Is JavaBeans ( EJBs ) important to Sun anymore as resuable RAD components ?

Considering the lack of answers to my own JavaBeans question on this forum, and to many others here, I have to question whether Javabeans is important to Sun anymore as a reusable component technology.

It seems as if Sun has just thrown in the towel when it comes to reusable software components, and ceded to Microsoft the ability to dominate the market in this area of reusable software in a RAD environment. Yet I can hardly believe that Sun does not see that component reusability is at the heart of success for Java and its platform, and that by not supporting it they will simply lose programmers to the dotnet environment.

Yet I have found not only are no answers to questions about JavaBeans given, as if JavaBeans is no longer an important or vital technology, but that Sun's support for JavaBeans in the Java API is half-hearted at best, as if they do not believe that the problems in using JavaBeans in Java programming environments and IDEs can even be addressed.

They added the good idea of a BeanContext, which allows beans and IDEs to work within the idea of many inter-related beans being dropped into a context in an IDE environment. But their documentation, and support for it, seems minimal. I asked a simple question how I can establish my own nested bean context as a property in my JavaBean, only to be met with no answer.

Visual RAD support for various standard Java properties, such as collectons, enumerations, enumsets as bit choices, or properties as nested beans, seems utterly non-existent in Javabeans and requires the programmer to write elaborate property editors for what should be basic property types and easily supplied by Sun as part of their distribution.

I keep getting the feeling that I am wasting my time developing JavaBeans, or even EJBs for that matter, as a RAD component because Sun has made only token effort at making the idea of reusable software components workable in RAD development environments.

Am I correct ? What do others think ?

[2032 byte] By [Eldienera] at [2007-11-27 3:00:37]
# 1
You might find this interesting, http://jcp.org/en/jsr/detail?id=273. Sun is leading the specification request.
developer_jbsa at 2007-7-12 3:41:54 > top of Java-index,Desktop,Developing for the Desktop...
# 2
That was over 2 years ago and nothing has shown up since in Java 5 or Java 6. This further leads me to believe that Sun has just given up on the idea of a RAD environment for JavaBeans or EJBs.
Eldienera at 2007-7-12 3:41:54 > top of Java-index,Desktop,Developing for the Desktop...
# 3

Check the timeline in Section 2...

2.1 Please describe the proposed Specification:

This JSR extends the JavaBeans specification and APIs to improve design-time functionality for use within IDEs.

Some key changes that are proposed are:

* Add an API for richer design-time manipulation of JavaBeans in an IDE

* Standardize common "extensions" to JavaBeans prevalent in current IDEs

* Clean up and specify common meta-data used by many tool/component vendors in BeanInfo

* Clarify expected behavior of IDEs with respect to PropertyEditors, Customizers, etc.

* Possible: Investigate a JSP tag library design-time rendering scheme for IDEs

It was a key goal of this JSR to be incorporated into version 6.0 of the J2SE platform (Mustang), but this technology will likely have to be released standalone after Mustang. It hopefully will be rolled into the next release of the J2SE platform.

A link to the public project

https://jbdt-spec-public.dev.java.net/

The javadoc was updated recently

developer_jbsa at 2007-7-12 3:41:54 > top of Java-index,Desktop,Developing for the Desktop...
# 4

Thanks for the further information. It gives me hope that JavaBeans will be more usable from the JavaBean developer's perspective in RAD IDEs. I won't ask when it will be completed but I do hope that those working on the proposal see the importance of getting out something in the near future, since two years does seem a long time to be working on such a spec without final results.

I also hope that the proposal will also offer a set of standard property editors which will be much richer than the current very limited set. It should also offer standard property editors for collections, enumerations, enumeration sets, nested beans, and bean contexts.

Eldienera at 2007-7-12 3:41:54 > top of Java-index,Desktop,Developing for the Desktop...
# 5

I had hoped the JSR would provide a standard localization framework. But, a contributor informed that the JSR is intended for IDE vendors, allowing them to provide new functionality based on the JSR's new standards.

I doubt richer property editors will be a direct product of the JSR, but it should allow for fuller features from IDE implementors thereafter. As for timeline, there is a big meeting this week (and next) in San Francisco. Maybe a release date will be discussed.

Message was edited by:

developer_jbs

developer_jbsa at 2007-7-12 3:41:54 > top of Java-index,Desktop,Developing for the Desktop...