Java the language is ripe for replacement. Java was a miracle. It was a safe, decently-designed language (thanks to Gosling et al.) that PHBs weren't scared of (thanks to Sun's marketing.) A decent language with modern libraries and no segfaults that your boss will actually let you use is no longer the novelty it once was. Now when people think "Java" they don't think "what a relief, no more C." They compare it to Ruby and C#, which is a little bit tougher standard.
As a heavy Java user it's really frustrating to see how little effort Sun (Oracle) has been putting into maintaining the JDK tools. It seems like only critical security bugs get fixed and everything else just languishes.
I submitted a new javac bug 10 days ago and, although it was confirmed fairly quickly, it still hasn't even shown up on the public web site.
Or look at Bug ID: 5093723 <http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=5093723&#.... It's been open for five years with no fix. I would be willing to pay for a fix but there doesn't even seem to be a way to do that.
> "I just want to remind people that many recent Java success stories are getting lost in all the gloom and doom. Among them are Google Collections, Guice, the JVM languages you referred to earlier, and Android"
People often overlook this strength. There are quite beautifully designed APIs (built on fairly few primitives e.g., monitors and condition variables) around the difficult problem of shared state concurrency (yet the Fork/Join APIs are also used by the Actor implementation in Scala 2.8 and I believe other Actor implementations).
I found it curious that they omitted J9 too. That said, I haven't kept up with what IBM has being doing with it lately and I've always been under the impression it is a closed project that isn't likely to influence the overall direction of Java.
I wish Microsoft would make the CLR open source and OS agnostic. That would make things really interesting.
Sadly, it's extremely hard to see that happening. Write once, run anywhere was a basic attempt at undermining the importance of Microsoft; oddly, it seems to have undermined differences in server OSes. CLR will always be a framework that makes Windows important for basic business reasons.
"InfoQ: This is probably provable (one way or another), but do you suspect that the JVM's use has decreased over all, or that Java proper has decreased while Ruby/Scala /Jython /Groovy have taken up the slack?
Actually none of these languages make Tiobe's top 20. That doesn't mean they aren't important or valuable; they are. But they aren't mainstream, yet. That said, I'm absolutely thrilled by the amount of language research and design going on these days. There are plenty of good ideas out there, and it's marvelous that people are exploring them."
Josh is a bright guy and Java isn't the terrible language everyone may think it to be (though it isn't my preferred language). Ruby is in the Tiobe top 20 though... good to see that he at least recognizes the good that's coming out of other languages.
I am eagerly waiting for JDK 7 with invokedynamic support. I will be good to see the improvements it will bring to dynamic languages like Jython and JRuby.
I wonder if Oracle is going to be conservative or go all out and consider features like tail calls.
I would very much like a very simple, basic, tail recursion to be available in Java. Not tail call elimination in the same way as is specified by Scheme. It doesn't even require the a JVM change, it can be done by the compiler: Scala does this. You could use an annotation (@Tailrecursive) to verify that it happens (or error out).
C does this with "-O2" (factorial is a bad example, unfortunately as there are algorithms that are much more naturally defined in terms of tail recursion):
long factorial_helper(int n, int acc)
{
if (n==1)
return acc;
return factorial_helper(n - 1, n * acc);
}
Looks very similar to what my own assembly would look like. Now with -O4 the assembly becomes completely unrecognizable (which is why higher level languages may be faster than lower level ones).
Continuations would be another improvement possible at language rather than VM level, with same effect (allowing efficient and elegant implementation of some algorithms).
Interesting comments from the guy who sets Java best practices for Google. I do most of my work now in Ruby but the Java platform is still very relevant to me because I can mix my libraries and other people's Java libraries with JRuby/Scala/Clojure applications. I still use Java for AppEngine development (BTW, but not JDO because of the hit of longer loading request times).
I'm not sure how much of that "focus and shift towards dynamic languages" is real, though, and how much is just an illusion created by lots of animated discussion between us geeks in blogs and on forums like this one.
Of course it's hard to judge the progress of a vast and fast-moving industry like ours, but what data there is always suggests that tried-and-tested static languages like C, Java, C++ and newer derivatives like C# and Objective-C still represent the lion's share of development. A few dynamic languages like Perl, Python and JavaScript have established themselves as mainstream and command significant user bases as well.
However, while we hear a lot about things like Clojure and Groovy because they are new and potentially interesting, I see little evidence that many people are actually using them yet. How much real innovation is coming out of these younger languages and platforms is an open question, and it will probably be a few more years before we know what their real contribution will be.
"I see little evidence that many people are actually using them yet." - My experience has been different , even in large Enterprise Java Shops you now have a smattering of different languages on the JVM , albeit all hidden away.
This is primarily because the new languages all come as jar files so taking a cursory glance the tendency is to say - oh that's another Java library , look a little deeper and you find Groovy,Beanshell,JRuby, AspectJ,Scala, Clojure et all, the main scaffolding is still in Java but the interesting parts are being written in other languages.
"in large Enterprise Java Shops you now have a smattering of different languages on the JVM , albeit all hidden away."
That's my take as well. I currently work in a J-A-V-A shop, but Python scripts are working their way in and there's the knowledge that in 5 years, we may need to look at creating our next generation of web applications in another framework.
Sure, but I think it's fair to say that Objective-C only really hit the mainstream in recent years, with the increasing prominence of Apple technologies.
