NOTE: Sun Microsystems is not an Entiva Group client.
Perhaps amongst all of the Web 2.0 hustle and bustle or the mounting anticipation for the seemingly inevitable release of Java into the open source domain, the impressive work that the GlassFish team has put into the first publicly available Java Enterprise Edition 5 (Java EE 5) implementation has gone unnoticed. However, Entiva has been tracking the GlassFish community as well as keeping in touch with its leaders on the Sun side of the equation for some time now, and have taken notice of what has developed into a well put together open source community.
Earlier this year, we made an analyst paper that we authored on exploring the capabilities of Java EE 5 using GlassFish available free of cost. And in the relatively short amount of time since the release of that publication, GlassFish has improved by comparative leaps and bounds. As we make light of in the paper, GlassFish happens to be more than just another Java application server, it is the only standing full implementation of the Java EE 5 specifications. Some analysts (no names, links to blogs, etc.) have written off Java EE 5, and as a fellow analyst I respect both their personal and professional reasons for doing so. Even JBoss CEO Marc Fleury joined the party by criticizing GlassFish as "irrelevant." An interesting point of view from someone who is at the head of a company that is in the midst of getting their own cut of Java EE 5 out the door, an implementation which will almost assuredly rely on Sun's reference implementation, a.k.a. GlassFish. It is a small world, indeed.
However, after investigating the new Java enterprise stack myself. I find it difficult to believe that very many others could make very many disparaging remarks after doing the same. No, it isn't perfect. And yes, it can be improved. But that categorization is inclusive of every technology known to mortal man. My point is that it seems as if most of the Java EE 5 bashing can be classified as thinly veiled attempts to take shots at Sun and their Java strategies of years gone by, or as 'Java is washed up' garble. If certain persons are bitter towards Sun for their past governance of Java, once again that's their personal business. As for the latter, Java EE 1.4 seems to be doing well enough with all of its flaws so I don't understand how an improved cut of the enterprise Java stack ends up being a negative.
Currently GlassFish v2 is available with some interesting features and improvements over v1, including enterprise quality clustering, high availability, and replication. While v1 is roughly equivalent to the reference implementation of Java EE 5, v2 is geared towards working the type of features that are already available in the Sun Java System Application Server. v2 also features support for Web 2.0-esque applications through project Phobos (The intersection of between the two promises to be critical for both as they vie for developer allegiance as a platform that eases, instead of complicating, application development). For more information about what's new and noteworthy with v2, visit the GlassFish community wiki.
The steady improvements made to the GlassFish experience coupled with an evident commitment to creating a solid Java EE application server (and not just a reference implementation), has made me somewhat of a believer in Sun's new 'Open Source as an Engine for Growth' strategy. GlassFish may not register on too many people's radar screen right now, but as Java EE 5 implementations start hitting the shelves by early to mid 2007 and the open sourcing of Java movement begins gather even more momentum. There will be a spike in interest in Java EE 5 capabilities and GlassFish will be primed for a significant surge in community participation as well as adoption rates. In the meantime, look for the GlassFish community to continue its steady push towards delivering a top notch open source application server that's capable of making some noise in the dense app server market.