Bruce Tate has some interesting comments on the announcements at JavaOne this week, on how developers have ‘hit the wall’ with the current programming model with overly bloated feature rich APIs:
“They’ve hit the wall. It’s ground-hog day all over again. Get excited. Learn a monolithic API. Start to build it. Make design mistakes. Dig hole. Make mess. Curse. Work overtime. Scratch and claw. Find silver bullet. Lather, rinse, and repeat”
We’ve all been there, done that. So whats next? Whereas OO programming was the last major shift to get past the last ‘wall’, he believe Aspect Oriented Programming (AOP) is going to be the next major change in development.
Major annoucements today included:
- NetBeans 4.0 released: now a complete J2EE development environment with support for developing JSPs, Servlets, EJBs, and Web Services
- Allied Irish Bank, one of Irelands largest financial institutions, is going to be migrating 7500 users to the Linux based Java Desktop System operating system.
In the first general session of the week, Jonathan Schwartz, President and COO of Sun gave an overview of the current state of the Java industry. In summary, Java continues to become widespread, not just in software development, but is becoming ever more present in the consumer electronics place:
- In the last 12 months, the Java mobile games industry generated $3 billion in revenues
- 350 million mobile phone handsets worldwide that support Java
- 600 million Java Cards shipped worldwide – used in phone access cards
- 650 million PCs running Java on the desktop
In the IT industry:
- 100 million Java Development Kits downloaded since Java was launched
- 4 million Java developers worldwide (and increase of 1M since this time last year)
The huge increase in number of Java-supporting phone handsets has created a whole new marketplace for building and delivering services to customers, including news and information services, and online and downloadable games.
Java technology is driving a $100 billion a year economy, covering many varied applications of the technology from the medical industry to entertainment.
The major trend predicted this year is the come back of the client application. So whereas is past years the big push and trend has been moving towards web-based applications accessed everywhere and the ability to access information from distributed systems running everywhere, this year according to Sun, the client is back. Not as much in the traditional desktop sense, but because the Java platform have the broadest reach of any development language, and now as it is available on millions of handsets across the world, suddenly a new platform has emerged and driven by Java technology. People want a rich compelling user experience, and todays technology being carried in the pockets of 350 million people, driven by Java.
What’s next? Java applications running in your car…
I expect J2SE 1.5 is going to be announced this week, as it has been in beta now for several months, and Sun usually reserves announcements like this for JavaOne..
I can’t image what other announcements there might be, as the big news around the POJO direction change for EJB3.0 is already common knowledge. Have to see what they announce….
TheServerSide.com have a book review of a book called ‘Covert Java’, which covers topics such as decompilation, JVM internals, debugging, and tracing.
These are topics which are not covered in many other books – some good reading for someone with a couple of years Java experience and wants to increase their understanding of the Java platform, although maybe not as much use for someone with more than a couple of years experience and has probably already used these techniques.
Bruce Tate makes an amusing analogy between the EJB spec and hunting elephants in this article.
Good comments here about how early adopters felt forced (and still do?) to take on all that the EJB spec offered, even though it was probably not the best fitting solution.
This article on TheServerSide.com discusses the recent announcents regarding EJB3.0 changes and support for POJO persistence, and also the big players voting down the JSR for JDO2.0, due to overlap with ‘other interests’, namely at the time the upcoming EJB3.0 recommentdations.