Web Beans: JBoss Seam being standardized for Java EE 6 release

Web Beans JSR 299 proposal by JBoss (RedHat Middleware LLC) has been approved unanimously by the JCP executive committee even though IBM and BEA Systems have expressed their concerns regarding it being (too?) ambitious. Following is what they had to say:

------------------------------------------------------------------------------
On 2006-06-05 BEA Systems voted Yes with the following comment:
This appears to be a sufficient challenge to achieve, but, in light of the overwhelming support at this stage of the process, we are prepared to see it go ahead.
------------------------------------------------------------------------------
On 2006-06-05 IBM voted Yes with the following comment:
Delivering on the deep level of integration that is proposed appears to be an ambitious task.

In short, the goal of this initiative is: to enable EJB 3.0 components to be used as JSF managed beans, unifying the two component models and enabling a considerable simplification to the programming model for web-based applications in Java.

The benefit will be: this work will provide a programming model suitable for rapid development of simple data-driven applications without sacrificing the full power of the Java EE 5 platform. This is a domain where Java EE has been perceived as overly complex.

Gavin King (Hibernate founder) is the spec lead for this effort which is targeted for release of a RI with Java EE 6 (fall 2007). This is also the first time that JBoss is leading a standard spec.

Some links to learn more about what Web Beans will offer are:
Gavin's interview by InfoQ on JBoss Seam
Proposed spec description for JSR 299
What Matt Raible (author of Spring Live and AppFuse founder) said about Seam
The JBoss Seam Demo
The JBoss Seam Documentation
JBoss Seam FAQ

What excites me?
Seam is an ambitious full stack framework (unifying and integrating technologies such as Asynchronous JavaScript and XML (AJAX), Java Server Faces (JSF), Enterprise Java Beans (EJB3), Java Portlets and Business Process Management (BPM)) which enables EJB 3.0 components to be JSF managed beans. So your EJB 3.0 beans could become your Action classes and the glue code which was required earlier for enabling Actions to invoke methods on beans in the business tier will go away thus simplifying the development effort required.
Some specifics from the proposal as quoted by Floyd Marinescu at InfoQ are:

• Changes to EJB 3 that will be needed for EJB's to act as JSF managed beans.
  
• Annotations for manipulating contextual variables in a stateful, contextual, component-based architecture.
• An enhanced context model including conversational and business process contexts.
• Extension points to allow the integration of business process management engines.
  
• Integration of Java Persistence API extended persistence contexts. (The other type of persistence context is transaction-scoped persistence context.)
  

Now thats what makes Web Beans a full stack framework! :) Here's one Seam Book by JBoss insiders.

Some notable aspects of Seam:

1. State management: Most other web frameworks store all application state in the HTTP session, which is inflexible, difficult to manage and a major source of memory leak. Seam can manage business and persistence components in several stateful scopes: components that only need to live across several pages are placed in the conversation scope; components that need to live with the current user session are placed in the session scope; components that require interactions from multiple users and last extended period of time (i.e., survive server reboots) are placed in the business process scope. The advanced state management facilities allow us to develop web application features that are previously not possible, or very difficult, to implement.
2. Multiple browser windows/tabs: Seam supports fine-grained user state management beyond the simple HTTP session. It isolates and manages user state associated with individual browser window or tab (in contrast, HTTP session is shared across all windows of the same browser). So, in a Seam application, each browser window / tab can become a separate workspace that has independent history and context. Multiple user "conversations" can be supported for the same browser. This behavior is similar to that in rich client applications.
3. Handling back-button navigation: Seam's nested conversation model makes it really easy to build complex, stateful applications that tolerate use of the back button.
4. Support for REST style bookmarkable URL: it is very easy to expose REST style bookmarkable URLs in a Seam application. In addition, the Seam application can initialize all the necessary backend business logic and persistence components when the user loads that URL. This way, the RESTful URL is not only an information endpoint but also an application interaction start point.
5. Support for JSR 168 portlets.
6. Support for internationalization.
7. JBoss Eclipse IDE provides a sophisticated, template-driven database reverse engineering tool which can generate an entire Seam application in minutes and a graphical jPDL editor for editing Seam pageflows and workflow definitions.