Posted by batate
on June 3, 2004 at 10:43 AM PDT
My mountain bike trip down Hermosa Creek reminded me of the current state of J2EE.
Over this past weekend, I took the family to Durango. Most of the time, I was a good boy, and dutiful father. No near drownings on the Piedra or Animas. No mountain bike overnights down the Colorado trail. Not even a difficult hike. In fact, my body is in no condition to do any of these things...
But I did break away from the wife and kids for just long enough to do a 22-mile ride down Hermosa Creek. It was stunningly beautiful, and fun. I like to challenge myself. Sometimes, when I get on a bike, it's too hard to be fun. But this trail rocked. It pretty much followed the creek. For you geographically challenged readers, water runs downhill, and so does the trail. I screamed down the hill with beautiful Colorado weather, a babbling stream that's frisky with snow run-off, and the mountains around me on all sides. Spring flowers were everywhere.
That's kind of like the early years of Java. Programming was fun. No pointers, a JVM to take out the trash, simple servlets, easy access to the Internet--all downhill. After c++ and CORBA, life was good again. So the first five years of Java were like the first ten miles of the hike. Sure, I might have to hop off of the bike to hop over a little snow drift, or step through a mountain stream, but the trail was simple. It followed the creek. I wasn't encumbered by a map, didn't have to do excessive training, or buy any special equiptment. (In truth, I in fact rented a full suspension bike, but I'll take a little artistic license here and pretend I was on my companion's old school Trek bike without even a front shock.) Java was like that. Simple and unencumbered.
Then, things got fast. The trail got steeper and more serious. The trail left the creek floor for short stretches, and we followed the cliffs along the creek. It was still fun...we were going mostly downhill. And it was still fast. We were insanely productive. We covered huge stretches of trail between stops. But it was getting a little more dangerous. We followed unprotected cliffs on either side of the creek, just a foot to the right or left. Java was like that, too. Though we whine about the problems with the Java platform, we actually built a tremendous number of serious applications. We had enough success that the whole industry jumped onto the trail. But some people crashed. Still, the risk-reward ratio was right in line.
And then we hit the hill. The creek went through a canyon that didn't leave enough space for a trail, so the trail went up, and we followed. After the first 3/4 of a mile, I thought I was going to die. I was having to get off the bike for longer and longer stretches to walk through rocky terrain that was difficult to peddle. It was definitely a whole lot less fun. Truth be told, we could have actually cut the trip short, but the promise of the trail held us captive. I'll admit that my next trip down that trail may be in a kayak.
Finally, we get to the point. This is where Java is today. After EJB and XML Schema and Namespaces and WebServices, we're getting maxed out. Some of us will spend all of our energy and die, right there on the trail. Others will stare longingly at the creek below, and wonder if there's not a better way to get down the trail. AOP, anyone? Maybe something a little bit less dramatic, like one of the dynamically typed languages? Some will take that uber-programmer and muscle their way up the hill. Others will try to find a motor to slap onto the mountain bike, like MDA or other frameworks that try to do all of the heavy lifting.
This is a serious point in Java's history. How will we handle the hill? Will we handle the hill? Some say that we're beyond a point of no return, and the creek will stay in the gorge, fully independent of any trail. It's a tall mountain that the average Java programmer will never climb. Others say that we just have to grind it out, and there are better times ahead.
I think that we've got to simplify. My book, Better, Faster, Lighter Java, comes out in a week or so. It's at the printers, and you can order it today. My premise is that you can make it easier to scale the hills by lightening your load. Most applications are simple, and we don't need to go through artificial steps to complicate them.
The real trail tapered off, screamed down a hill, and went right back down to a natural rhythm of a little up, and a whole lot of down. I think that the Java community actually has a fork in the trail. On the left is a narrow but simple trail that goes back down to the creek. On the right is a huge, wide trail that will accept any type of bike, but it goes straight up. We can abandon the current trail, of heavyweight EJB, and full-blown J2EE containers, but we've got to lighten up. We can use the work that we've done, and the knowledge that we've accumulated, and ride the momentum down for a while. Some fantastic simple, yet sophisticated frameworks lead the way. Spring, Hibernate, Kodo JDO, and IDEA are a mix of open source and commercial products that do it right. Or, we could get overly ambitious, and die on the hill. Think Tomcat with Spring. Think Pico. Maybe some developers are strong enough to make it up that hill, and many will doubtlessly try.
I'm going to lighten up. I'm ready to go downhill again.