7 ms·
Kotlin though. Everything you said, but a joy to write instead of a chore.
by euyyn 8y ago
Kotlin though. Everything you said, but a joy to write instead of a chore.
- pjmlp 8y agoI am betting Kotlin is the new Groovy. Lets see where it stands 5 years from now, specially if Fuchsia actually gets released.
- ptx 8y agoKotlin is different because of the focus on tooling, which is the advantage Java still had over all the dynamically typed JVM languages. Also, Kotlin/Native is in beta now and could target Fuchsia (compiling AOT to native code using LLVM).
- pjmlp 8y agoWhat tooling? Being forced to use InteliJ, without any proper support on Eclipse and Netbeans? Still not able to use several of Android Studio features available to Java, like incremental compilation and slim APKs? Kotlin advocates seem to forget JVM will never be rewritten in Kotlin, the language is just yet another guest, with the usual syndrome to wrap existing libraries, having to take care about FFI for Java access, not having all features taking advatange of the latest bytecodes, e.g. lambdas implementation. As for Kotlin/Native, there is nothing to worry about versus what Go, Rust, C++, Dart, D, Nim offer in terms of performance, libraries and in some cases tooling. Having to buy CLion for a graphical debugger isn't a selling point versus the established alternatives. Fuchsia is being written in Go, Rust, C++ and Dart, with the team now hiring for node.js support. https://www.androidpolice.com/2019/03/19/google-working-to-bring-javascript-app-support-to-fuschia/ https://www.androidpolice.com/2019/03/19/google-working-to-b...
- coldtea 8y ago>What tooling? Being forced to use InteliJ, without any proper support on Eclipse and Netbeans? Seeing that both Eclipse and Netbeans are now more or less dead (and speaking as a long time Eclipse user, from the very first version to around 4), yes, first class vendor-direct InteliJ support is more than enough. And more than most languages (including Groovy) ever had. >As for Kotlin/Native, there is nothing to worry about versus what Go, Rust, C++, Dart, D, Nim offer in terms of performance, libraries and in some cases tooling. IMHO, Rust will always be kind of niche as hard to tackle, Dart we'll see, D never went anywhere, and Nim will remain niche, it's a little too idiosyncratic to catch on. Kotlin is already more popular than all of the above except perhaps Go. >Fuchsia is being written in Go, Rust, C++ and Dart, with the team now hiring for node.js support. Fuchsia is still vaporware or at least irrelevant. It's not even in the market yet. And the fact that it's written in 4 (and looking for 5th) languages doesn't really bring much confidence.
- pjmlp 8y agoGroovy had Netbeans and Eclipse support, which was dropped when it started fading away. We move in different worlds, no InteliJ installations around here. I remember when Groovy was popular, with every JUG in Germany having weekly talks and Sun talking how the next JEE revision would support Groovy for writing beans. Popularity doesn't write software.
- olavgg 8y agoI don't understand how Groovy is fading away when it is getting more and more popular. Groovy is also very similar to Java. There is no learning curve, you are immediately creating value.
- pjmlp 8y agoMore popular where? Had it not been for Google's adoption of Gradle for Android and it would have been left for Grails maintenance projects. And even then, the pressure for perfomance has been so much that it is being superceded by a Kotlin based DSL. How to improve Gradle has been a common talk at every Android conference since Android Studio has been introduced.
- olavgg 8y agoEverywhere? Google Trends says Groovy is just as popular as it was 10 years ago. The number of downloads has increased significantly but it is not because of Gradle, as Gradle bundles Groovy. It is also currently placed #16. on TIOBE Index And last, Groovy with compile static is in most cases, just as fast and memory efficient as Java. Some people always talk down Groovy, yet the numbers speak for them self. Before Kotlin, Groovy was attacked by Scala developers. The major thing Groovy has over both Scala and Kotlin is strong similarity to Java syntax which makes the learning curve non-existing. And that is a bigger deal than a random cryptic/confusing programming expression. Code readability is very important.
- pjmlp 8y ago
- coldtea 8y agoGroovy was a random JVM-based dynamic language with no major company support and no special tooling. Kotlin has IntelliJ (and thus a great IDE) and Google standing behind it, and Steve Yegge's nod of approval.
- pjmlp 8y agoNot every Java shop is going to drop Eclipse and Netbeans just to make JetBrains happy. As for Android, lets see what happens with Fuchsia. The overwhelming majority of Java developers don't even know who Steve Yegge is.
- coldtea 8y ago>Not every Java shop is going to drop Eclipse and Netbeans just to make JetBrains happy. True. They will drop them because Eclipse has been faltering for ages and has been dropped by IBM, and Netbeans has always been a subpar unloved stepchild used by the kind of devs that don't know better and think SlickEdit or Notepad++ are great editors. >The overwhelming majority of Java developers don't even know who Steve Yegge is. That's on them.
- pjmlp 8y agoI guess Fortune 500's have missed that memo, including IBM own subsidiaries. Lets talk back in 5 years from now. I am betting Kotlin would become yet another language that happens to compile to the JVM, after the Android honeymoon goes away.
- coldtea 8y ago>I guess Fortune 500's have missed that memo, including IBM own subsidiaries. They probably did. They're late for all memos. The Fortune 500 is not were you'll go to gauge adoption. Heck, half of them still have IE6 only apps. There are hundreds of thousands of companies, and just 500, well, Fortune 500. Apparently you use the most established and boring development-wise of companies (Fortune 500) to prove Kotlin is not used widely, but you're OK with vaporware like Fucscia that's not even beta yet as an argument that Dart and co are. >I am betting Kotlin would become yet another language that happens to compile to the JVM And I bet you're wrong. Let's see in 5 years. We were on HN 10 years ago, we'll probably be here then. I'll set a reminder.