« Terracotta Announces Partnership with Interface21 to Create Transparent Spring Clustering | Main | Conflicting Galactic Forces in Java »
January 6, 2006
2006, The Year of the True POJO Entity
posted by ari
Spent the last month of 2005 traveling. I was in London to speak with Rod Johnson--learned a lot. I was in Miami for the Spring Experience. I was in Belgium for Javapolis. I felt it a great honor to hang out with Jonas Bonér and help him get Terracotta's message out there. In Miami we demonstrated our new product, Terracotta for Spring. At Javapolis, we had two 15 minute sessions where we did tech-heavy demonstrations as well.
I learned quite a bit about people's relative level of interest in our view of clustered Java. Jonas was able to draw over 200 people to each of his 15 minute sessions in Belgium. We showed clustered Java apps with no code changes (drop Terracotta in and out, etc.). Not only did people come to our sessions, they approached both Jonas and me after the sessions and asked tons of questions, and had really hard problems that they were trying to solve with either JCache or JMS (yes, a messaging solution to a distributed / shared state problem).
In general, I feel the market is accepting of a "no new API's approach" where true POJO's can be the development model yet clustering is injected at runtime, albeit declaratively. Everyone I spoke to understands what Terracotta does, and why we do it. And, inevitably the same lightbulb goes off.
"Terracotta should do this for Servlet (HTTP) Session. I would love fine grained object replication and persistence. I would love to get away from the side effects of serialization of POJOs and [large] graphs in session."
Of course, we have Terracotta for Sessions (Weblogic and Tomcat at the moment) which does exactly that. I believe after our Spring Experience session in Miami, 50 folks from around the globe downloaded our Session product. Spring's presence and global reach is amazing. And people's openness to try new things, now that they have seen the benefits Spring introduces into their application code, is also nice to see.
The most consistent pattern of problems out there, however, is the one of bridging the gap between entities (EJBs, Hibernate managed objects, etc.) and POJOs. Seems every developer I speak to wants to put a Hibernate object in container-managed clustered session or hang that object off a graph of POJOs that might end up anywhere. Hibernate, EJB 3, and others only allow the developer to disconnect an object from its manager and then merge out-of-band changes back to the manager context. Perfect opportunity for Terracotta's technology to help. I am very eager to get started in 2006 on bridging the gap between POJO's and entities.