Today, I forgot my laptop at home and therefore didn't get a chance to do any real-time blogging. It was quite nice in fact. I realized I forgot everything about halfway into my drive (30 minutes) and decided it wasn't worth turning around. Therefore, it was a day for just sitting back and soaking it all in.
This was quite a conference. Never have I seen so many great minds and influential Java Gurus in one place. The speakers were awesome, and to make things even better - I got to meet and shake hands with most of them. Who says the community in Java sucks? I'll tell you what - the Java community in Colorado is rockin'! The online community in Java might be a little terse - but the face-to-face community is a different story.
Ugh - I just had one of those "write a huge long paragraph and then lose everything" moments. Who was at fault: a combination of IE and Roller. I accidently hit the back button and Roller kicked me back to the login page. When I clicked the forward button - Roller/IE didn't return me to my post -> hence I lost everything. Sometimes I hate webapps. Time to finish the rest of this post in an HTML Editor where I can hit Ctrl+S all the time.
The first session I attended this morning was Dave Thomas's Herding Racehorses and Racing Sheep. Thanks to
Carl Fyffe for recommending it. The first part of it was about how developers
have different levels of expertise and how they learn. Novices are productive and learn well if they have a specific set of instructions. This is in contrast to Experts - who are more efficient if they're given freedom and lack of
rules. Most folks in the room considered themselves Experts. Dave estimated the room's average was 4.5. From there, we
proceeded to talk about the fact that most developers think they're smarter than they really are. Lastly, Dave touched on
Personal Knowledge Investment and how we all need to take time to invest in our own knowledge. It's your #1 defense
against offshoring. You know its working if you still have a job at the end of the year. You know it's really
working if you're doing what you want to be doing, programming in your preferred platform. This talked rocked and I
highly recommend it.
After that I attended Rick Hightower's TDD for the Web Tier. It was a good talk and I learned about jWebUnit. This tool is similar to HttpUnit, but it simplifies the API. It has the same functionality as Canoo's WebTest - allowing to easily
test interactions on a web page. It looks like a great way to test your view layer, though I prefer Canoo. The main reason is
because Canoo Tests are a bit simpler to write - especially if your QA folks are writing them. I had some success with this at Comcast - where our QA guy actually wrote quite a bit of XML to test our webapp. Rick also covered
Cactus fairly extensively (with examples) and a I learned some things about Cactus I didn't know. Cactus's beginXXX is for setting up client parameters, testXXX is for executing and testing
server-side code and endXXX is for testing the client side response (i.e. an HTML page). Rick definitely has a lot of experience with test-driven development and has motivated me to learn to use my IDE more
for testing. It's simply faster and therefore will make me a more efficient developer. Rick was saying that he
uses AppFuse with MyEclipse all the time. He says he hardly ever has to restart Tomcat and he can easily run XDoclet on a single file. The update for Eclipse 3.0 M9 is due out in a week - I might have to make an effort to get AppFuse to be MyEclipse-friendly then.
Julie and Abbie showed up for lunch and Abbie got to run around and wink at all the boys. It was cool being able to introduce them to all my "Java buddies."
After lunch, there was an Expert Panel Discussion where we talked about all kinds of topics: Hibernate, Personal Knowledge Investment, Tools/IDEs and of course, .NET. From there, I attended Rick's talk on AppFuse. It was great to hear Rick's perspective of AppFuse. When I talk about AppFuse, I tend to talk a lot about the
reasons I created it and how it came to be. Then I talk about the technologies it uses. Rick talks more about
his experience with it and how you develop applications with it. He had a lot of code to show and the talk went really well. Then again, I'm biased. I've never heard Rick speak before this weekend, and I have to say - he's a lot better than I expected. He has a lot of humor and experience in his talks and was quite enjoyable to listen to.
I hope to work with him someday, that'd be a good time. If anyone is learning more about AppFuse, and you're local - I'll be talking about it at the Denver JUG and Boulder JUG next month. I also plan on releasing AppFuse 1.5 and AppFuse Light this week.
For the last session of the day, I attended Rod Cope's Easy Enterprise Applications with JBoss, Hibernate, AspectJ, and XDoclet. Rod is going to be my new boss in a week, so it was definitely cool to hang out with him this weekend. Rod is a good speaker, but his presentation wasn't teaching me much, so I left. He was talking about how extracting 75 lines of Hibernate code (many try/catches, lots of transaction/rollback code) into aspects can save you a lot of time. I don't use aspects, I use Spring - and my Hibernate methods are typically 1-3 lines of code. I'm sure if I'd stayed, I could've learned more about AspectJ - but frankly, it doesn't interest me that much.
I snuck out of the room and went to the 2nd half of Spring in Action by Craig Walls. I met Craig
for the first time at this conference and I thoroughly enjoyed talking with him. He seems to be a very easy going guy and he definitely knows a lot about Spring. I learned some different ways of writing controllers - extending the Abstract Controllers, rather than using the Controller interface and SimpleFormController. It was interesting to see how
flexible Spring is and how his way takes about the same amount of code as the way I'm doing it. It was a good talk and I look forward to downloading his sample app. Hopefully he'll post a link soon on his blog. I'd be happy to host a demo.
What a great conference. I left with a headache today because my brain was so full.
. I also made improvements to Spring and its context file loading so you should be able to