainstream. Go, Julia, Rust, TypeScript, Solidity, etc. along with Swift. I think most of us only have space to pick up one or two of these cool-and-new languages every few years.
> could plug into the C/C++ ecosystem. Hence all the numeric stuff people were doing in Python powered by C++ libraries could've been done with Swift.
In 2015-2017 you could interop with C, C++ support wasn't added until very recently.
I do agree with you though and I am not sure what the exact reasoning is, but Swift is definitely an Apple ecosystem language despite the random efforts to gain traction elsewhere.
> around 2015-17 - Swift could have easily dethroned Python.
Why could it?
> it was simple enough - very fast - could plug into the C/C++ ecosystem. Hence all the numeric stuff people were doing in Python powered by C++ libraries could've been done with Swift.
Half a dozen languages fit this description.
> the server ecosystem was starting to come to life, even supported by IBM.
No, not at all. Kitura, Vapor (a fitting name) were just a toys that no serious player ever touched.
After that, and IBM losing interest, Apple did hire a few competent people (including contributors to Netty and Akka) to build the Swift Server Workgroup.
But I don't know why I'd pick Swift on the server when Rust is better in almost every dimension, with a thriving and more community-driven ecosystem.
I think it's not about that but about dogfooding Swift on the server. Apple uses Go, Java etc for a lot of its server components and refused to invest in hiring people that would extend the ecosystem for server Swift.
Thats the problem.
It certainly doesn't help, but among big tech, Apple is not the only company where teams are siloed and independent. Microsoft has people writing Java or Go instead of C# too.
I assume the server side usage is not zero, but not enough to reach a critical mass, you're probably right there.
Maybe Chris Lattner leaving and creating Mojo also didn’t help in that regard.
Swift for TensorFlow was a cool idea in that time …
It remains to be seen how much Mojo has learnt from that effort.
NVidia, AMD and Intel now have doubled now into giving Python GPU JITs, and Julia, the same capabilities as their CUDA, ROCm, and SYSCL offerings with C++.
With Julia and Python having their 1.0 long behind them.
Lattner probably left because Apple didn't give the team any breathing room to properly implement the language. It was "we must have this feature yesterday". A lot of Swift is the equivalent of Javascrip's "we have 10 days to implement and ship it":
https://youtu.be/ovYbgbrQ-v8?si=tAko6n88PmpWrzvO&t=1400
--- start quote ---
Swift has turned into a gigantic super complicated bag of special cases, special syntax, special stuff...
We had a ton of users, it had a ton of iternal technical debt... the whole team was behind, and instead of fixing the core, what the team did is they started adding all these special cases.
--- end quote ---
For this language to become default at Apple they had to be doing a massive amount of internal promotion - in other words they knew where it was going.
And then if that's the case, how were they not ready to solve the many problems that a big organization would run into? And all the schedule constraints that come with it?
> Swift has turned into a gigantic super complicated bag of special cases, special syntax, special stuff...
That's true, but only partly true. It already was a gigantic super complicated bag of special cases right from the start.
Rob Rix noted the following 10 years ago:
Swift is a crescendo of special cases stopping just short of the general; the result is complexity in the semantics, complexity in the behaviour (i.e. bugs), and complexity in use (i.e. workarounds).
https://www.quora.com/Which-features-overcomplicate-Swift-Wh...
Me, 2014:
Apple's new Swift language has taken a page from the C++ and Java playbooks and made initialization a special case. Well, lots of special cases actually. The Swift book has 30 pages on initialization, and they aren't just illustration and explanation, they are dense with rules and special cases
https://blog.metaobject.com/2014/06/remove-features-for-grea...
Of course, that doesn't mean that it didn't get worse. It got lot worse. For example (me again, 2020):
I was really surprised to learn that Swift recently adopted Smalltalk keyword syntax ... Of course, Swift wouldn't be Swift if this weren't a special case of a special case, specifically the case of multiple trailing closures, which is a special case of trailing closures, which are weird and special-casey enough by themselves.
https://blog.metaobject.com/2020/06/the-curious-case-of-swif...
Oh, and Function Builders (2020, also me):
A prediction I made was that these rules, despite or more likely because of their complexity, would not be sufficient. And that turned out to be correct, as predicted, people turned to workarounds, just like they did with C++ and Java constructors.
https://blog.metaobject.com/2020/04/swift-initialization-swi...
So it is true that it is now bad and that it has gotten worse. It's just not the case that it was ever simple to start with. And the further explosion of complexity was not some accidental thing that happened to what was otherwise a good beginning. That very explosion was already pretty much predetermined in the language as it existed from inception and in the values that were visible.
From my exchange with Chris regarding initializers:
"Chris Lattner said...
Marcel, I totally agree with your simplicity goal, but this isn't practical unless you are willing to sacrifice non-default initializable types (e.g. non-nullable pointers) or memory safety."
Part of my response:
"Let me turn it around: Chris, I totally agree with your goal of initializable types, but it is just not practical unless you are willing to sacrifice simplicity, parsimony and power (and ignore the fact that it doesn't actually work)."
Simplicity is not the easy option. Simplicity is hard. Swift took the easy route.
[...] when you first attack a problem it seems really simple because you don't understand it. Then when you start to really understand it, you come up with these very complicated solutions because it's really hairy. Most people stop there. But a few people keep burning the midnight oil and finally understand the underlying principles of the problem and come up with an elegantly simple solution for it. But very few people go the distance to get there.
-- Steve Jobs (borrowed and adapted from Heinelein)
https://blog.metaobject.com/2014/04/sophisticated-simplicity...
To be fair, I think such a fate in inevitable for most languages after many years of changes and development.
The thing what people don't get with C++'s complexity is that complexity is unavoidable.
It is also there in Ada, C#, Java, Python, Common Lisp,....
Even if the languages started tiny, complexity eventually grows on them.
C23 + compiler extensions is quite far from where K&R C was.
Scheme R7 is quite far from where Scheme started.
Go's warts are directly related to ignoring history of growing pains from other ecosystems.
> Even if the languages started tiny, complexity eventually grows on them.
And then of course the case that proves the opposite, Clojure. Sure, new ideas appear, but core language is more or less unchanged since introduced, rock solid and decades old projects still run just fine, although usually a bit faster.
That is because Clojure is done, there is hardly anything being done other than probably what matters to NuBank and Datomic mostly.
Also its market share adoption kind of shows it.
That's a pretty far cry from "complexity is unavoidable". Reading that to me implies that the complexity is inherent in programming language design, whereas this follow-up argument seems to say that complexity is the result of tacking on new features.
The latter is a bit tautological, since the size of the language grammar is itself a measure of complexity.
I think they even haven't adopted newer JVM features, it is a hosted language designed to depend on its host, plus it is a Lisp.
The complexity would be to grow like Common Lisp, instead it is up to Clojure folks to write Java, C#, JavaScript code, therein lies the complexity.
The current market share shows how far you can go with just being a better Java.
If (or when? I haven't checked recently) a decent and well-thought-out LLVM backend emerges for it, ideally with some new underlying complexity seeping through, the market share might expand overnight.
And as for C++, while some complexity is certainly unavoidable, a rigorous complexity control is desperately needed. Ideally, the same way Bell Labs folks did when they initially conceived Go from Algol68 and C and similar (before or after joining Google; I couldn't tell), and Rich Hickey did when he initially designed Clojure. Some people are managing the complexity using style guides and clang-tidy checks. Which is great in that doing so doesn't need lengthy language committee decisions. But that approach hasn't been enough to make code _sufficiently_ safe; every now and then an enterprising engineer or team finds a way to abuse a feature in a way that produces unsafe or unpredictable results. Rust is a bit better and solves a few of the common problems, but sadly the list of potential issues (of using Rust in a codebase at scale; Engineers' faults, not Rust's) is long and growing. My verdict is we need both complex and simple LLVM languages, ideally co-designed to have no interop problems by design, while allowing expressing some logic in the simple parts and some logic in the complex parts. Or better, a 3 tier design would be nearly perfect: expressive config language, glue and research language, and core building blocks language. I think a clojure-style language can be designed to achieve all three.
All good, except that going forward the new languages to be designed are going to be specifications and formal verification for agents.
I think the way of classical programming languages is behind us, unless AI implodes and we are back to programming without it.
> That is because Clojure is done
Yes, that's one approach to avoiding ever growing complexity, maybe the other languages should try it sometime ;)
With that said, everything around Clojure keeps improving and getting better. While the language doesn't have static types, clojure.spec offers something that is even better than static typing (imo), and doesn't even require any changes to the core language. Something else other mainstream languages could learn too.
Is Typed.Clojure finally stable and sound?
In theory we only need parentheses, prefix operators and a REPL, but mainstream never went down that route.
Anyway the complexity then ends up being custom DSLs and macros.
> Swift could have easily dethroned Python.
Just IMO, but... no. To me a "could have easily" requires n-1 things to have happened, and 1 thing not happening. Like, we "could have easily" had a nuclear exchange with the USSR, were it not for the ONE Russian guy who decided to wait for more evidence. https://en.wikipedia.org/wiki/1983_Soviet_nuclear_false_alar...
But even in '15-'17, there were too many people doing too many things with Python (the big shift to data orientation started in the mid/late 90's which paved the way to ML and massive python usage) by then.
The 'n' was large, and not nearly of the 'n' things were in Swift's favor then.
Again, IMO.
That's my read too.
Swift was feeling pretty exciting around ~v3. It was small and easy to learn, felt modern, and had solid interop with ObjC/C++.
...but then absolutely exploded in complexity. New features and syntax thrown in make it feel like C++. 10 ways of doing the same thing. I wish they'd kept the language simple and lean, and wrapped additional complexity as optional packages. It just feels like such a small amount of what the Swift language does actually needs to be part of the language.
I get this feeling with C#. I have been here since its release. I looked at Swift and then they moved very quickly at the beginning, so the book I had to teach me was out of date moments after it was printed. With all the complexity being thrown in, I stuck with C++ because at least it was only 1 language I had to keep track of (barely)!
C# is the other direction, IMO.
I've been using C# since the first release in 2003/4 timeline?
Aside from a few high profile language features like LINQ, generics, `async/await`, the syntax has grown, but the key additions have made the language simpler to use and more terse. Tuples and destructuring for example. Spread operators for collections. Switch expressions and pattern matching. These are mostly syntactic affordances.
You don't have to use any of them; you can write C# exactly as you wrote it in 2003...if you want to. But I'm not sure why one would forgo the improved terseness of modern C#.
Next big language addition will be discriminated unions and even that is really "opt-in" if you want to use it.
> Next big language addition will be discriminated unions and even that is really "opt-in" if you want to use it.
I was excited for DU until I saw the most recent implementation reveal.
https://github.com/dotnet/csharplang/blob/main/proposals/uni...
Compared to the beauty of Swift:
https://docs.swift.org/swift-book/documentation/the-swift-pr...
The C# impl is still early and I think what will end up happening is that a lot of the boilerplate will end up being owned by source generators in the long term. C# team has a habit of "make it work, make it better". Whatever v1 gets released is some base capability that v2+ will end up making more terse. I'm happy and OK with that; I'd rather have ugly unions than no unions (yes, I already use OneOf)
Ah Source Generators, after all these years still badly documented, when searching you most likely will find the original implemenation meanwhile deprecated, have poor tooling with string concatenation, and only have a few great blog posts from .NET MVPs to rely on.
:shrug: we're using them very effectively and there are plenty of resources at this point.
Very useful for reducing boilerplate and we can do some interesting things with it. One use case: we generate strongly typed "LLM command" classes from prompt strings.
Which keywords would you get rid of and why? You don't have to use all of them!
I would remove result builders and all other uses of @attributes that change the semantics of the code (e.g property wrappers).
I would remove the distinction between value types and reference types at the type level. This has caused so many bugs in my code. This distinction should be made where the types are used not where they are defined.
I would remove everything related to concurrency from the language itself. The idea to let code execute on random threads without any explicit hint at the call site is ridiculous. It's far too complicated and error prone, which is why Swift designers had to radically change the defaults between Swift 6.0 and 6.2 and it's still a mess.
I would remove properties that are really functions (and of course property wrappers). I want to see at the call site whether I'm calling a function or accessing a variable.
I would probably remove async/await as well, but this is a broader debate beyond Swift.
And yes you absolutely do have to know and use all features that a language has, especially if it's a corporate language where features are introduced in order to support platform APIs.
I agree with you about result builders, silly feature that only exists for SwiftUI.
But a lot of what you said, except for the concurrency and property wrapper stuff, largely exists for Obj-C interop. The generated interface is more readable, and swift structs act like const C structs. It’s nice.
I'm not a Swift user, but I can tell you from C++ experience that this logic doesn't mitigate a complex programming language.
* If you're in a team (or reading code in a third-party repo) then you need to know whatever features are used in that code, even if they're not in "your" subset of the language.
* Different codebases using different subsets of the language can feel quite different, which is annoying even if you know all the features used in them.
* Even if you're writing code entirely on your own, you still end up needing to learn about more language features than you need to for your code in order that you can make an informed decision about what goes in "your" subset.
But you have to know all of them to read other people's code.
To answer your question: I would immediately get rid of guard.
Also, I think the complexity and interplay of structs, classes, enums, protocols and now actors is staggering.
I'm surprised, guard is really useful, especially when unwrapping optionals. It's terse, explicit and encourages defensive programming.
internal should definitely go though.
The absence of guard in Kotlin is one of those things that regularly trips me up when bouncing between it and Swift. Rather than Swift losing guard I’d prefer if Kotlin gained it.
I think the ?: operator ends up being a decent alternative, e.g.
Unless there's a use case for guard I'm not thinking ofi would get rid of associatedtype, borrowing, consuming, deinit, extension, fileprivate, init, inout, internal, nonisolated, open, operator, precedencegroup, protocol, rethrows, subscript, typealias, #available, #colorLiteral, #else, #elseif, #endif, #fileLiteral, #if, #imageLiteral, #keyPath, #selector, #sourceLocation, #unavailable, associativity, convenience, didSet, dynamic, indirect, infix, lazy, left, mutating, nonmutating, postfix, precedence, prefix, right, unowned, weak, and willSet
It's true that internal is pointless.
Focusing on the keywords rather than the macros, I think the rest of them have legitimate use cases, though they're often misused, especially fileprivate.
You can take this approach in personal projects - with teams you need to decide on this and then on-board people into your use of the language. This does not work.
Yes exactly, it’s easy to blame a language when really it’s a team problem.
1. You don't have to use it all, but someone will. And there are over 200 keywords in the language: https://x.com/jacobtechtavern/status/1841251621004538183
2. On top of that many of the features in the language exist not because they were carefully designed, but because they were rushed: https://news.ycombinator.com/item?id=47529006
That number is unfairly exaggerated. The list includes ~40 internal keywords used only by language developers, plus dozens of tokens that would be called preprocessor directives, attributes, or annotations in other languages (e.g. `canImport` as in `#if canImport(...) #endif`; `available` and `deprecated` as in `@available(*, deprecated) func`).
are there actually 217 keywords? Just wondering what the difference between that file and https://docs.swift.org/swift-book/documentation/the-swift-pr... (a mere 102 keywords)
That file is the compiler's list of reserved keywords, so some of them may not have been added to docs, or they're experimental/internal/...
I'm not 100% sure but I think the swift doc you linked is missing at least a dozen keywords so the truth probably lies in the middle
Ah makes sense, personally I wouldn't consider reserved but unused words as keywords in the sense that you don't need to know them to read the language (even though they're keywords in some other technical sense). I was curious because I just tried counting number of keywords by language and it seemed surprisingly ambiguous/subjective/up to the language to say what's a "keyword" vs some type of core module. So my attempt (https://correctarity.com/keywords) probably has mistakes...
I felt that too many smart people were getting involved in the evolution of the language. There should have been a benevolent dictator to say NO.
>"around 2015-17 - Swift could have easily dethroned Python."
NumPy, SciPy, Pandas, and Pytorch are what drove the mass adoption of Python over the last few years. No language feature could touch those libraries. I now know how the C++/Java people felt when JS started taking over. It's a nightmare to watch a joke language (literally; Python being named for Monty Python) become the default simply because of platform limitations.
> Haven't used it since v3 though.
Since 5.10 it's been worth picking back up if you're on MacOS.
Eh, I don't think Swift would ever have dethroned Python. What pain point would it practically solve? I don't use Python often but I don't hear folks complaining about it much.
I do, though, think Swift had/has(?) a chance to dethrone Rust in the non-garbage collected space. Rust is incredibly powerful but sometimes you don't really need that complexity, you just need something that can compile cross-platform and maintain great performance. Before now I've written Rust projects that heavily use Rc<> just so I don't have to spend forever thinking about lifetimes, when I do that I think "I wish I could just use Swift for this" sometimes.
You're right, though, that Swift remains Apple's language and they don't have a lot of interest in non-Apple uses of it (e.g. Swift SDK for Android was only released late last year). They're much happier to bend the language in weird ways to create things like SwiftUI.
> just need something that can compile cross-platform and maintain great performance.
I think Go has already taken that part of the cake.
Go is garbage collected, though. Rust and Swift still occupy a niche Go doesn't.
ARC is a form of garbage collection. Swift does not fare better than Go usually.
> Swift could have easily dethroned Python
No way something that compiles as slowly as Swift dethrones Python.
Edit: Plus Swift goes directly against the Zen of Python
> Explicit is better than implicit.
> Namespaces are one honking great idea -- let's do more of those!
coupled with shitty LSP support (even to this day) makes code even harder to understand than when you `import *` in Python.
Edit 2: To expand a little on how shitty the LSP support is for those who don't work with Swift: any trivial iOS or macOS project that builds fine in Xcode can have a bunch of SourceKit-LSP (the official Swift LSP) errors because it fails to resolve frameworks/libraries. The only sane way to work with Swift in VS Code or derivatives I've found is to turn off SourceKit diagnostics altogether and only keep swiftc diagnostics. And I have the swift-lsp plugin in Claude Code, there's a routine baseline of SourceKit errors ignored. So you have symbols without explicit namespaces, and the LSP simply can't resolve lots of them, so no lookup for you. Good luck.
>No way something that compiles as slowly as Swift dethrones Python.
This must have pushed Chris Lattner towards making Mojo both interpreted and compiled at the same time.
> Explicit is better than implicit.
That's funny. To me magic is implicit by definition and Python strikes me as a very magical language compared to something like Java that is way more explicit.
Until you start using frameworks like Spring and then everything is so painfully magic that no one knows how the program actually runs.
Magical language how? And you should see what reflection based Java monstrosities do in the background.
Plus Swift goes directly against the Zen of Python
The Zen of Python is how we got crap like argparse where arguments are placed in the namespace instead of a dict.
I wouldn't change that in any way. I'd might make it an Arguments class, but I wound't make what parser returns merely a dict.
Yeah, so what happens when you have an option with a '-' in it that isn't valid as a variable name (I know what happens). It's just stupid.
The same thing you'd do yoursef if you wanted to assign it to a namesake local variable even if it was in a dict to begin with: you'd make the dash an underscore.
It would be extremely unlikely that you would replicate the name as a local variable if it was in a dict, but regardless a dict doesn't have that limitation. The namespace thing is atrocious and bad design -- no straightforward way to iterate over them, merging/updating them is awful, collides with keyword methods (keys, items, etc.), and so on; thankfully more modern argument parsing libraries didn't repeat this mistake. It's just a shame this ended up in the standard library, but then Python standard library has never really been any good, e.g. logging and urllib1234567.
>It would be extremely unlikely that you would replicate the name as a local variable if it was in a dict
If you had some feature flag args, you'd keep accessing them via the dict? Highly unlikely...
Dethroned Python? The Apple language, seriously. Where is numpy for swift?
I spent last week (with Opus, of course) porting the xv6-riscv teaching operating system to a bunch of different languages. Zig, Nim, LISP, and Swift.
The improvements in embedded Swift have definitely made it one of the most enjoyable/productive languages to work on the OS. I feel like I can build useful abstractions that wrap raw memory access and make the userland code feel very neat.
On the other hand, the compilation times are SO bad, that I'm really focusing on the Nim port anyway.
It's been a long time since I came across Nim. I thought it was really interesting about 12 years ago. What made you land on Nim instead of any of the more obvious alternatives?
I was looking for something that allows easy access to direct memory, with a syntax thats a little easier to explain than C. Frankly, zig was not actually a real viable option based on that syntax requirement but I still wanted to explore it.
Nim really is clean and simple.
Nim is really incredible. The only things I cannot get over is the fact that it goes the inheritance route in a way I find to be hacky and fragile (no more than one level, really?) and traits are not a core feature. If Nim's primary approach was composition + Rust-style traits (that work at compiletime and runtime), I'd have a hard time wanting to use anything else.
yeah, Nim is great for that... much easier to explain to others than C or Zig especially for math code
Yeah, for a language that claims to be a better modern alternative to C, zig verbose syntax is really an eyesore to look at compared the very same codebase written in C...
I lost immediately any interest on it
How about Odin?
LISP like McCarthy LISP?
> Swift is designed to be the language you reach for at every layer of the software stack.
It's a nice lang for sure, but this will never be true with the way things are. Such wasted opportunity by Apple.
How so? I can indeed target every layer of the software stack using Swift, today.
E.g. ClearSurgery[0] is written fully in Swift, including the real-time components running on the Linux boxes.
[0] https://clearsurgery.vision
I _can_ do the same with Rust, doesn't mean it's "the language I reach for" for making e.g. a website. Because the tooling, ergonomics, hireability factor, etc. are still very harshly against it.
Same with Swift, but I'd call that more of a wasted opportunity because Apple, unlike Rust Foundation, has a mountain of money to make it happen, and yet they don't seem to care.
> They don't seem to care.
I don’t believe that’s true. Things are moving constantly, and in the right direction. Then again it would help if you cited particular grievances, because being a regular (cross-platform/cross-target) Swift user I am not sure what you are talking about…
I did not choose ClearSurgery’s example randomly. I was at a conference recently where the CTO was here, and he explicitly told us they were moving fast thanks to the Swift ecosystem. (I am not working there personally, nor am I affiliated.)
they seem to be adding more and more keywords
if they really want me to use this lang for everything, they'd have to 1. massively improve compilation speed, 2. get the ecosystem going (what's the correct way to spin up an http server like with express?) and 3. get rid of roughly 150 of the 200 keywords there are
especially w.r.t. the last one, of course everyone frets at huge breaking changes like this, so it won't happen, so people won't use it
> 3. get rid of roughly 150 of the 200 keywords there are
I don't understand this point. Could you explain?
The new keywords enable new language features (ex: async/await, any, actor), and these features are opt-in. If you don't want to use them, you don't have to.
What are they keywords you think should be removed?
> these features are opt-in. If you don't want to use them, you don't have to.
Using a language is more than just writing it with a pre-established knowledge of what subset of features you think is worth the tradeoffs. More keywords/features means when you try to figure out how to do something new, there may be 15 different ways and you need to analyze and figure out which is the best one for this scenario, which ones are nonstarters, etc.
That's was more or less the whole design goal of Go. It was made by C++ programmers who were fed up with how many features were in the language, so they kept the feature set limited. Even the formatting is decided by the language. You may not agree with every decision, but what matters is decisions were made and they're standardized, so everyone is on the same page. You can read anyone else's code, and you know exactly what's going on.
> I don’t believe that’s true. Things are moving constantly, and in the right direction.
Hah! I'll use that argument if I ever get PIP'd.
No but seriously, constantly moving doesn't mean fast enough. Swift took took long to have cross-platform support.
And it is still uberslow to compile. To the point of language servers giving up on analyzing it and timeout.
Not just uber slow to compile, because as a Rust dev I could take that. But it rejects correct programs without telling you why! The compiler will just time out and ask you to refactor so it has a better shot. I understand that kind of pathological behavior is present in many compilers but I hit it way too often in Swift on seemingly benign code.
I don't know why anyone would want to use Apple tools if they are not developing for Apple platforms. Apple barely maintains compatibility for their own platforms, using Swift on a non-Apple platform is setting yourself up for doubule pain.
> Apple barely maintains compatibility for their own platforms...
You're commenting on a post about an update... that they apparently don't do? What?
Why are you interpreting this comment as "never receives updates"? It takes great effort to maintain API compatibility, some things aren't improved or are implicitly deprecated.
2/10
That was true for Swift 2, maybe a little for Swift 3, but it has not been true since a long time now…
In a way it still is true. Swift works on Windows and Linux until it doesn't. It's taken until a couple years ago for other build systems to get swift support (which I suppose is the fault of said build system, but Swift taking so long to be cross-platform contributed to that), and even now it (still) doesn't quite work right. C interop is a mess requiring hacks to generate clang modules to actually get Swift to see them (and CMake for example provides no easy way of doing this, or last time I checked it didn't). Oh and Swift tends to take over the linker and compilation pipelines when you enable it, at least with CMake, because... Reasons? I honestly don't know why. It causes very weird errors when I integrated Swift code into my C++ project that were a pain to actually diagnose. I eventually got it working, but still, it wasn't simple or seamless.
If cross platform support took so long, it's a major red flag.
Plus Swift is arguably too unnecessarily complex now.
And there's Rust/Zig so why use Swift for low level?
https://github.com/finagolfin/swift-android-sdk
At least 6 years old.
That it's designed for a thing and becoming the go-to choice for that thing can be far apart indeed.
It just works. One language. Many platforms. Incredible performance.
With a simple tooling. No ugly script. Everything is naturally integrated.
> No ugly script
What’s that supposed to mean?
The typical Apple sales pitch. Forgive me for assuming it’s only surface level.
Isn’t that Go?
Go and “simple tooling” don’t really belong in the same sentence. Powerful tooling, sure, but simple?
Would be helpful if you elaborate which part is not simple.
Coming from C++ and JavaScript, there aren't many languages that can claim to have "simpler" tooling than Go.
The tools aren’t bad any more, but you do need a few liners to write safe code. But that’s the case for most languages
No mention of compilation speed improvements? Very unfortunate. Compilation times slower than rust really hampers the devx of this otherwise decent language.
I tried Swift a few months ago in a project that made use of a bunch of bigger dependencies and I was instantly shocked by the compilation times. It's quite unimaginable to me using Swift for everyday work because of that. Especially when coming from the fast compile times of Go. But it's really unfortunate because I really enjoyed writing Swift because it feels like a very well made language. But iterating on some code and trying to get quick feedback is pure pain.
> Swift 6.3 introduces the @c attribute, which lets you expose Swift functions and enums to C code in your project. Annotating a function or enum with @c prompts Swift to include a corresponding declaration in the generated C header that you can include in your C/C++ files
Why did this take so long to be added? Such strange priorities. Adding an entire C++ compiler for C++ interoperability before adding... C exports. Bizarre.
C++ interop got attention because it helps Apple absorb low-level codebases that already moved past pure C. Exporting Swift to plain C mostly means more DIY FFI spaghetti.
Once enums, ownership rules, and nullability cross that boundary, the generated header stops looking like a neat bridge and starts looking like one more place for ABI bugs to hide. Closures make it weirder fast, because now your error handling and calling conventions can drift just enough to produce the kind of bug that wastes a whole afernoon.
They had it earlier, as an underscored attribute.
You already had ObjC export so it was arguably low priority given the crossover
it's been there for a while as an experimental feature. I used it in a project
I remember building dylibs in Swift for use in C programs, had to use @cdecl annotation iirc to achieve that, which was experimental. Good to see it's finally official
> Swift 6.3 includes the first official release of the Swift SDK for Android.
Anything similar for Windows and Linux?
For Windows there's a 5 year old blog post: https://www.swift.org/blog/swift-on-windows/
For Linux there's a guide for GNOME: https://www.swift.org/blog/adwaita-swift/
It would be really nice if instead we could just do one style of development and then ship a set of libraries as used to work for OpenSTEP (which was why it had "OPEN" in the name).
Swift on Windows has been part of the official distribution for a long time:
https://www.swift.org/install/windows/
The blog posts seems to place higher on search results --- maybe arrange to have it edited?
Which GUI toolkit(s) does that install support?
None, you’re on your own. The Browser Company made bindings for WinRT so you can use WinUI imperatively but it’s non trivial to set up.
I haven't shipped any Swift on Windows myself but I have a production Linux system using Swift (and C++ interop) and it works really well
That is going to be used... less than Swift for the servers
Interestingly, Kotlin has a pretty solid cross-platform story.
I'd pick it over Swift if targeting Android since it can build and run in the JVM as well as natively -- and has Swift/ObjC interop. Its also very usable on the server if you wanted to, since you can use it in place of Java and tap into the very mature JVM ecosystem. If that's what you're into.
And I have a lot more faith in JetBrains being good stewards of the language rather than Apple, who have a weird collection of priorities.
Kotlin is practically a no-brainer when you have JVM at your finger tips, versus something like Swift which is comparatively young.
I tried to use Vapor with Swift recently and struggled to get something working because the documentation looked comprehensive, but had a lot of gaps. I ended up throwing it out because I didn't have the time to dig through the source to understand how to do something, when I could use a mature framework in any other language instead.
The promise is there but I'm just not ready to invest. My youthful days of unbounded curiosity are coming to an end and these days I just want to get something done without much faff.
Mind you, Kotlin/Native (which is what gets used when you're compiling for iOS) doesn't have access to the JVM.
However, the Kotlin community is fundamentally all about open source, whereas Apple & iOS Devs have an allergy to it. The quality and quantity is already miles above the vast majority of what's in the Swift ecosystem. https://klibs.io has all the native compatible libs. And if you're targeting a platform where the JVM is available then yeah, it's massive. Compose makes UI tolerable compared to JWT too. Even large projects like Spring are Kotlin first nowadays.
JetBrains has monetary interest in promoting Kotlin beyond Android, there’s zero incentive to promote Swift as the language outside of iOS and Mac. They don’t need to capture minds of devs for them to develop for Apple devices.
I don't know. Could be nice for those developers that prioritize iOS and now they could keep writing Swift also for Android.
Is it gonna be what you primarily use if you wanna write an Android app? Probably not.
Is it gonna displace react Native? Probably not. Is it gonna reach the levels of flutter? Maybe.
The language doesn’t really matter. The underlying SDK/framework is where the action is at.
However, I suspect that we may not be too far off, from LLMs being the true cross-platform system. You feed the same requirements, with different targets, and it generates full native apps.
> from LLMs being the true cross-platform system
Fully agree. I have zero Swift knowledge and currently use LLM to write a native app. I'm well aware of the SDKs and concepts in iOS development, so even if something's wrong I got intuition where to look and how to make the LLM fix it.
I’m not sure about that but porting libraries from one language to another seems well within their capabilities.
I don't think we're there yet, but I suspect that it's just a matter of time.
It would certainly be quite profitable. Money tends to drive progress.
> Is it gonna reach the levels of flutter? Maybe.
Never. It won’t even reach Compose level, Flutter level DX is unattainable for any framework outside Flutter.
This is going to be used much more than Swift for servers. Swift is a primarily client-side mobile language. It makes sense that you tap into reusing the logic.
Just like .NET for linux... right? RIGHT?
The noncopyable types improvements are the most underrated part of this release. Finally makes it practical to model unique ownership in Swift without fighting the compiler.
How is the toolchain? Does Swift Lint and Swift Format support the newest version. Honestly, the modern program language should have the built-in formatter and recommend lint rules. It is not just shipping a program language, it is a while ecosystem.
Both are apart of the toolchain now, no external dependencies required: `swift format` and `swift format lint` to access them
i worked with SwiftUI for about two years, and i think it's a really nice language. the compiler is very slow though.
but i think it's too coupled to Apple still. when i tried getting anything running on non-Apple, i had so much trouble i decided then to not even bother.
In an effort to use swift for scripting, without the startup cost:
Swift Caching Compiler - https://github.com/jrz/tools
Whats the stdlib situation for swift in comparison to newish languages like go or rust. I know its not batteries included lke python - and doesnt have a massive dev ecosystem of helper libs seeming to be mostly tied to macOS/iOS operating system API/ABI.
There are still challenges with basics like compression, which tends to involve trawling Github for the least dubious toy project. Even Apple's Compression framework is missing important algorithms like ZSTD.
Another problem is the Apache Software Foundation don't seem to have any Swift maintainers, which means there really aren't any good pure Swift libraries for Arrow or Parquet.
There are some really good open-source libraries from Apple like Swift Collections or Swift Binary Parsing.
> There are still challenges with basics like compression
FWIW, there is an active discussion on this very topic: https://forums.swift.org/t/proposal-compression-library/8541...
As of very recently, the entire stdlib (i.e. "Foundation") is open source and available on all platforms Swift targets. For a while, the Linux builds had a much smaller/limited version of Foundation, but it's fully supported now.
A good source of available packages is the Swift Package Index. You can search here packages compatible with Linux[0].
[0] https://swiftpackageindex.com/search?query=platform%3Alinux
And also Android[1]!
[1] https://swiftpackageindex.com/search?query=platform%3Aandroi...
did they ever add #define? bridging constants from the build system to swift with static NSString * const kConstValue = @XSTR(CONST_VALUE); is soooo annoying.
>nocopy for disabling copy-to-clipboard
Im curious how is this used?
It is a markdown thing. It removes the "copy" button from the code block when the markdown is rendered. Search for "nocopy" on this page.
https://docs.instruqt.com/tracks/challenges/using-markdown-e...
Re: module name selectors, wasn't this already possible, e.g. ModuleA.getValue()? Though I suppose this disambiguates if you also have a type called ModuleA.
Anyone else think the weather in the screen shot at the top of the page is a bit off? Snow in Lisbon (apparently it snowed there once in 2006), rain in Reykjavik at -1°. AI slop?
If you look at the code above, the temperature and weather are selected independently randomly. That alone is not indicative of AI-generated code; a human could write something similar for demo/learning purposes.
My chances of landing at Apple increased when they launched Swift. I am too dumb to learn objective C but swift I can do.
Swift truly is one of the languages of all time. I started a mobile app with the UI built in Swift and the core in Rust. The amount of implicit and hidden behaviour, magical fields being generated on objects because of certain annotations, the massive amount of @decorators...it's too much. I'm going to have an LLM generate the SwiftUI and touch as little of it as possible.
It's also strange because before I learned Swift, I heard about how beautiful it is but I find it much noisier than Rust.
I find rust much noisier, especially at its basics where you write “for item in items” instead of using iter()
Idk, I don't swift is objectively noisy but there is something about reading it that just hurts my brain. Stuff like
\. looks like a syntax error to me, wtf. Also, "in" instead of -> ??? I also really dislike how you can leave off the type name, meaning you never actually know what that Type is unless you literally inspect the signature of the caller. ???Yeah I fully accept it's just a me problem. But I hate it haha.
What are you using for rust-swift interop?
uniffi
I want to like swift so bad
The Lord's language keeps getting better and better. The easier C interop could not have come at a more serendipitous time in my life.
Lack of Mingw support keeps me away from it, and Odin.
> @c @specialized @inline(always) @export(implementation)
Reminds me of "In case you forgot, Swift has 217 keywords now" https://x.com/jacobtechtavern/status/1841251621004538183
I remember jumping into Swift from Obj-C in 2014 and using words like "beautiful" and "expressive" to describe syntax, and saying things like "you can tell what language someone is coming from by how they code in Swift". Now it's grown as it has - sometimes feels like season 4 of Lost. Makes me feel old too.
I'm glad Chris Lattner moved on and founded Mojo. It's such a cool language with ton of potential.
I was excited about it when it first came out, but haven't heard anything about it since.
Swift reminds me a lot of Flash back in the day.
While the Flash guys had to use a native development environment and compile their stuff, I could just edit JavaScript in a plain text file and hit reload.
20 years later, and some of the same friends now swear by Swift. And have to use a native development environment and compile their stuff. While I still prefer to just edit JavaScript in a plain text file and hit reload.
Given that Swift has one of the slowest compilation times, this is a valid criticism.
Swift lives only for macOS,iOS and besides those ecosystems does not have a solid and robust ecosystem to be used for anything else.
It's a shame but it for sure needs BigTech for it to be used anywhere else.
I still have hope Swift will break free like C# has. I hope one day something like Vapor or Kitura takes off.
If Swift was paired with smh like QT/GTK (could have replaced Vala) it would've been interesting.
It has been: https://www.swift.org/blog/adwaita-swift/
smh?
Nice update, but it's a shame I can't write on it, not because I don't want to, but because it doesn't support it.
but still a good update