It seems that lately lots of writers have been bashing Java. Some people seem to think that Java has stagnated, and that the "hyper-enthusiasts" have left. Well, the rest of us are just quietly coding on a platform that is more exciting than ever. Other technologies like Ruby, PHP, .NET, or AJAX might get more hype and acclaim lately, but hype aside, Java still produces the majority of innovation in the software profession.

You can get a sense of Java's health by simply looking at how many projects are out there using Java. Pop on over to Sourceforge and look at the number of projects by language: http://sourceforge.net/softwaremap/trove_list.php?form_cat=160.

Certainly Sourceforge is not the final measure, but it's a reasonable reality check. Anyway, you'll find 17300+ Java projects, the most of any language. That's more than PHP (12675), Ruby (425), and .NET (3094 C#, 347 VB.Net) combined.

Because of its prominence, Java gets a lot of attention and with it much criticism, some of it valid. What many may not realize is that some big breakthroughs have arrived and that the Java development landscape is solving important problems. Let's take a view of where Java is going to go in the next year or two as these ideas gain traction.

The first big arena of innovation is the addition of scripting support. Some people rightly claim Ruby or Python is better the Java for some tasks. Groovy and Beanshell solve these same problems and will become a standard (in the JSR sense) part of the Java stack. Each offers something better than standalone scripting. Both integrate into a truly mixed environment with compiled bytecode and interpreted scripts interoperating smoothly. Beanshell's syntax offers as little surprise as possible for the Java developer and Groovy gives a Ruby-like syntactic efficiency, but can also be compiled to pure bytecode and reused seamlessly, a big improvement over JRuby or Jython.

Second, the days of heavyweight frameworks are gone. J2EE-style EJBs are out. Now, the lightweight frameworks like Hibernate, Spring, and HiveMind are the future of Java. The heavy Struts of the past is being replaced with two new directions: Struts Shale based on JSF and Struts Ti is a wholesale merge with Webwork. These projects share a common idea that the framework should stay out of the way of your object model. EJB3 takes the same view.

A third major innovation where Java leads is aspect-oriented programming (AOP). AOP will unclutter your code by moving "cross-cutting" concerns like logging and security to transparent method interceptors. AspectJ is the Rolls Royce for AOP, but Spring has lightweight AOP themes as does the new JBoss.

Fourth, the PHP crowd had a legitimate gripe that banging out a simple web page took too much work if you were using JSP and Struts. But our templating solutions like Velocity, Freemarker, and now Groovy don't make you choose between ultralight text generation and powerful mid-tier solutions. Groovlets should make for some rapid servlet solutions. For those people who really like PHP, keep it: Java 1.6 makes it the reference implementation for the new web scripting language bindings, so you can do your web views in PHP and use Java for model and controller.

Finally, Java's traditional strength in the tools arena just keeps increasing. Eclipse is the best IDE on the planet, and it appears that Eclipse has reached critical mass like Ant and JUnit did, in that people simply expect solutions that extend it. Eclipse's team coding capabilities for CVS and Subversion (via the Subclipse plugin) are first-rate. The refactoring and debugging support are top notch. Integration of standard tools like ant, compilers, JUnit, and javadocs are about as good as you can imagine.

So perhaps the hyper-enthusiasts have left and the buzzword mongers will proclaim that Java is dead. If Java is "dead," it's only because it was laid to rest so that a better, lighter, more agile, form of Java could replace it, so Long Live Java!