Finally a post that seems to be sticking up for the Spring framework. (nice change)
Sure EJB 3.1 does make things easier, but it has taken a lot of time to catchup to where Spring was. And as the blogger points out, as new versions of the technologies that JEE use like JPA come out, how long will it take for JEE to implement these.
Also, one thing that does bother me with all the ‘my JEE is better than your Spring’ articles is, and the blogger alludes to here, is that they compare annotations in EJB to Springs old XML config which just like EJB is seldom needed these days. This is a little unfair.
As for myself, I’m technology agnositic, but easily develop biases based on the path of least resistance factoring in long sightedness too. Like many, if a tool does what I want and is scalable, I’m going to use it. I think both approaches are fine but they aren’t necessarily compatible. Unfortunately we aren’t unified yet although the standards are in place to get it to work (eg Spring components <-> JEE components). Getting a Grails app to talk locally to EJB components hasn’t been easy as I learnt last week primarily due to having the app log to the right place in a container its not used to, and thats just at the container level, but thats another post.