Sun Microsystems recently filed a $1 billion lawsuit against Microsoft, ostensibly to recover “damages” from Microsoft’s monopoly and to force Microsoft to distribute Sun’s Java Virtual Machine (JVM) with Windows XP and future versions of Microsoft Windows. The lawsuit has nothing to do with Java on the client, and everything to do with Sun’s anxiety over the positive corporate attention garnered by Microsoft’s .NET and XML Web Services initiatives. In order to understand the rationale for the lawsuit, it’s important to first understand where Sun’s Java initiative stands.
Java on the client is unimportant
These technologies allow for richer client interfaces and faster page loading times without resorting to Java applets. Organizations that develop systems that require client-side Java support typically choose a particular JVM to be installed on the client rather than assuming that any JVM will do. In fact, it was Sun’s insistence on maintaining the “purity” of Java that resulted in a court injunction in late 2000 under which Microsoft had to remove Java support for over 6,800 products in order to comply.
In its January 2001 settlement with Sun over the inclusion of Java with the OS, Microsoft retained the right—but not the obligation—to include a specific, down-level version of the JVM for up to seven years. Currently, Windows XP users who attempt to load a Java application are directed to a Microsoft site where they can automatically download this version of Java.
Java on the server is significant
There are literally hundreds of examples of successful, highly reliable, and scalable implementations of Java and Enterprise Java Beans (EJB) at the server. The primary beneficiaries of this business, however, have been IBM, Oracle, and BEA Systems, and not Sun.
IBM’s focus—rebuilding as a services company with hardware and software solutions—has strengthened its position against both Sun and Microsoft. IBM is not only enjoying revenues from sales of its WebSphere product line, but it is also gaining revenues from implementation of WebShere as well as other non-IBM products.
IBM also has no particular allegiance to Sun hardware and operating systems. It gives the same attention to Windows 2000, Linux, and its own versions of Linux and OS/400—all running on lower-cost IBM hardware, with IBM service and maintenance revenues tacked on. Oracle is still the database bucket of choice for EJB solutions and is giving BEA a run for its money as the preferred middleware for EJB implementations. Sun has spawned a vibrant industry around Java at the server, but, as yet, has been unable to capitalize on its adoption.
Another major blow to Sun’s revenue stream is the collapse of the dot-com phenomenon. Sun was the leading provider of high-end hardware and software to dot coms, but since hundreds have disappeared, the hardware is now popping up in the resale market. This has put great revenue pressure on Sun given that many businesses that used to buy directly from Sun can get the same hardware for pennies on the dollar in the used-equipment market. Sun’s dependence on hardware revenues is one reason that the once highly valued stock is currently trading under $10 a share.
Clouds on the horizon
Despite the current success as a server programming platform, Java and EJB are beginning to show some vulnerability. In order to get acceptable performance, users are forced to tie systems more closely to specific hardware and software platforms, minimizing or eliminating the “write once, run anywhere” promise of Java.
Current implementations of the Java platform don’t natively support Web services or SOAP and so require developers who want to participate in the groundswell of new Web services development. (This support is planned for an upcoming, end-of-year release.) Although Sun participated in early discussions around—and ultimately agreed to—the W3C. SOAP recommendations, it is the only major technology company that has not signed on to the Web Services Infrastructure (WSI) consortium. While Microsoft, IBM, Oracle, and other major technology companies hammer out the standards that will allow customers to develop secure, routable, and transactional Web services between their respective technologies, Sun sits on the sidelines pursuing an “innovation through litigation” strategy.
History repeats itself
Sun would do well to study history before pursuing this dangerous course of action. The landscape is littered with the corpses of companies who chose to fixate on Microsoft (Borland, WordPerfect, Ashton Tate) rather than continue adding their own value to the industry.
IBM and Apple are great examples of companies that retooled themselves into successful entities by focusing on what they do well rather than attempting to beat Microsoft by going head-to-head.
Sun’s real motivation in this lawsuit has nothing to do with the distribution of Java on the client. It’s a desperate attempt to prevent Microsoft from including the .NET Framework in future versions of Windows or in future service packs.
Sun recognizes that Microsoft has a compelling platform for developers and corporations. The availability of the .NET Framework on every Windows desktop will ignite resurgence in the development of native Windows applications, since the Framework eliminates the deployment and support costs normally associated with Windows applications.
Like Java before it, the Framework can automatically download, install, run, and update client-side applications over a standard HTTP connection, accessed through a browser. .NET gives developers a common development paradigm for the server and the client, as well as their choice of any computer language for development (including the Java language). By removing the most significant costs associated with standard Windows applications (deployment and maintenance), Sun’s position as the provider of the “platform of choice” for distributed applications becomes significantly weaker.
The irony is that rather than focus on improving its platform, Sun’s use of the courts to further its agenda simply serves to highlight its shortcomings.