Evolving MGM Resorts announces leadership shuffle

I'm happy to know that Chris (Ratner) is heading. I chatted with him in the last WWDC shortly after the main technical session, but he was willing to open, but could not see if the power could be passed.

The support of the standard library on Linux is definitely surprising.

Linux's support is surprising, but I think it's their great action.

Think: How much iOS app is the front end of the API server? And how much of those APIs are moving on the Linux server? Swift on Linux means swift on Linux.~All iOS client server app code can be written in the same language.

The problem is here. They will release Apple Music for Android. And is there a full support for Linux? Is the Android app partially fast?

It is better not to read too much.

Android is equipped with Bionic Libc and is different from the Glibc, which is usually installed in Linux distribution. There is definitely a difference between these two.

In addition, the average Android application is far from the average Linux application. If you were aiming to support Android, I would have said that instead of Linux.

The keyword here is "partially" swift. I will never forget that Apple will use the intermediate framework to host hig h-speed runtime and Android's native API calls. As a result, logic other than the UI is held in the code base.

More or less, they are doing iTunes. I remember that there was an incomplete library packaged with iTunes (it was like a nervous implementation of Grand Central Dispatch, neither wonderful nor par).

iTunes has a lot of Mac OS 9/ carbon, at least, and it's actually a Windows base. & amp; gt? Implementation of Grand Central Dispatch, which was neither great nor mission

Thank you for the hit suitable for R/ProgramMinghumor ;-)

What is "stdlib", maybe you're talking about something different?

That's a great news. I hope that a nod e-type ecosystem for writing small services on Linux can be done immediately. F#, Swift, and C#can now be used, so I hope the spread of JavaScript can be slow down on the server (because there are no running people).

The . NET framework has actually declined, Swift is still off the server, and there is a framework like Meteor.

Nevertheless, I support Clojure + Clojurescript. Theoretically, the principle of the same code exchange can build a much more advanced framework than Meteor.

How to reduce something? Net? I have never heard of it.

This is completely based on numbers, but it may be a bit subjective. https: // OneDrive. live. com/view. ASPX? RESID = 1e5aa35a965d3234%2.

I'd like to ask a little, why do you say so (about Go)?

From my preference, it feels too realistic. However, it is necessary to obtain from Google to solve this kind of problem (scalability, development is much more important than the code itself). Personally, I don't have a problem like Google (few people have it!), And I really like the excellent type system with algebraic mathematics/ pattern matches/ generics. 。

My taste is very realistic.

The C language was a null and realistic language created to increase the transplant of UNIX. Nevertheless, C became one of the most popular and most important languages.

I also prefer ADT/ Generic, but many people don't forget that many people are not basic and that realistic elements have a great effect on language selection (tools, ecosystem, popularity, familiarity).

GT? Nevertheless, the C language became one of the most popular and influential languages.

This was a sid e-b y-side effect of the successful emerging companies (SUN, SGI, etc.) adopted UNIX as operating systems.

UNIX's success is an inseparable relationship with C language, and vice versa. The C language has made UNIX as one of the first portable operating systems, and provided performance by at a low level.

The same can be said for a great variety using algor and mesa. However, there was no successful company that succeeded using them.

You want to say that the success of C language and UNIX is a historical coincidence chosen by two emerging companies, and that choice is a function of UNIX and C language, but of course it is nonsense. 。

Sun adopted UNIX because Sun was founded by Bill Joy. If she sympathized with UNIX and C and could not take on the task, they would have made a different choice.

As many programming languages ​​triggered, SGI and Sun were just one of the triggers.

You just supported me what I said.

The emerging companies that chose UNIX were part of the UNIX University culture in the United States.

If the background is different, the background of the mainstream OS and their system programming languages ​​will also differ significantly.

In Europe, the C language had little to do with most companies, until most companies began replacing their main frames with these companies' UNIX servers.

I have been touching both C and Pascal, so I really liked working in C.

I think there is an essential place in C language than other modern system language.

I often use GO for backend hardware, but I agree with your opinion. There are many things you can do with the interface to improve the lack of generics.

GT? I feel too realistic

I don't feel realistic at all. It depends on the user who does the work of the compiler (such as a claim).

I feel "realistic" in that GO almost faithful to "yagni". In order to increase the clarity of the code, it may be a better choice to repeat yourself than to increase the degree of abstraction. It is good to see that other languages ​​are working in the same field as GO and have made different choices in this field.

If you write GO code, you'll notice that you don't really face this problem.

At present, the most common problem is that the [] foo can match the . name (character string) attribute and the [] string cannot be obtained. I always do this on Ruby, but in Golang, it's really the worst to make a for loop to collect what is added to the new table.

 People. map (& amp; amp;: Name) = & amp; gt? [Joe, John] 

a: = make ([] string, 0) for _, p: = people range

It's not at all realistic, and it's a language for mining workers who are busy thinking. I just use Go when I need to think deeply, that is, when it's a difficult problem.

If the problem is boring, use a fun and challenging language. If the problem is fun and worthwhile, use a boring language.

Jabaskrpt is amazing

It may be interesting to see if @objc will be supported by Linux. Apple's runtime requires a dynamic link to notify Libobjc when the image is loaded, so it is difficult to support.

The Objective-C support in Linux is already excellent, and the compiler is processing it well. The problem is that the Foundation library has not been transplanted and does not use the old and unsupported relics of the OpenStep project.

CoreFoundation has a MakeFileLinux [1] for 635 or later versions (equivalent to 10. 7). Obviously it is based on Clang (because various extensions are used).

Not all CF is there, some of the glue in objc is missing, especially the last drop. I could put most of it if it came back.

What language is the link in? Looks like bash, but isn't it? This is a makefile [1]. It's very simple. It's defined like this:

 Target:DependencyCommand 

This command is executed from your shell (e. g. bash!) when the target is called and executed. For example

 All: Hello World Echo "!" Hello: Echo "Hello" World: Echo "World" 

Now when you run "make" you will see each command executed. By default, if you call "make" with no arguments it will execute the "All" target. You can also have it execute just the "World" target, for example by calling "Make World". As you can see the dependencies of the target are called first, then the target's commands are executed.

This is often used to manage dependencies in C/C++ projects, but it can be used for anything really.

If you're interested in how this tutorial works and why it's useful, see here. http://mrbook. org/blog/tutorials/make/

Thanks for breaking it. I've played and promoted Linux for years, and this was very helpful. Interesting and funny.

It's a ... makefile for ... make.

I feel like this answer was more helpful and less sarcastic. Some people are still learning.

Everyone is always learning (I hope). But some people learn something useful or become less sarcastic. )

Always just a smooch, but my gut feeling is that Swift is too integrated with the Apple Runtime to be used. I wouldn't be surprised if the Linux port of Swift has limited support for OBJC, but I'm glad to find that out to be wrong. (I started porting the Apple runtimes to Linux last year, but ran into the linking issue mentioned above.

Is the Apple dialect supported, or only in the next version?

Objective-C 2. 0 via Clang and the GNUSTEP OBJC2 runtime supports blocks, GCD, fast enumeration, declared properties, and introspection.

I've been wondering for a while now why no one has created an OBJC backend for their applications:

Linux development capabilities might be just what we need!

I have long dreamed of using object C to build a server application. The support of the compiler has been supported for many years (GCC, later Clang), but there is no standard library. Everything you want from the foundation is missing. Contains, sockets, I/O files, coding support, threads, etc. There is GNUSTEP, but when I finally saw it, it was stagnant and very GU I-oriented.

I think if you use Swift for the back end, it will be better. In recent evolved languages ​​(including Go and Rush), Swift was the closest to my ideal language.

OS OS people have been working on Gnustep for a while. http: // eTOILEOS. com/

It is recently come out of this group, and I want to see it in a serve r-side project. http: // coreObject. Org/

The best web development framework in history was the next webObjects. It was basically a web cocoa. It was wonderful. When Apple acquired WebOBJECTS, they continued this for a while and tried to make it more platformed in Java. Despite the fact that the price was greatly reduced and the license restriction was reduced, it was a commercial product, so it did not actually spread. This was the point that open source solutions like Ruby on Rails began to spread.

It is a pity that Apple opened in time, even though there was a great community around WebObjects.

For example, iTunes Store is an application for WebObjects, and the App Store is the same.

At that time, Sun had a partner with Next, installed an object C on Solaris (before the oak became Java), and gained the basic idea of ​​J2EE from Webobjects.

I think so too. It is a logical language to accomplish things, and it feels easy to read. It's much cleaner than GO, so let's leave the story of JS.

Because there is a server support. I want to write Ob j-C at the back end. Libgcd and compile dispatch alone are worthwhile, and of course ARC can write pure C online (not UNSAFE blocks).

The fact that I did not use OSX was a barrier to improving iOS development. Hackintoshing has been considerably vague, and the VMware and Vbox USB layers on Linux do not convince Virtualized OSX enough to enter my iDevices. I am not convinced that strict testing the emulator will give you high quality.

Someone has an old Mac that can execute the latest Xcode (maybe you may know what Xcode recommends more than me), and for a completely unspecified future project If you want to donate it to an open source professional developer, please email me (my handle name HN)@(Googles Email Service). thank you.

Xcode is written as Xcode instead of Xcode. Thank you, and good night. If you have a friend working at an Apple retail store, there is a big cleaning of internal machines twice a year. You can get a pretty good machine very cheap. Last time, the white MacBook made in 2009 was around $ 170.

The cheapest official retail option is $ 499 Mac mini. Not a laptop, but the latest Xcode moves. It may be necessary for the first app to make you a millionaire.

That's a great hint. My main laptop is a white MacBook made in 2009, and the SSD was eaten from another old laptop. The only nostalgic V culture is good, and everything else is good.

Does CIV V not work on Mac? Yes, but probably not on the 2009 MacBook. GT? It has been proven that hackintosh is quite ambiguous

I have been using hackintosh since 2008. It is from 2008 to use hackintosh.

Many people reported that they had succeeded, but my 2 0-yea r-old hal f-dad's attempt was a terrible failure.

I don't know what magic book they use.

Is the web just biased to success, as people who have failed generally have nothing to say?

If so, I want you to draw the whole picture here.

Skyline OS X is a great resource to install a stable and simple hackintosh. Therefore, I recommend that you try the driver.

Not difficult. Apple's computer still uses commodity hardware. The closer you are, the easier it is, but there is a forum that focuses on fine adjustments and updates to move anything. Even a CPU that Apple has never used.

If you don't feel comfortable with it (or your time is valuable enough), you should invest in a real deal. :)

With the right hardware, it works just like a real Mac.

But some hardware is incompatible and there's nothing you can do about it.

I have an old white MacBook that runs SL but not the latest Xcode. It'd be great for Python Dev stuff though.

So let's start writing an article about Swift in the same mold!

I'll do it too.

Apple needs to put the rest of OS X on servers. They could either recreate their own servers (unlikely) or at least license OS X to Dell/HP/IBM/VMware etc.

Why, when the aftermarket can do that?

OS X Server is really just a shell of what it used to be. It was great as a small business/design studio server but has lost most of its usefulness.

Almost all the documentation and best case scenarios assume that the Macs will connect to and be managed by an Active Directory/MS environment.

As a typical web/lamp host, OS X is not very performant. UNIX tools run faster on RHEL/CentOS boxes than on OS X (even if the OS X box has higher specs). MySQL is particularly bad if not tuned, and most of the literature seems to indicate that OS X (more specifically Mach) is not optimized for server workloads.

That said, there are times when you can get great value with good performance using an OS X box (see booting a Mac Pro as an image handling server).

I think Apple realized that the consumer side was more profitable at the same time they were making servers. That said, it seems like they don't need to be hostile to a partnership with a vendor that provides servers, rather than not making servers, and I think that would be more profitable for Apple than getting into servers.

Right now they're moving to barebone style Docker VMs, and applications are services on the servers.

But if they get into the server market, they'll have less credibility working with server-focused vendors. VMware would probably be just as happy, if not more, and to that extent. IBM has done so much of both that it upsets me because of the arrangement they have now.

When I was making a server, I was making a server for two purposes, a farm and studio networking.

Most of the render farms are replaced by Linux, and the studio networking can now be processed on a mini on the shelf or a Windows server in the closet.

That would have been more when Jobs was immersed in computer animations. This and the excellent audio driver and the API setup have helped to maintain Apple in the audio DAW industry. As you say, many things change and you can see why they focused. I want to have a server flavor at least. However, confusing consumers and being exposed to serious mixing up may have a great deal of support/ costs for confirming 100 % correct, and the number of customers is reduced.

Nevertheless, they like consumers, and if they are targeting consumers, the personal iCloud house at home will probably be a potential server. If there is any route, it may be realized.

Is it already offered at the airport? I remember that the link with iCloud was providing file storage / backup.

Then, if you have a laptop computer and have too many music or video, you will be able to stream to Airport Express/Appletv/iTunes Home Sharing, even if you do not have a projec t-only machine.

I know that you can add a disk via a USB port, you can use a hub to have some disks and share or back up them. However, there was nothing to be able to synchronize with iCloud, not the Apple user, other than the desktop option. However, he found this: https: // support. Apple. Com/en-us/ht204618 Remote desktops and related functions are written through an ICLOUD account, so maybe they need to access them through ICLOUD. There may be, but it may be okay if there is an option from the backup option.

I agree that the demand for Swift developers is increasing due to the growth that Swift has shown in the past. Swift may dominate mobile / ease of use. I already added it to my #todo list. )

I don't think Swift will dominate the web. JavaScript has not yet been destroyed by any other language that appeared before that.

I can't control the Internet. JavaScript has been a standard for a long time. Since JS is not a perfect language, dozens of JS pricon pillas have been made so far, and competition is quite intense. Swift is a way to be functional and provides a method of using an inseparable data structure, but not as functional or logical to force imminance. However, I personally don't think it will bring new things to the web ecosystem. However, it is certainly excellent for the OBJ-C alternative.

When I mentioned the web, I was mainly thinking about the web backend on the web server. JavaScript is definitely the standard for the front end.

I intend to double over the Swift to position Swift as a common language replacing JavaScript. Front end and back end.

I pray for good luck. One of the reasons JS is a common language is that all browsers have JS runtime, only plain text editor needs to be started, and only cheap / free web space is needed to publish an app. 。

I want you to remember my words: Swift will not be replaced by JS, and there will be no substitute for PHP.

Editorial Department: Also, if it is not developed for iOS, does it make sense to write Swift code on Linux? As a developer other than iOS, there is nothing to force Swift. Swift does not have anything that JS, Python, Ruby, etc. are no longer better. Except for iOS support. HN tends to forget that in the United States, it is not internationally the same because iOS is default. Apple is a luxury brand, not a product.

The meaning of using Swift on other platforms is that you can write software other than iOS/OSX. Swift is a fun and modern language that brings many things not found in JS/PY/RB (the ability to distribute type checks and binary at the beginning).

However, the developers do not want better languages ​​[0]. Most developers are looking for what they already have. Looking at it, there is a more fun and modern language, "just saying this kind of resistance cannot be overcome.

Programmers want a better language, but it will be difficult to persuade to switch to a new language alone. The only specific advantage of Swift is that it can be used instead of iOS's Objective-C-Objective-C is important for iOS, as it creates even bigger hurdles for its unfamiliar syntax. Swift wins on iOS because it competes with the language that no one wanted to learn in the first place.

In terms of friendliness, only Ruby seems to be familiar (based on the first impression), except for more familiar C type syntax (that is, parentheses). Ruby developers are very noticeable (especially among the startups and valleys in the valley), but not so muc h-it goes without saying that some have already moved to Rust or JS.

Let's refute

What is? But developers are looking for better languages

TypeScript and ES6 (7)+Babel. These have a fairly healthy user base, and there are so many excitement around ES6, so I think the developers really want a better language.

I also want a better language. I don't know how to persuade me, and I don't think I'm the only one who feels like that.

They want a "different" language. My comment that Apple "positions" as a substitute for JavaScript (to use this word) is very meaningful for those who are developing for Apple, IOS and OS X. Maybe.

First, Apple does not need to acquire from other browsers. They can provide Swift runtime on Safari (desktop and laptop), and I don't think it's harder to write Swift Transpila on ES5 than to write ES6/7 Transpila on ES5.

Second, Swift has many developer support. Native apps are increasingly dependent on the corresponding web APIs, so it's reasonable to be satisfied with share the code between the front and ends of the app. 。 I think the biggest advantage of Node in other serve r-side languages ​​is that they can use Node packages on both sides.

Third, Apple doesn't think the web developer cares about this problem. They are interested in providing tools and solutions to those who develop and use their platforms. If they can claim that Swift is better than JavaScript, with some indicators we don't know, they will do so. Everyone else is crying in the corner.

Also, opposition votes? Really? It's not the most radical idea in the world, and I don't think there's anybody offend.

Postscript: I'm sorry, I don't have time to watch this video properly, so I hope it doesn't end the wrong ending.

Partially agree.

Developers want better languages, but they are extremely allergic to friction.

Some of the ES6/7 are relatively low friction thanks to Babel. But most of the modern JS JS developers are still very slow and long processes.

Swift can be dented if Swift-to-JS is something, but most of the languages ​​ported to JS are experiments (see below). The advantage of Babel to other To-JS compilers is simply compiling JS to JS and making the promise that the collection steps will be completely unnecessary at all.

"JavaScript is an ASM of the web" was attractive, but could not be realized again and again. Even if iOS developers can write web apps, they don't think Swift can succeed where other developers fail.

The words may be cloudy, but I don't think Swift-in-the-browser will compromise JS. Also, I don't think Swift-on-the-server will have a big impact, but I can imagine that a shop that uses iOS wants to use Swift when developing the app API.

I agree that Apple will continue on its own. Apple is doing all the control of the ecosystem, so it's not surprising to put Swift as an alternative to JS (like a unilateral CSS extension).

About videos: Basically, Douglas Clockford (famous for the notomers of JSON and JSLINT) prefer (as a whole) programmers (as a whole) prefer "similar things" rather than "bigger things". Is claimed to be historically proven. Even if it is almost all of the point, it tends to choose almost the same as what you already know than the one that needs to adapt your mental model. This series is worth seeing, which helps JS to understand why it has built today's position.

According to your logic, JavaScript began to be developed on the server side because many fron t-end developers only wanted what they had. There are a huge number of mobile developers, and they need to write a server code in some way. If Apple supports the open source Swift), what do you think they choose?

There are much more people who write JavaScript than those who write iOS apps on Swift.

I agree that the server side Swift may be for iOS developers who write their own server API, but at best it will be another option alongside Node, Ruby, Python, and PHP. Only. After all, Swift will be a reasonable option only if you are already using Swift as a major language, that is, if you are writing a native iOS app.

It is important to remember that the web is not only JavaScript only for JavaScript for the story of "Swift in the browser. HTML, CSS, and even if you can omit the DOM operation like React. Replace JS with Swift, which is necessary to be a web app, and you need to be a web developer. It is no different.

Considering how the previous attempt was handled without actually writing JS (Ruby developers use CoffeeScript, Java developers use Dart and GWT, and Python developers use Pyjs. ) I don't think JS in the browser will go anywhere, no matter how some people want it.

Too late. If Apple did so from the beginning, I might have been able to get my interest, but at this time I am struggling to get excited. Half of the reasons I encountered a new language are the culture and community surrounding the language. At first glance, you may think it's ridiculous. It is a programming language, not a club activity. But please ask.

Golang is a realistic gathering. If you go to the#Go-nuts of FreeNode, you will notice that the conversation has hardly deviated from the problem solving. There are many links in the playground of GO. From the beginning, Go has formed their community like this. Consciously? No one knows. However, I think that the result of the startup was that the community became like that.

Haskell is a gathering of computer scientists. Haskell is a purely functional language. What is the discussion about Freenode? Very dense. People are hoping to easily understand the topics, and the community initially benefits that they can understand it until they seek more help.

Looking at Swift, it's a new iOS language. You won't come out of it. I won't write one line because I don't make an iOS app. When I first saw Swift, my impression was that it was a more strange, slightly functional language, and something unique. Maybe you are interested. But at this time, there is no need to ask iO S-specific questions to get what you need. I don't want to be a beta tester for Linux implementation. I don't want to use a language that would probably be used in a completely different way from other communities. My interest will be secon d-class.

It is not always decided, but I think so. In a world where new languages ​​appear every day, I don't know why I choose Swift.

Hey, look at me a year later and explain why you were wrong.

Six years after Go was released, StackoverFlow had less than 10. 000 questions. http: // StackoverFlow. Com/Questions/Tagged/Go

I am a GO fan. I made a website with Go and wrote some small applications.

But you overlook how much the larger swift community will make.

Whether Google officially supports the transition to Android is all betting.

Anyway, 1, 000, 000 Swift developers will change everything.

I think some of the points in the parent comments were that Swift would have many users / hold. Now, these 10. 000 questions cover many serve r-related topics. Most of the 37. 000 Swift questions are iOS questions.

If I start writing an HTTP server on Swift on the day of Swift, I won't get much help. In essence, adopting Swift for iOS apps does not help, and may impair the efforts of using Swift elsewhere.

Look at Object-C: Writing software for Apple is very common. Although it is not limited to Apple, it hardly exists elsewhere.

Objective-C did not come out of the Apple/Next Eco System because his greatest strength was not the language itself, but in the application context. When iOS spread Objective-C, C ++, Java, and . NET eco systems have been established. Swift is facing the same task, but is fighting a younger generation competitor who has not been much established. Swift wins an external recruitment, probably a painful battle, but does not fall off the cliff.

Needless to say, Swift is basically a Rust language, with a memory management model (all ARC) and a Scala style plugmatism (buil t-in OOP model). Developers from OBJC have high interest in Swift. If Swift is released for Linux, I might look for it!

I carefully compared the two languages, but I have no similarities than GO, Swift or other languages. It mentions different memory management models, which is the main feature of Rust. The only common point is to use LLVM.

Why do you stick to Android's GO when you can write an app on kotlin? Kotlin can be completely interoperated with Java. I don't have much experience with Swift, but Kotlin is very cool. It has the advantages of many languages, such as Ruby and C#. There is also support because it is made by Jetbrains, which makes the AndroidStudio core. Kotlin's runtime is as small as 200KB and is as fast as Java. Almost all magic occurs during build, so the build time is a little longer. Kotlin also works in existing Java libraries and annotation editing libraries.

Again, why do Android developers choose Swift? Kotlin works well on the server and can be compiled to JavaScript.

If I was a startup that handles iOS, Android, and backend, I would use GO or Rust in the backend. You will be able to distribute native libraries that share networks and model logic with Android and iOS. Both Go and Rust will support iOS and Android cros s-compilation.

If you don't use Java, you will use Kotlin/Groovy/Clojure/Scala for the Android app viewing layer. Since Google does not show that Java 8 is supported, lambda cannot be used unless you use another JVM language. In iOS, you will write a view layer on Swift. Both apps use shared binary libraries.

This comparison between Swift and Kotlin was interesting, and it was interesting to point out that it would be easier to switch between Android and iOS.

Kotlin is really great, but when I tested it (certainly last year), it wasn't likely to be used for production. As long as things were simple, I used Realm. io and wrote some unit tests, and all kinds of NOCLASSDEFFFFFFFFFFFFFFFUNDERROR exceptions and Dexmaker errors occurred. In short, I couldn't fix it. As I said, it's been a while. I hope that things will be improved and will continue to be improved. The language itself is wonderful

1. 0 is quite closed. The M12 has been released, supporting the interaction with the Java library that uses annotation processing like Dagger.

With the release of annotation processing support, Kotlin has been able to cooperate with the following: Kingdom.

I've been playing in Kotlin for the last few weeks. At least, the official practice repositories use this feature everywhere, so they seem to have rejected this feature and changed to an interface.

If you are particular about the type check at the time of compilation, it seems to have the advantages of C#and Ruby. But I'm not so good at JVM language, so I'm not sure if it's better than other options.

If you are using Scala for view layers, why do you want to use GO for the back end?

That's right. I don't understand the meaning of Go. Go is about the same performance as the JVM language, and JVM is easy to develop. JVM can choose from Clojure, Scala, Ruby, Groovy, Kotlin, and many others. JVM is as easy to develop as binary GO. Personally, I'm going to stick to the JVM language and Rust.

See Go points:

Open the Rosettacode website. Please choose the algorithm you know. GO version and Java version are arranged. Compare the number of words and lines.

Compare how Maven's hell and Go solve the same problem.

The version of Clojure and Groovy must be simpler. I agree that Scala is very complicated. ) Maven is the worst. Gradle wraps it, not the worst. Probably the best build tool I've ever seen.

So, what is a more concise Java that can't use the JVM ecosystem? It may be more concise than Java, but why is it better than Scala who can use the JVM Eco System?

Go is much simpler than Scala.

(I can't agree that the development background of JVM and GO is similarly complicated, but that's another debate).

Android? Why are you particular about Android when you can write an app on kotlin?

Android Java is compiled with native code at the time of installation, so I don't think there is any merit of using GO in terms of power consumption.

With Android SDK, you can guarantee that you can write more efficient applications than NDK. GitHub has 70, 000 GO repo, while Swift 31, 000 (vs.)~Swift is 5K)

If you update the page many times, the number of results will change significantly. GO is 39k-79K, Swift is 12k-42K, and Rust is in 3k-9K. Also, if you do not know exactly what is measured (for example, it contains a fork), we probably need to interpret these numbers as simply comparing the size.

Swift Repos will pass within 12 months. I think most iOS developers continued to use Swift. It was the beginning of this year that Xcode and Swift could be used.

Most people do not write open source iOS apps. No, but there are many components.

I don't think this will be backed by your hypothesis-70, 000 km in six years, but 31, 000 km in one year means that Swift has much more momentum than GO. It means.

If Google pushes new app development languages ​​for Android platforms, Dart seems to be a more obvious option. It is close to JavaScript and has a GU I-oriented library. In addition, it is natural to put such an app into an HTML5 container for iOS/Windows.

It's no surprise that there is a browser in the container in the container of the desktop application. Anyway, 1, 000, 000 Swift developers will change everything.

Object C developers could not change anything for users other than iOS. Why is Swift different?

Linux support. GCC was released at least 2. 0, 1992

This is very correct. I remember trying to find out the Objective C for years and years ago in my Linux box. However, it seemed impossible because there was no standard GUI library.

Of course, if GO becomes an Android firs t-class citizen, all bets will be missing. It will not happen soon.

I'm not saying because it is related to Apple's Swift's presentation. It is a great benefit for Android.

GT? It's a big advantage for Android.

The language has a weaker type system than Java, doesn't support exceptions so you have to debug every other line, and has very poor tooling support (because the compiler wasn't designed with IDEs in mind).

I don't really see what it brings to Android other than maybe it will add a few more developers who refused to code on Android because of Java.

It shouldn't die. The idea of ​​rewriting the whole Android framework is dumb when you can use better languages ​​like Kotlin, Scala, groovy, clojure. Go can be compiled to the Android architecture so you can share code between iOS and Android like C++ LIB. It makes sense.

What about gt? If you don't use java then you'll use Kotlin/groovy/clojure/scala for the view layer of your Android app.

So that's it. It's silly when you can use better languages ​​like Kotlin, scala, groovy, clojure.

It seems to piggyback an aspiring JVM language (Kotlin) on top of three other alternatives that have already pulled it off (Scala, Groovy, Clojure). The list of examples of alternative JVM languages ​​looks unsettling because this trick has already been done before with them (i. e. Groovy piggybacked on the qualities of Scala and Clojure).

Sorry. They all have tools that work on Android. I think Kotlin is a reasonable choice. It doesn't affect app startup and has a much smaller runtime size than the others. For example, Clojure Android adds 3 seconds to startup time. I've never used Scala, so I don't know how well it suits Android.

Did someone say something like rewriting the entire Android framework? No. It was about becoming a first-class citizen of Android. That doesn't mean that people who want a fancy IDE can coexist with Java.

It's already there. You can compile Android NDK libraries with GO. NDK is not linked with Android framework so you don't get lifecycle events, so it's not that useful. The entire Android framework written in Java in the last decade or so writes this.

The strength of the type system has no relation to the adoption rate of a language. Otherwise nobody would use C, Python, PHP, Javascript, Perl, Ruby, C++, C#, Java are foreign languages, and almost everyone would use Ada, Haskell, Ocaml.

But I agree with your last sentence.

Unless all developers who are developing apps on Android want to port all their code. That would be nice, but all indications are that Google isn't interested in that right now.

GT? too late. Apple might have been able to get my interest if he did it from the beginning, but at this time I have a hard time getting excited.

This is the same as your opinion.

(A) Because it is proposed as a language for the development of iOS apps, it has already attracted millions of developers (B) (B) and (with languages ​​such as GO, Scala, Rust) very hot. It is a wonderful modern language to do.

GT? Looking at Swift is the new iOS language.

Hn and other places have stated that many people want their tastes to this language and open source so that they can be used elsewhere.

Microsoft and Apple seem to be quite intentional to own the following ubiquitous language (the predecessor is JavaScript, PHP, Java, C ++, C, FORTRAN, COBOL). This is a language where computer science students spend time, and the language is modified by children and hobbyasts. This is a big deal.

C#was strong and improved by Microsoft's recent cros s-platform movements. Swift was late, but very good.

Fortunately, these two languages ​​are not separated. Transfer from one to the other is not terrible.

Very realistic. Can both win the same way? Here in the UK, C#seems to be a high salary.

It's interesting how effective the transition to serve r-side development would be. There is a Node on the server side since the web browser became ubiquitous.

And Swift has no stigma of "bad" language than JS.

Does that mean? And Swift has no stigma that it is a "bad" language than JS.

However, Swift has a stigma that it is "dwarfed" Object-C.

Are you trying to compare JS and Swift using node. js?

Swift? Swift is a recommended language for iOS app development, and has already attracted millions of developers.

What I pointed out is not always good. Rails makes Ruby's shadow thinner because many of Ruby use is related to Rails. Swift will follow exactly the same fate. I am not interested in developing iOS front ends and iOS backends in this language. I believe that the web wins, and I think it will be faster than many people.

And I'm sure Google, Apple and Microsoft are terrified of this idea. Why do you think there's been such a renaissance in programming language development and usage recently? What did these companies put in place between 2000 and 2010? Microsoft had . net at the dawn of time. From what I've googled, Google doesn't seem to have put anything in it at all. Apple brought back Objective-C. No development or innovation whatsoever.

Fast forward to the last 5 years. Microsoft dabbled in F# style functional programming, even though their programmers are in business and probably not interested in this functional programming virtuosity. In the last few months they released Visual Studio and . NET on UNIX! Meanwhile Apple is making Swift, a language simple enough to be used by Apple hipster developers building iOS apps, but complex enough to have decent functional elements. Is Google developing Golang and Dart? They decided to use Java for Android, but now they're interested in spending money on language development with names like Rob Pike and Ken Freaking Thompson?

Ballmer called Linux a "cancer". Now Microsoft is hosting Linux VMS for cheap and paying people to develop tools for UNIX users. When Apple launched the iPhone in 2007, they only had 8. 1% of the PC market. But they risked the success of the iPhone instead of maintaining a closed ecosystem and requiring iOS developers to own a Mac instead of giving them development tools for Windows. Now, when that risk has proven to have paid off and they have a dedicated developer community, will they suddenly decide that SWIFT multi-format support is important?

What happens when you can make apps with ReactJS and they cut Google off from Android AdSense money because it's just a webview? What if the iPhone advantage suddenly went to zero because we could build apps with ReactJs and every app could run on every device and look exactly the same on every device? What if we interacted with PCs only through web and mobile and Windows continued to lose market share on PCs and maintain a pathetic 2. 5% on Windows Phones?

This is acceptance, expansion, fade-in, and purer. Microsoft wants you to develop in their language with their tools because they know that . NET is not really a free license. It's definitely a stretch, but they're afraid.

Apple works on the same principle, but their license is actually a permissive license. Either way, I'm sure SWIFT will come later for iOS and OSX developers. SWIFT is a first-class product for Apple devices, but will they ask the community to help them provide support for other devices? Enjoy relying on SWIFT.

I don't understand Google enough. Their goal is probably to bring Golang to Android. As you can see from the recent Android releases, they are invested in providing a very unified Android experience, strongly enhanced by the power of Google Now and Google Services. If you can write your backend in Golang, host it on Google Services, and write the client side as well, that's a very attractive environment. Honestly, when it comes to Google, I believe they act altruistically sometimes. I don't think Go is an altruistic act, but look at how they give back to open source through GSOC. Nobody's done anything close to that. The only big name I think they're doing better than Google is Mozilla.

Speaking of Mozilla, why do you think they developed Rust, servo, and have a bunch of experimental rendering techniques? Because they're betting on the web and they're best at it.

TL: DR: They give us new languages ​​and development environments because they're hanging on for dear life, and I'm not going to accept a flirtation with a closed development environment. State-of-the-art technology is a direct result of truly free software, determined by the community and only the community. I won't accept anything less.

"The web wins"

If you mean "everything but HTML-based sites", then that's accurate. If you mean actual websites, then I don't think so. The web is huge, so it's not dying, but it's certainly not winning.

Native Mobile's "win" is due to the much better UX and overwhelming demand. Natives still use web architecture elements (URI and HTTP) and hyperlink UX in many cases (thankfully). However, as you know, the HTML web cannot keep up with the native experience. This tendency has been set up with a huge amount of investment in IoT alongside Facebook, Google, Microsoft and VR/AR experiences, so unless someone releases new innovative hyper media and clients, it is not popular. It will continue.

"They provide new languages ​​and development environments because they have a very good life, and I don't want to be fascinated by a closed development environment."

I want to applaud your commitment to your openness, I really think so. However, it is recommended that you take a closer look at the world market. Such companies do not last for a lifetime. They are increasing their strength. I attended the WWDC keynote speech. When Swift was announced that it would be an open source, the audience was almost like a kitten. Twitter was on the verge of explosion. These are certain anecdotes, but numbers support Swift's rapid growth.

Apple has vested interests to push Swift everywhere. Probably, as Sun did in Java, you would even do failure with wisdom. Google, on the other hand, was able to do this in Golang, but seems to be secret for unknown reasons. The best theory I have is that Apple is a product company and Google is a technical company. Golang has made good progress by several companies other than Google as a server language.

Golang? Native Mobile's "victory" is usually due to much better UX and overwhelming demand.

I can enter the URL here, and you can see it.

Instead, if you hide the content behind the app, you will not be able to link it deeply. The possibility of installing an app to see the content is much lower. To install my app, your mobile device requires space. Perhaps a password is entered in the installation. It must be compatible with your device. I need to approve my permission. The content of my app cannot be searched by search engine. You cannot add a bookmark to where you are.

I am not convinced that Native Mobile has a much better UX. Certainly, there is definitely its aspect.

What are the pure numbers demand? Again, I'm not convinced. In fact, I am convinced that you are wrong.

Why would you want to type in a URL? It's as easy as getting the URL of the app in the AP store and saving it there. Saving an app on your device is a plus unless you want a web app to be downloaded in full even on a crappy mobile connection. And do you really want your app content to be indexed in external search engines? I like being able to control the permissions of my app. Or why doesn't your web app ask for location, camera, and microphone permissions? I don't have to bookmark where I was in the app: I'm there as soon as I open it.

You probably don't want to type in a URL, especially on mobile, but the fact that every item on HN's homepage is a link to a web page, not a link to a screen app, should point to a crucial advantage of the web over apps. And why having a website is a must and having an app is optional.

If the web and apps were in the same race, I would say the web couldn't lose. Google Docs and the web and apps overlap in some ways, but they serve different purposes.

gt? Why would you want to type in a URL?

Just type in the URL and click. Go to your local library and someone might demonstrate how the internet works.

ggt? You can just as easily get the URL of the app in the AP store and have it there.

You're ignoring the deeper connection I'm talking about. Not just to the app (like CNN. com) but to the actual content on the app, like a specific article.

Is that so? Saving the app on your device is a plus unless you want the web app to be fully downloaded even on a crappy mobile connection.

Web apps can currently store most of their content in the browser. So additional views don't download the full app.

Is that so. Also, do you really want the content of your app to be indexed by external search engines?

Yes, I want the content I produce to be indexed by search engines and actually be found by users. Sure it's not 100%, but I have control over it. If my content is inside an app, please explain how I can get it indexed by search engines.

gt? I don't need to bookmark where I was in the app.

If you want multiple bookmarks, you're not right. If you want to share bookmarks, that's not possible.

Or are you just meditating to be controlling?

It depends on what the application is. If you want to browse content via HTTP, the native app will probably be a stupid idea (unless the browser function is very different like Android, you can only write a stupid website because you do not know what will come out. do not have).

But if the app is something useful (not just content viewer), native is much better.

For example, let's say I'm looking for a house.

Comparing the smartphone redfin. com and the native app redfin. com, there is a clear difference in UX. The native app has far surpassed mobile apps in terms of performance and usability.

This is an example, but many sites that are not just displaying static content.

I feel like people ignore my fundamental points.

On the web, there are things that the current native cannot do. That kind of thing is really important for me.

It doesn't always say that the web is better than native. it's not. But I want to say that the web is better than native. Even mobile.

I am a web architecture great recommender, but I think HTML is dying due to the politicalization of standardization processes and the slow pace of technological innovation.

"If you hide the content in the shadow of the application, I can't connect with that content."

That's not the case. When I enter the URL on the safari on the iPhone, it is dee p-linked to the app related to the domain. I just saw a deep connection between the apps from Siri and others in the WWDC keynote speech.

"The possibility of installing an app to see the content is much lower."

This is also contrary to the data. People love apps and download and use apps like Mad. This year, more than $ 100 billion profits are expected from mobile apps. This is three times that of Hollywood's box office revenue, 10 times that of book publishing, and larger than PC software. Eclipse is not the only packaging / shipping software.

Installing and using apps is less efficient than using a web on a mobile device: I have to deal with a slow web browser each time I use it, but I do it only once. You can do it.

If you have an app, people choose the app.

"The password may have been entered to install. I need to approve my permission."

The same is true for sites and sites that must be registered in OAuth2.

In the case of an app that has an i n-app bill, I do nothing (just log in to the store or use fingerprints).

"I can't search the content of my app with any search engine. I can't add a bookmark to the place I was."

It's interesting. Most of the apps I use can search in the app or on the web. That doesn't mean you use a web browser to make most of the exchanges.

Again, I don't think the web will be gone. If something comes out through an open hyper media, it may change. However, considering how political and rich the HTML standardization process is, it is invisible.

"What is pure numerical demand? Again, I don't think so. I'm convinced that you're wrong."

What kind of scale? In other words, I do not claim infalla here, (a) smartphones are becoming the most common device on the earth, and are more common than toothbrushes, cars and toilets. (B) They all support apps, (c) Most companies with services and content have apps, and almost all users prefer apps than websites. , So I want to say that almost all funds spent on consumer software are moving to native mobile. I am pursuing money and the number of use.

I am pursuing money and numbers. I just saw the WWDC keynote speech, but Siri and others showed a lot of deep connection between apps.

It is true that the app has begun to understand this. But in general, I can easily share links to any website. Sharing a deep link to the content in the app is hard to say that it is a "solved problem."

It is contrary to the data. People love apps and download and use apps like crazy.

Is it true? If you share a link to CNN and HN articles, will you be less likely to be seen? I don't think it's right. I think it depends on the type of content. It just says that there are cases where the web wins a lot.

GT? The same is true for sites that must be registered with forms and OAuth2.

Yes, we have created a situation where others are wrong. But I think you agree with my premise that the web can completely destroy the application.

GT? I am pursuing money and the number of usage.

I sit here and type with ycombinator. com. After spending time on reddit. com, facebook. com, plus, Google. com, and cnn. com. Certainly, there is a version of these apps, but I sit on a desk for 8 to 10 hours a day, and my computer experience is much better than mobile experiences.

He doesn't say that the web is better, but he says that the web is better.

"Really? If I share a link to a CNN or HN article, am I less likely to see it? I don't think so. I think it depends on the type of content. I'm just saying there are some cases where the web wins big."

I'm not saying people will stop looking at links if T is a website, or that most people will avoid it. Web architecture (links) is alive and well.

I'm saying that many sites will have app links at the top of the content as soon as you click on them, and if you find yourself visiting the site a lot, you'll click "Get" to get the app.

Then future links will open the app instead of opening the browser.

"After spending time on reddit. com. facebook. com. plus. google. com., I'm sitting here typing on ycombinator. com.

Right - it's a habit of mine for years. But IME, most people use the app versions of these sites on their mobile devices (except HN - and the UX suffers because of it).

But I sit at a desk 8-10 hours a day and the desktop experience is way better than the mobile experience.

Here we fundamentally disagree. The mobile experience far surpasses the desktop experience for most casual computing tasks. I use my phone and iPad more than my laptop for reading and replying. I use my laptop more for coding and sysadmin.

It is contrary to the data. People love apps and download and use apps like crazy.

I think we're talking differently too.

Right now, if you look at the main page of HN, you'll see this URL: