TSSJS Day 3 Keynote (Panel - Future of J2EE)
Our third and final day at TheServerSide Java Symposium did not feature a keynote address as the other days have. Instead, we had a panel discussion from some of the top brass attending the session. Several of the names escape me, but the most important contributors were Rod Johnson, Linda DeMichiel and Mark Hapner.
The panel was actually quite disappointing in that everyone tended to agree with each other far too much. Each question received one good answer that would have sufficed on its own, but a parade of tag-on answers tended to follow every time. The most "juicy" conversation centered around whether .NET still posed a thread to J2EE. The answer was essentially that it will always be a strong competitor, but that the open source community behind J2EE and the power of IBM's support are probably enough to keep .NET at bay.
Toward the end, the question of the Persistence Wars (JDO vs. Hibernate vs. Toplink vs. EJB CMP vs ....) was raised. Specifically, the audience member had asked why we needed a new JSR spec for a persistence API since that's exactly what JDO is. Really, there is no good political answer to this. The real answer is that JDO failed to take off ENOUGH in the marketplace to hold off Hibernate and TopLink. If JDO was not a JCP certification framework, we would be having none of this discussion.... It would have been declared a 3-way tie, and no winner would have emerged. The creation of a new API at that point would have been noncontroversial. However, since JDO was a Java certficiation, it's a bit of a slap in the face to it that it is being subsumed along with the other frameworks. Tough luck, JDO... The future is EJB 3.0.
Anyway, my thoughts on this notwithstanding, the panel did finally disagree on something when this came up. Linda DeMichiel (spec lead for EJB 3.0) said she considered what has happened with the EJB 3.0 persistence API a validation of the strength of JCP. Different groups came together to work to create one unified standard. Again, I think that's a more political answer than a candid one, but I like Linda a lot so I'll give her the benefit of the doubt. I do have to tip my hat to Rod Johnson for being the only person to say that what happened with JDO was a bit of a disaster. Linda quickly (and correctly) ended the conversation by saying that we need to put the Persistence Wars behind us and move on, and I think we're all grateful for that. But it was amusing to see a little death rattle of the old JDO controversy rear its ugly head one last (?) time.
Getting back to the basic message gleaned from the panel discussion, it was that the vision for J2EE is clearer than ever, projects are adopting J2EE with more success than ever, and that we are definitely in execution mode at this point. It's time to take new technologies (JSF, AOP, EJB 3.0, etc) and put them to use.
The panel was actually quite disappointing in that everyone tended to agree with each other far too much. Each question received one good answer that would have sufficed on its own, but a parade of tag-on answers tended to follow every time. The most "juicy" conversation centered around whether .NET still posed a thread to J2EE. The answer was essentially that it will always be a strong competitor, but that the open source community behind J2EE and the power of IBM's support are probably enough to keep .NET at bay.
Toward the end, the question of the Persistence Wars (JDO vs. Hibernate vs. Toplink vs. EJB CMP vs ....) was raised. Specifically, the audience member had asked why we needed a new JSR spec for a persistence API since that's exactly what JDO is. Really, there is no good political answer to this. The real answer is that JDO failed to take off ENOUGH in the marketplace to hold off Hibernate and TopLink. If JDO was not a JCP certification framework, we would be having none of this discussion.... It would have been declared a 3-way tie, and no winner would have emerged. The creation of a new API at that point would have been noncontroversial. However, since JDO was a Java certficiation, it's a bit of a slap in the face to it that it is being subsumed along with the other frameworks. Tough luck, JDO... The future is EJB 3.0.
Anyway, my thoughts on this notwithstanding, the panel did finally disagree on something when this came up. Linda DeMichiel (spec lead for EJB 3.0) said she considered what has happened with the EJB 3.0 persistence API a validation of the strength of JCP. Different groups came together to work to create one unified standard. Again, I think that's a more political answer than a candid one, but I like Linda a lot so I'll give her the benefit of the doubt. I do have to tip my hat to Rod Johnson for being the only person to say that what happened with JDO was a bit of a disaster. Linda quickly (and correctly) ended the conversation by saying that we need to put the Persistence Wars behind us and move on, and I think we're all grateful for that. But it was amusing to see a little death rattle of the old JDO controversy rear its ugly head one last (?) time.
Getting back to the basic message gleaned from the panel discussion, it was that the vision for J2EE is clearer than ever, projects are adopting J2EE with more success than ever, and that we are definitely in execution mode at this point. It's time to take new technologies (JSF, AOP, EJB 3.0, etc) and put them to use.
posted by Anonymous at 10:42 AM
0 Comments:
Post a Comment
<< Home