Ask HN: Is Java Dying?
What do you think the future of the Java Language would be? What's a good replacement? Would the replacement run on the JVM?
14 comments
The JVM is going strong and will stick around as a common platform for a variety of interesting languages for quite a while. Java the language is on the downward slope of the hype curve, but still has a huge base of existing installs so even though it is bleeding it will take a while for shock to set in. IMHO, the biggest future surprise will be how Java turns out to not be the COBOL of the future. COBOL is an antediluvian artifact that got the job done but lingers on in constant maintenance mode until enterprises bite the bullet and spend capital to do a wholesale replacement of the legacy mainframe system. The JVM is what will eliminate this refuge and last bastion for Java the language; the fact that other languages can run on the JVM and interface with old Java code means that incremental replacement is an option, and that option will be exercised more and more until the Java language becomes a distant memory.
No, it's not dying. The installed base is huge - it's currently the most popular language on http://langpop.com . Perhaps, if Oracle messes things up enough, its rate of adoption for new projects might conceivably slow a bit.
Indeed, there's no way you could say its dying.
That said, Oracle has rolled a grenade into the greater Java ecosystem and we're waiting to see what happens next. If "Oracle messes things up enough" I suspect we'll see a lot more than a slow down in new projects, or at least the sort we're interested in.
That said, Oracle has rolled a grenade into the greater Java ecosystem and we're waiting to see what happens next. If "Oracle messes things up enough" I suspect we'll see a lot more than a slow down in new projects, or at least the sort we're interested in.
I've been working with Java since '98. I worked for Sun Micro, then Netscape, then the Sun-Netscape alliance, then iPlanet. I also worked for Oracle before I joined Sun and would have lost a lot of money on the wager that Oracle would buy Sun one day.
Java is not dying. The enterprise footprint alone will sustain the community (and developers) for years. And there is still plenty of innovation rolling out (SpringMVC is a good example) and I see some great angles for VMWare.
That said, I finally broke down and started learning Ruby last week (after the Oracle litigation announcement). All I can say is...wow. In less than a week, I was pretty comfortable with RoR (OO and MVC felt right at home). Over the weekend, I built an app that I'm actually proud of and planning to share with the HN community for feedback this week. The community is friendly and I'm blown away by the ecosystem (Heroku is ridiculously great).
Bottom-line: I built my home and fed my family with Java over the past decade and feel almost a sense of loyalty. I thought I was pretty fast in Java, but my productivity is up in RoR, even with the learning curve. I've loved every minute and have decided to build subsequent projects using RoR unless I run into some type of crazy roadblock along the way.
The early adopters that embraced Java and pushed it before it was proven (.97b anyone?) are the same people that love trying new things now. That other options are so enjoyable is a bigger threat to Java than Oracle. Oracle might have just provided the nudge people needed to see for themselves.
Java is not dying. The enterprise footprint alone will sustain the community (and developers) for years. And there is still plenty of innovation rolling out (SpringMVC is a good example) and I see some great angles for VMWare.
That said, I finally broke down and started learning Ruby last week (after the Oracle litigation announcement). All I can say is...wow. In less than a week, I was pretty comfortable with RoR (OO and MVC felt right at home). Over the weekend, I built an app that I'm actually proud of and planning to share with the HN community for feedback this week. The community is friendly and I'm blown away by the ecosystem (Heroku is ridiculously great).
Bottom-line: I built my home and fed my family with Java over the past decade and feel almost a sense of loyalty. I thought I was pretty fast in Java, but my productivity is up in RoR, even with the learning curve. I've loved every minute and have decided to build subsequent projects using RoR unless I run into some type of crazy roadblock along the way.
The early adopters that embraced Java and pushed it before it was proven (.97b anyone?) are the same people that love trying new things now. That other options are so enjoyable is a bigger threat to Java than Oracle. Oracle might have just provided the nudge people needed to see for themselves.
This discussion often happens around JVM languages or dead-end projects that will need to be maintained until the end of the universe. If that's all that remains, then yeah, I'd say Java is dying, because people won't be actively writing new, interesting code in Java.
A couple of projects that suggest Java may not be dying (quite yet) are Roo and GWT. Roo is a spring-based framework that is very Rails-esque in terms of getting a web application up and running very quickly, though I haven't looked through much of the code. GWT is (as almost everyone knows) a very powerful framework for creating rich internet applications. It's a really major departure from page-based technologies.
Aside from that, there are still other projects like Guice, Gin, Struts 2 (and other MVC frameworks), ORM type work, and so forth (many of which are often used with GWT and are rolled into Roo).
While I've looked into these, none of them are enough to make me want to make a switch (back) from Ruby to Java programming. But I do think there's still enough going on there that I really wouldn't say Java is dying as a language that people use to build web applications (and I don't mean purely as a JVM or legacy code thing...) Some of the top innovation is still happening in Javaland.
A couple of projects that suggest Java may not be dying (quite yet) are Roo and GWT. Roo is a spring-based framework that is very Rails-esque in terms of getting a web application up and running very quickly, though I haven't looked through much of the code. GWT is (as almost everyone knows) a very powerful framework for creating rich internet applications. It's a really major departure from page-based technologies.
Aside from that, there are still other projects like Guice, Gin, Struts 2 (and other MVC frameworks), ORM type work, and so forth (many of which are often used with GWT and are rolled into Roo).
While I've looked into these, none of them are enough to make me want to make a switch (back) from Ruby to Java programming. But I do think there's still enough going on there that I really wouldn't say Java is dying as a language that people use to build web applications (and I don't mean purely as a JVM or legacy code thing...) Some of the top innovation is still happening in Javaland.
Theres also grails, but even that is a bit clunky at times. I was ready to jump ship to Ruby land and then I discovered Play framework (www.playframework.org). It goes against many Java conventions but thats a good thing! Development time definitely lot faster! The only downside is that the community around it isn't quite as big as the other frameworks (hopefully that will change!).
Not a chance, though I wish it would.
It was born in sin, marketed beyond merit, and has a horrible culture of bureaucracy and overhead.
It was born in sin, marketed beyond merit, and has a horrible culture of bureaucracy and overhead.
We use JRuby with a bit of Java. The JVM is a very interesting place to be, with relative safety in the language and runtime, C-ish performance, and access to lots of lovely libraries. And you don't have to program in Java to use it.
Here's how we went from Ruby to JRuby + Java: http://bens.me.uk/2010/jruby-in-practice
Here's how we went from Ruby to JRuby + Java: http://bens.me.uk/2010/jruby-in-practice
We're writing our startup using Java and Google App Engine, so we're obviously stuck in the middle of something. But in the end, we're confident that Java is giving us the best performance for what we're doing (big data), and no matter what happens between Oracle and Google, there's still an Open Source programming language that can't be taken back.
Yes, that's what I am thinking, with Larry's style of management/acquisition style, the future is uncertain for us Java Developers
Java is now deeply-rooted in enterprise software, and in this field, once a project have been developed, radical technology changes occours on a 10-year+ basis. You won't believe how many big businesses still relies on Cobol development for manteniance. So, imho, Java isn't going away anytime soon.
Some information at:
http://www.michaelnygard.com/blog/2010/04/the_future_of_soft...
"The Future of Software Development"
http://www.michaelnygard.com/blog/2010/04/the_future_of_soft...
"The Future of Software Development"
I had dropped java/j2ee since 2006 and moved to ruby and erlang community. I hated the deployment of j2ee application and the dependencies of jars, i known ant/maven would solve the problem somehow, but i enjoy the way ror/otp does
It stands pretty strong. Lot of enterprise softwares are Java based. Most schools teach this as their main language. So, at least I don't see it happening anytime soon.
Yes! Still there will be a remaining huge code base to be maintained.
dying? no. growing? no, not really, either. or, at least, nowhere near the pace that it had to get to where it is now.