Live from the first day of the JAX:
Rod Johnson, SpringSource:
The
Future of Enterprise Java
Does enterprise Java have a future?
Third answer: "No one will date
a Java Programmer" 
Java market is not shrinking and still
huge in comparison to e. g. Ruby.
Still growing, but there are strong
competitive threats, and Java hasn't fully delivered on its promise, yet.
Forces for change
- Current trends
- Introduction to Java EE 6
- The market scenario
Technology
forces
Six predictions for the future
- plus two totally free bonus predictions
The productivity challange
- Ruby on Rails
- no great leaps forward in enterprise Java for some time
- Java community tends to miss low hanging fruits
Enterprise Java slowed by "baggage",
conception is over 10 years old.
Need of a reaular clean out.
Java EE 6 is an attempt to clean
house; will come out late 2008 or eraly 2009.
Two new philosophies
- Extensibility: more extension points and SPIs, extensions should be as easy to use as build-in features
- Profiles: different sets of platform technology for different purposes, three profiles planed
see
JSR-316
Profile A - minimal, lightweight
- Tomcat
Key element: Servlet 3.0
biggest change relates to extensibility
extension at runtime instead of web.xml
Profile B - adds persistance
and two (?) component models
still relativaly lightweight
many points still unclear, like EJB
3.1 "lite" and Web Beans
Profile C - "Full platform"
Java EE "classic"
"Titan nuclear Missle of enterprise
Java"
Opposing forces: Modularity and Monopoly
Java EE 6 promotes modularity
Current AS market quite consolidated:
WebLogic + WebSphere ~70%
Both market leaders are part of much
bigger plays
Oracle: aggressive vertical play
IBM: services
Both see their Java middleware as part
of a Microsoft-like full-stack solution
How to measure the market?
old measurements are obsolete
Predictions
1. Real competition will return
to the application server market
Bonus 1: Economic value will become
more aligned with what people actually use
2. Tomorrow's application server will
be lightweight and modular
3. Tomorrow's application server will
not merely implement JCP specifications
Bonus 2: The JCP will change to be run
through open source; Sun is becoming an open source company
4. The market will address the gap between
Tomcat and WebLogic/WebSphere
5. The gap between application servers
and ESBs will be bridged
6. The Black Knight will be defeated
EJB is dying
Why does it matter?
EJB is the centerpiece of "old
J2EE"
It's vital that it is put aside for
progress to be made
Conclusion
Wer are in for a period of rapid change
Java EE 6 may keep Java EE relevantt,
but Java doesn't really shape the future.
One of the key technologies that will
shape the future is OSGi
It will be an exciting time.
