At a certain point, we'll collectively have to realize that the people who jump to write opinionated articles on a technology that is 2 days old probably don't have a lot of facts or an opinion worth reading.
So far, every article I've read about Swift has been opinionated. Should they all not have been written? I actually agree, but this could be said about pretty much all tech journalism. If it's more than 2 days old, it's old news.
As for Swift, which Slate compares to Clojure, OCaml, and Scala, one difference is that exactly _none_ of those languages have native support for Cocoa APIs, something you might care a lot about if you're an iOS or OS X developer.
I agree completely regarding Slate and their tendency for sensationalist headlines, but I think the author's point regarding Clojure et al is that Apple could have provided native Cocoa APIs for all of those languages, but they chose to develop Swift instead. At that point, IOS or OS X devs could have used any or all of the listed languages that had official APIs.
I suppose time will tell if they made a good choice there, but I tend to think that open source languages will always win over proprietary languages, as they can adapt faster, and get much larger community adoption. It's a pity, because Swift looks pretty nice.
The author comes across as someone who has grown a little older and is now so stuck in their old way of doing things that they become ignorant and oblivious to improvements.
I really don't get where the hate for Objective-C is coming from. Objective-C is definitely showing its age and I'm glad to see a new language that improves on it, but I really don't understand why the author's slagging it so much. I've worked in Java and Obj-C and I'd much rather work with the latter any day. I'm not saying it's perfect by any stretch, and I'm not saying it's objectively better than Java, just that there's a case for both. But I definitely don't think iOS would have better off if it had adopted Java as its language of choice instead of Objective-C.
C++ made certain decisions about syntax, Java kept many of them. They have been dominant languages for many years.
Objective-C made different decisions. They aren't clearly better or worse, just different. So C++/Java developers have to learn to read a new syntax and have to give up all of their favorite tools.
After all that work, they end using something that's basically the same as their old language. So they aren't thrilled about the change.
You're right, the syntax is the main thing I've heard complaints about. I find that really bizarre though. I've been working in C++ longer than Objective-C, and more than Objective-C, but I don't find anything offensive about the latter's syntax. So it's different. For me the measure of a language is how hard or easy it is to get stuff done. Objective-C holds up pretty well in that regard.
Wow, what a terrible article. First of all, the obvious biases of the author as an ex-Googler and java programmer shine through strong enough to make you blind, but the arguments he puts forward are just awful.
Really, java is leaps and bounds better than Objective-C? That must be why all those java feature-phones were such wonders to program for, and why android apps are so much more responsive and well-designed than iOS apps. Or maybe, that's simply not the case, and the language hasn't been that much of an issue for developers. On the contrary, binary compatibility with C and C++ has made it trivial for game developers to port their games and engines to iOS, whereas it took years for Google to belatedly release a native SDK for android.
Are android developers really less miserable than iOS developers? I've developed for both. Rather, I've looked at developing for the hodge-podge mess of vendors, phone sizes, capabilities and general horror of Android and fled from it screaming, while I've released apps on the app store despite not really buying into the walled garden of apple. It's just too easy to make apps on that platform.
"Rust creator Graydon Hoare points out that Swift “appears to have ‘Objective-C object model interop’ as a hard design constraint.” In English, that means Apple decided that Swift still needed to look as much like Objective-C as possible, wherever possible."
That's just not at all what that sentence means. Hoare is talking about binary/runtime interop, not language design.
I really doubt the creators of Go would stand behind the claims made in this article. Swift is a very impressive first release of a new programming language regardless of who is behind it, especially one that achieves binary compatibility with C (how many other modern languages manage that? Go doesn't. It needs a special C compiler for seamless interop.). Yes, the proprietary nature of it is worrying and potentially a limiting factor for the success of Swift. Every other claim this article makes is utter nonsense.
His points are not invalid, but when you're stuck using C++98 day to day because the technical board don't have nearly as much clout as a sole owner such as Apple, the grass seems so much greener on that side to me. Bringing ADTs and such to a practical language is definitely something to praise; it will take C++ so many more years to get there, if it even happens at all.
The only point I can identify with here is the implied one about avoiding swift due to its license. Why anyone would prefer a proprietary-licensed programming language is unfathomable.
12 comments
[ 3.0 ms ] story [ 33.2 ms ] threadUntil that day, this guy will have a job.
"Is There Any Truth to the Claim That Einstein Was a Fraud?"
To summarize the article: "no."
Here's an entire article on Slate's link bait tendencies: http://matthewdicks.com/2013/11/benedikt-ridiculous-offensiv...
As for Swift, which Slate compares to Clojure, OCaml, and Scala, one difference is that exactly _none_ of those languages have native support for Cocoa APIs, something you might care a lot about if you're an iOS or OS X developer.
Swift is new, but Slate is not. Total nonsense.
[1] http://www.slate.com/blogs/quora/2014/03/18/is_there_any_tru...
I suppose time will tell if they made a good choice there, but I tend to think that open source languages will always win over proprietary languages, as they can adapt faster, and get much larger community adoption. It's a pity, because Swift looks pretty nice.
Objective-C made different decisions. They aren't clearly better or worse, just different. So C++/Java developers have to learn to read a new syntax and have to give up all of their favorite tools.
After all that work, they end using something that's basically the same as their old language. So they aren't thrilled about the change.
Really, java is leaps and bounds better than Objective-C? That must be why all those java feature-phones were such wonders to program for, and why android apps are so much more responsive and well-designed than iOS apps. Or maybe, that's simply not the case, and the language hasn't been that much of an issue for developers. On the contrary, binary compatibility with C and C++ has made it trivial for game developers to port their games and engines to iOS, whereas it took years for Google to belatedly release a native SDK for android.
Are android developers really less miserable than iOS developers? I've developed for both. Rather, I've looked at developing for the hodge-podge mess of vendors, phone sizes, capabilities and general horror of Android and fled from it screaming, while I've released apps on the app store despite not really buying into the walled garden of apple. It's just too easy to make apps on that platform.
"Rust creator Graydon Hoare points out that Swift “appears to have ‘Objective-C object model interop’ as a hard design constraint.” In English, that means Apple decided that Swift still needed to look as much like Objective-C as possible, wherever possible."
That's just not at all what that sentence means. Hoare is talking about binary/runtime interop, not language design.
I really doubt the creators of Go would stand behind the claims made in this article. Swift is a very impressive first release of a new programming language regardless of who is behind it, especially one that achieves binary compatibility with C (how many other modern languages manage that? Go doesn't. It needs a special C compiler for seamless interop.). Yes, the proprietary nature of it is worrying and potentially a limiting factor for the success of Swift. Every other claim this article makes is utter nonsense.