Far be it from me to say yet another change in technology will address this problem. However, I will suggest you take a hard look at Java as the development language of choice.

Holy Wars First, let's get the religious issues out of the way. Industry support breaks into two camps: Microsoft Corp. and everyone else, and both camps have blinders on. Microsoft insists on nonproductive changes to the Java object model while everyone else insists that Java will take over the world--all the while ignoring the fact that no one is buying off-the-shelf productivity tools written in Java. Frankly, I have no interest in talking about it any more because I don't own stock in either Sun or Microsoft.

Both Java and VB offer fine programming tools to address business problems at a moderately high level using commonly available (in other words, affordable and trainable) programming talent. IS can use both Java and VB to create compelling intranet and client/server applications.

I'm sure many of you must be disappointed by now. I've mentioned both Java and VB and sidestepped all the arguments that feed daily water-cooler debate. So let's see if we can start a new one.

All Hail Java Java will eventually supersede VB, PowerBuilder and the like for two reasons: First, it doesn't cost more in terms of infrastructure, talent or developer tools--an important consideration given the overhead associated with C++ and other object-oriented (OO) languages. Second, and more important, it is the first environment that has combined the power of OO software development and the talents of rank-and-file programmers. Given the constant gripes about the scarcity of talent, management must work smarter by harnessing technology and the talent it has and finally implementing software reuse. That hasn't happened with VB during the past few years.

You might say SmallTalk, C++ and the like all have more elegant and complete OO facilities. Others might point out VB's new "implements" keyword and its applicability to developing DCOM (Distributed Component Object Model) servers. All valid, but the facts (observe any IS shop) show that SmallTalk and C++ have remained languages for the elite developer (read: high-priced), and for whatever reason, VB does not encourage development and deployment of reusable objects.

Java's culture has led both commercial and in-house developers to deliver solutions in the form of classes that the developer can inherit and deploy--classes as opposed to libraries, APIs, DLLs (Dynamic Link Libraries), OCXs (OLE Custom Controls) and the rest of the package of convoluted interdependencies you get from the VB community.

This inclination toward OO solutions fostered by the Java camp within their ranks has placed them, in my estimation, at least two years ahead of the VB camp. Furthermore, the next two years will bring an explosion in available bandwidth, resulting in more ambitious distribution of Java classes and frameworks. There is sufficient competition among Java tool vendors to keep pace with productivity enhancements and price reductions. Meanwhile, the demand for portability and agile redeployment to support server applications, embedded stored procedures and network-aware agents will raise the bar of expected functionality. All of this will favor Java.