With 7 concurrent tracks (see below), it has been hard to choose.Key Note - by Ted Neward
Ted's opening key note was excellently presented, talking about the The Next Big Thing. And his start with "I have no clue!" and the promise that he was done in 60 seconds... But No! He went on talking about the emergence of new technologies and how long that cycle really is. "Objects" took 25 years, but AOP took 5-7 years, and Ted predicts that the cycle will get shorter, as the feedback loop between academia/researchers and the practitioners (us). He pointed out that JVM/CLR helps language designers, many tools are available for language design. So, I think what he is trying to say is that we can expect to see further explosion of languages available to us.
One could argue that Qi4j is a intermediary step between 'traditional' languages and the new wave that Ted thinks will happen.
First Session
I went to listen to Jimmy Nilsson presenting Domain-Driven Design at a fairly brief level. Running some analogies, such as Rugby "messiness" vs Soccer "elegance". I like Jimmy's book and I bet he is great guy, but frankly speaking he is not the best presenter. It wasn't really, really bad, but I walked away wondering why I went in there in the first place.
Second Session
Rickard Öberg was up with a Qi4j session, and after a quick intro to Composite Oriented Programming and Qi4j, he went straight to the point. Showing how the Spring/Hibernate dddsample.sourceforge.net project look like in Qi4j. Those of you who follows the commits to our source repositories knows that code has been written over the last few weeks. He shows how API concerns, such as constraints are deeply embedded in the original code, where the Qi4j way will expose that in the public view of a Composite. He shows how Qi4j provide Structural enforcement throught the layers and modules, and even shoot up the Visualizer, drilling down to Concerns on methods. All in all, Rickard did a good job. Not bogging down too much in details, not too cursory, and touching just enough on UnitOfWork to wet the apetite. Well done - again!
Third Session
Dan Bergh Johnsson is giving me a entertaining presentation of Value Objects. No, not the Data Transfer Objects and container of data opeated on by services, but the real, original defintion of Value Objects. Immutable, behavior-rich and type-safe. He shows a fairly complex set of code, and shows how a relatively simple refactoring makes the code much simpler yet encapsulates the same business logic, and finding that in the right places. A good presentation, and good manifestation of our vision of "all objects are not the same".
Fourth Session
I decided I needed a break, so nothing here...
Fifth Session
Eric Evans runs through Strategic Design in Domain Driven Design. The room is full, so I decide to have a look at "Compiled DSL Languages", but that turned into a course in Accelerated Graphics programming, so 20 minutes later I squeezed into Evans room anyway.
Evans pace and humility is excellent and he tries to give concrete advice on how to get from Ball Of Mud to DDD Nirvana, both in terms of people as well as existing systems. Don't rewrite... Stablize and transform the Core Domain via an Anti-Corruption Layer. All-in-all, I liked the part I saw.
Fifth Session
It hasn't started yet, but I think I will go and listen to my friend Bruce Snyder talking about ActiveMQ. The report from that has to wait until tomorrow.
- Niclas
Tracks
- .NET
- Agile Ways
- Aspects of Leadership
- Cool Languages
- Domain-Driven Design
- Java
- Mobile 2.0
