JUMPing the Java Habit

  • Smaller Small Medium Big Bigger
  • Default Helvetica Segoe Georgia Times
Microsoft has released a beta migration tool that will enable organizations to abandon Java and move to Microsoft's C# programming language. Called the Java Language Conversion Assistant (JLCA), this tool is a key part of Microsoft's Java User Migration Path (JUMP). The idea is that the JLCA will enable developers to JUMP off of Java and into Microsoft's Web-based .Net computing architecture.

The JCLA is used with the forthcoming Microsoft Visual Studio .Net and is touted to automatically convert basic Java source code into C#, Microsoft's proprietary Web application development language designed specifically for the .Net Framework architecture.

Migrating from Java

Essentially, the JCLA turns a straightforward Java application or Java Server Page (JSP) into an application that will run on a Microsoft server. This removes the requirement of running a Java-based server to host the application. The JCLA purportedly converts as much of the Java application as it can and then reports on the remaining parts that must be rewritten manually in C#. The first beta version of the JLCA is available for download on Microsoft's developers' Web site at http://msdn.microsoft.com/. The final version of the JLCA is due out mid-year, and an enterprise version is also being developed to convert Enterprise JavaBeans (EJBs).

How will JUMP and the JCLA help Microsoft attract new converts to its .Net strategy? Why would anyone want to move from the relatively open standards of Java into the jungles of a proprietary architecture such as .Net? The answer often has as much to do with IBM's grandiose WebSphere brand as it does with Microsoft's culpability. In the midmarket arena, this is the platform debate that is being waged for the hearts and minds of the developer community. The question: Which platform--WebSphere or .Net--will gain the most loyal adherents?

Open or Closed Web Application Server Platforms

On the surface, anyone who is investing real money in building a Web e-commerce application will pay lip service to the idea of open standards for the Web application server. It only makes sense: 1) You want your application to run on as many different platforms as possible, 2) You don't want to have to rewrite code if your platform vendor fails or loses favor, and 3) You want to make certain that what you build will work with other middleware tools that are available.

And, indeed, this is the path that IBM has followed from the inception of the WebSphere Application Server: WebSphere is "open," it's standards-based, and everything integrates together.

However, what happens if you, as a developer, start to follow this path toward open systems only to discover that WebSphere's a lot more expensive, it's an extremely complex and difficult strategy to follow, and Microsoft can provide most of the middleware your customer needs anyway?

This seems to be the big complaint that midmarket customers are starting to voice about IBM's WebSphere branding: It costs too much, it's overwhelming and confusing, and the customers don't really need all that horsepower. To these customers, moving away from Java and WebSphere on small to medium-sized projects might represent a real labor and time-saving decision.

What Will IBM Do with the Midmarket?

Meanwhile, IBM, for its part, seems clueless about how to satisfy the finicky needs of the midmarket. Will the Web application server needs of the small and medium-sized business really require the bulletproof, stainless steel middleware products that WebSphere delivers? Or will these customers' needs be more readily met by the burgeoning application service provider (ASP) and Web services technologies? If ASPs and Web services are the wave of the future for the midmarket, then that--in and of itself --is a computing cultural shift that may prove a bit too revolutionary for midmarket customers. These are small to medium-sized businesses who want to have their critical apps in-house. They're not about to place all their critical apps in the hands of an ASP a thousand miles away.

History Repeats?and Repeats?and Re?

And, in a curious way, this same midmarket customer dilemma still haunts IBM from the days when SanFrancisco was promising everything to everybody. After all, wasn't SanFrancisco supposed to deliver the new architectures that everyone would be using: open, flexible, standards-based e-Business apps?

In reality, SanFrancisco delivered so much complexity that most--if not all--midmarket customers could never use it. In fact, SanFrancisco delivered WebSphere, and indeed, IBM WebSphere branding is following in exactly the same path: great technology, ripe for genius, but maybe not exactly the stuff of everyday developers.

Market Niche for Microsoft .Net

Into this market niche steps Microsoft with .Net and C#. "Sure, it's proprietary!" is the Microsoft argument. "That's how we'll make you so productive! We'll pull everything together into a neat, simple package. And we'll make it work because we own everything inside! So JUMP the Java habit! Get decaffeinated, move to C#, and get on the .Net bandwagon!"

And midmarket customers are going to buy it, too. Even if Microsoft owns it lock, stock, and barrel. Why? Because it's simple, and it's what they think they need now!

Thomas M. Stockwell is Editor in Chief of MCMagOnline. He can be reached at This email address is being protected from spambots. You need JavaScript enabled to view it..