There is nothing wrong with Java. Then why arent more shops using it? IMHO, whats slowing the adoption rate of Java in traditional RPG shops has absolutely nothing to do with the language itself. It has everything to do with how programmers learn to program.
Think back to your first job on the System/36, System/38, or AS/400. How did you learn RPG? If youre like most people Ive met in the field, you started in a shop that already had some experienced programmers. You were assigned tasks of increasing complexity. You probably started out modifying simple reports and screens, then modifying or cloning more complex programs. Then, after you got good enough at that, you were able to write new programs from scratch. All the while, you had a nice base of code you could look at or clone as needed to help you along the way. The obvious advantages to this approach include a nice gradual increase in complexity and exposure to the expertise of highly experienced mentors.
There is another benefit to learning the old way that is rarely acknowledged. While you were learning how to program as I described, you were also learning other things at the same time without even thinking about it. You learned how to edit, compile, and debug; how to launch applications; how application systems are structured; how to set up a test environment, and a whole host of other things about the operating system.
Now think about how a lot of experienced RPG shops attempt to move to Java. They get a bunch of books and videos and send some people to a class. There these same people are then assigned to write their first system from the ground up. Does this sound like a recipe for success? Ive seen a few shops try this approach in the past with RPG, and it didnt work then either. Books, videos, and classes are extremely valuable tools for learning a language, but they tend to teach syntax and language-specific things as opposed to the more complex structures necessary to create useful business applications.
If Java were just another green-screen language, learning it would not be too difficult. You could just concentrate on the new language elements. But Java is a whole new ball game for green-screen programmers. The skills that make you a good RPG programmer will also make you a good Java programmer. Unfortunately, a lot of green- screen-specific skills youve learned wont help much with Java. Editing, compiling, and debugging are different. The applications are structured differently, they are deployed differently, and they execute differently. Throw in new concepts such as object orientation, event-driven programming, and the new syntax, and you can see why learning Java is such a daunting task. Oh, and dont forget that you still have 40 or more hours a week of green-
screen programming to keep up with the day-to-day demands of your employers. You may observe that, if there were more shops out there with experienced Java programmers, then you would not have this problem. Thats correct. But how do we get there?
The best way would be for those businesses with a vested interest in the success of Java, such as Sun Microsystems and IBM Corp., to provide some free Java applications to shops. Im not talking about demos or tutorial-type applications either, although they can be useful. I mean real-world applications that demonstrate at least some of the things that programmers have to do as part of their everyday jobs.
IBM could release some TAA tool-like programs written in Java. If these tools came with instructions on how to install and run the sample applications, then each shop that had them would have a good code base to clone from. Typically, TAA tool programs tend to be short on database access, so a better choice for a Java application might be a lightweight tape management system. Im sure Sun has a few things lying around that it could package and give away.
The good news is that Sun Microsystems and IBM have made large commitments to Java. The investment needed to package some applications and give them away should be small by comparison. Just look at what Sun Microsystems did with Star Division. It bought the entire company just so it could give away its softwarefor free. Surely it could do something on a smaller scale to jump-start adoption of Java.
With a modest investment of time and money by IBM, Sun Microsystems, and AS/400 programmers, someday soon well be hearing, Ooohhh, so thats how you do it.... I get it.
John A. Brown is an independent consultant and vice president of the Fairfield, Connecticut, AS/400 Users Group. He can be reached at
LATEST COMMENTS
MC Press Online