Solving Real Problems

I am surprised to read less and less in BYTE about fourth-generation language (4GL) products that help build everyday applications, while I follow with lassitude your day-to-day analysis of the new Java trend. Sockets and threads have existed in Unix for a while, but when the Web exploded, Windows became a 32-bit development platform with Unix features. Suddenly, developers saw new opportunities, and the Windows NT versus Unix battle to control the Internet started.

What about Java? It's a new server-side language that is as cross-platform as C++ should be, but it is still a platform for the elite. Whether your 4GL is produc ing Java bytecode or proprietary p-code is just a matter of price for deployment.

The battle is not about the language or the platform. It is about writing stable and scalable software in multiuser environments. Let's face it, the tools made little progress in the last 10 years. That's why most applications that still run today were written in COBOL, C, assembly language, Xbase, and RPG. Replacing them with Java code may solve some technical issues such as portability, but it won't solve the real problem: addressing new business needs. Only a high-level tool can move information technology (IT) forward and deal intelligently with such issues as the Y2K bug, Euro currency, mobile users, and market globalization.

Alain Stouder
alain.stouder@smartway.ch