The project claims to have a "radically different" design and "wildly faster" performance than anything else avaialble today. Will this project rock the EJB world? Or pass by unnoticed?
Since any EJB Application Server has to closely follow the specs, it is difficult to imagine radical departures in the design. I think the departures might be in the area of Persistence and O/R mappings. At any rate, anything Open Source is welcome as it can only benefit the rest of the development community.
I am supportive of open source projects, however, I find myself wondering why people end up creating several different competing projects. I think progress would be faster and the product would be better if everyone worked together. Why should there be so many open source EJB containers? Why don't they guys at OpenEJB and JBoss get together. Their goals are very similar. I would suspect that there would be a lot of synergy between them. Both entities are working on EJB containers; both entities are working on JMS implementations; both entities aim to be radically different. Sounds like a match made in heaven to me...
I'm agree with you, Nathan.
There are already 3 Open EJB servers available:
jBoss, JOnAs and now OpenEJB.
I think, jBoss and JOnAs are already mature, because I already tested them intensively as I wrote articles about them in a german Java Magazin.
What we need is a really "killer" EJB server that can compete with WebLogic or Persistence.
I think, Enhydra + JOnAs = Enhydra Enterprise could be the one.
I don't agree with this one. I think one should not
compromise a design. jBoss is for me at least the most
interesting application server around. It's open-source
(GPLed!), it has a very clever concept and it's moving
at an incredible pace.
Of course jBoss is not as feature-rich as Weblogic but
it's easy to learn and its community is very supportive.
jBoss is said to be on production level next month and
I'll hold my breath for this one :)
Well, Enhydra project could be the first complete opensource implementation of J2EE specification.
I hope that orionserver project will open source code and change license type. They could be the greatest players on opensource EJB container market.
I'm with you, if Orion was open source everyone else could go home.
I agree, Orion steals the show. It is so simple, easy and full of features...it almost seems too good to be true.
What about the scalability of these open solutions...?
How many concurrent users?
Why doesn't sun add the functionality to jserv/tomcat?
What u said about Orion is definitly true.If this server was an open source no other server will beat this product. I 've been using Orion for 2 months .It seems like an unfinished great ejb server and the guys who made it seems to be absent for the time being
I switched to jboss - I am still paying days and nights as price for using an open source product
I also agree. It seems to be one of the odd bits about Open Source that it doesn't have much incentive to get different groups working together unless there is a strong person in the middle pulling it all together. In one sense it is good since we get many different implementations. The bad side is how long it takes them to get done, and their ability to compete with commercial products in an ongoing feature race.
I disagree with you on one point. It's not the quantity of people that makes a project fast. It's the quality and a common vision on how things should be done. As far as I know Richard Monson Haefel (the MAN behind OpenEJB) was once on board of jBoss (back in the 1.0 days) but for whatever reason they (RMH and others) could not agree on several key issues.
In my opinion system architecture should be flexible and open but democracy? Hell, no!
We'll see which of those open source projects will be the best. Right now I'll bet a few cents on jBoss. But then who knows.
What about ejboss at http://www.ejboss.org
? Their open source EJB server is coming along quite nicely.