JavaOne, Day Three

IBM Keynote

I slept late and skipped the IBM keynote presentation. I wanted to go, but I had to get some rest. My BOF was scheduled for 8:30 that evening and I didn't want to be yawning as I was doing it. As it turns out, the "IBM Keynote" was really an "Eclipse keynote." More details here.

Collaboration

So the day started out at a slower pace. Geertjan asked me to be signed on to the share.java.net server with the collaboration tool in the NetBeans IDE. At about 9:00 we did a quick demo of that feature. I was in my hotel room and he was at a restaurant somewhere nearby demoing his Wicket functionality for some Wicket developers.

Reality Bites

When I did finally make it over to the Moscone Center I immediately got in line for the Ajax, Web 2.0, SOA panel discussion. While waiting I struck up a conversation with a senior level developer from a big-name manufacturing company. I have to be honest: the conversation was a bit sobering. He's struggling with legacy systems that cannot be migrated quickly or easily. The most telling comment he made: "We hope to have all the applications migrated from J2EE 1.3 to J2EE 1.4 by the end of this year." So all the talk about Java EE 5 and Ajax, etc. is interesting to him but not relevant anytime soon. I've talked with other developers who have told me similar stories. And while there are abundant examples of organizations that do keep pace with the latest JDK and J2EE releases, the reality is that it is not always possible. Sometimes, reality bites.

The panel discussion did not bite. It was actually pretty good. Edward Ort has a good recap of it here.

JAXX

After a quick lunch I headed over to the JAXX presentation by Ethan Nicholas of Yahoo. I had attended Ethan's session last year on Swing development and I also enjoy reading his blog. His JAXX framework is an open source project that uses a declarative approach to Swing GUI development. In other words, specify what you want in an XML file and the JAXX compiler will turn that into the necessary .java file. The .java file is then just compiled with javac and then is deleted. It looks intriguing for three reasons: in some cases it looks like the declarative version is more compact than the equivalent Java source files (especially if the defaults the framework provides are adquate), a CSS approach is used for style information, and Ethan likes it. That last attribute is, admittedly, a bit of fuzzy logic. But whether I agree with him or not, I have a great respect for Ethan's ability as a Swing developer, so if he likes something it at a minimum is worth considering. I'm a bit turned off by the fact that there is no visual GUI builder for it, but JAXX is still worth investigating.

Debugging Across Tiers

Next up: Debugging Across Tiers. This was done by three members of the NetBeans development team: Roman Ondruska, Martin Entlicher, and Adam Sotona. It's worth noting that Adam was asked at the last minute to fill-in for Petr Suchomel. Petr had developed most of the demos for this session but a conflict came up: he had to go to a rehearsal for the next day's general session. Adam did a good job of filling in. The basic idea behind the presentation was to show how a single development tool (in this case the NetBeans IDE) could be used to debug multiple applications at one time. The demo showed an application being debugged as it ran on a cell phone. When the application called into a web service that was running on a separate machine the debugger could handle that too.

Back In the Booth

Then it was time to work the NetBeans booth. Traffic was heavy at times, but we also had some lulls in activity. One demo I did was for a very skeptical vi user and he seemed less skeptical when the demo was over than when it started. He said he might install it and give it a try. Ludo stopped by and wrestled the mouse away from me long enough to turn off the IDE's support for Tomcat. Ludo prefers Glassfish. :-) After he left I turned the Tomcat support back on - the NetBeans IDE is a big tent.

Meeting With James Gosling

I attended a customer meeting after working the booth. Charlie Hunt and I accompanied Warren Sheaffer of St. Paul College to a meeting with James Gosling. The topics discussed were all over the map: development tools, teaching methodolgies, and a bit of ice hockey.

BOF 1: Java Champions

After that it was time for supper. I bought a meal instead of waiting for the after dark party to start. That ended up being a good idea because the food lines were pretty long once the party started. I wandered around the party for a while and chatted with a few folks. Then I headed over to BOF number one of the evening: Java Champions. Matt Thompson described the history of the program and then turned it over to Yakov Fain who described how to be a Java community leader. Then they opened it up to questions.

BOF 2: Memory Leaks In Java Applications, Different Tools for Different Types of Leaks

I left shortly after that to head over to the room where my BOF would be held. For those of you who attended (it looked like about 300 folks), I will have a separate blog entry (real soon now) that recaps the presentation, complete with the slides. So watch this space.

BOF 3: Creating an IDE for Your Own Favorite Web Framework

After that it was time for Geertjan's BOF with Petr Pisl where they described how to build IDE support for a web framework. There are, needless to say, a rather large number of Java frameworks for developing web applications. IDE suppport for them ranges from non-existent to comprehensive. Geertjan and Petr have built some basic support in the NetBeans IDE for the Wicket framework and their BOF described the process involved in doing that. More here.

The best part of that BOF, however, was that I got to meet Manfred Riem. Manfred is a long time member of the NetBeans community and has made significant contributions (including translating the IDE into Dutch). He was one of the community members who won an award at last year's Net Beans Day. He had attended my BOF and introduced himself after it so we were able to chat on the walk over to Geertjan and Petr's BOF.

BOF 4: JConsole

At that point it was 10:30 PM. I had not attended a 10:30 PM BOF all week, but I really wanted to see the JConsole BOF. So I went over and am glad that I did. There are some interesting features coming in the next release of JConsole - for more check out Mandy Chung's blog entry. As a bonus, I got to meet Alan Bateman who has taught me some lessons through his excellent blog and I also got to meet Jean-Francois Denise and Leif Samuelsson of the JMX team.

And with that, the day finally came to an end.