As someone who's just moved to Java myself... I suggest starting here: http://grails.org the Grails framework will get you started in Java based web development and you can learn more diving deeper into Java as you get more and more proficient. Grails is built on Spring which is a Java Enterprise platform but Grails saves you from having to learn all those "enterprisey" bits. Instead Grails picks defaults that are the most sensible for web development.
As you get better you can graduate to adding your own features to the framework or even beginning to use the more advanced Java frameworks outside Grails tying everything back together using Spring's IoC and Java JNDI. Or you can keep things light weight and web2.0 by using various remoting technologies inside Grails. It's a quick way to learn that doesn't put training wheels on you.
Get going in an afternoon... scale out using Java as your project grows.
... looks like I'll be packing up my bags and leaving the Perl kingdom for the lands of Java. Forever this time. Not entirely from choice but from pragmatic concern over my career. All the cool vibe has rubbed off the Perl kingdom. Most of it dribbled onto the Rails zealots. I'm beginning to think the Django fans may have gotten some of it too. But, I'll be making money by day doing huge scale Java applications for medical, clinical, and biotech applications from now on.
My only hope is that the lands of Java will be invaded by the "cool vibe" again
For reasons that are just becoming apparent Groovy/Grails is not as popular in my region. There is no technical reason in fact if you are running in a Java environment you'd be myopic to not pickup Groovy... and you would be well served to at least build a demo app in Grails. This region however is a Python/Jython stronghold. I didn't know why, but, I just learned that a major Jython contributor is a member of my Jug. That explains a lot. I know RedHat is primarily a Python shop so Java to Jython in the region surrounding the Red Hat campus makes perfect sense. I suspect that Java running Jython code would be the main way Red Hatters will use Java going forward. They are big on Python. And so too are some parts of Google apparently...
I'm beginning to learn that political forces more than technical merit can shape the technological landscape. It's a sad thing to learn because we techies want to believe we're more like mathematicians and only proof and merit rule our decisions. It's not true. For example, Slashdot is among the last strongholds of the Perl kingdom. Outside these hallowed walls the PerlBe are under constant threat.
Each of the major technologies have their own little world they live in and very few technologists actually have a realistic view of the whole technical landscape. I've found that http://www.indeed.com/ is an invaluable tool for looking into the reality of what the tech landscape was. The fact is companies hire for the technologies that they were using three to five years ago. Most do not hire for what they will use or what they are going to use. The hiring managers simply don't know where tech is going tomorrow. If the job is on a job board it's probably a legacy application that's being supported and that means it's a few years old.
You start to bundle these technologies into sets. Perl and C go together better than Perl and Java for example. Python, Perl, C and C++ all roll together for example. So to me it looks like if you were after sheer job security you should learn C really well. In fact from this data C is the unassailable emperor of computing languages. And even though I'm packing my bags for Java land that's the truth of the tech landscape
Truth hurts. I'm going to have to go and play with some open source C again at some point. I figure if I can cross between Java and C I'll be a sight better off than someone trapped on either island. Both of these are pretty big islands now though.
I really do.
Each kind of karma point would be aligned along it's own axis in N-Dimensional space. The resulting total karma score would be expressible as an N-dimensional manifold. You could then filter the resultant N-ary manifolds using your own total karma manifold. The results would be utterly transparent to you... instead as you accumulate karma points it informs the shape of your own manifold that is being projected onto the comment moderation system results.
In short you would experience the Slashdot that best fits your perceptions of reality.
All I ask is a chance to prove that money can't make me happy.