Network Computingwww.networkcomputing.com

So I wonder: Should Oracle be taking a lesson from Novell's history, or is Oracle doomed to walk the path that has been Novell's decline?

The Old Red Mare Don't get me wrong. I don't think Novell is going away soon. However, even the most die-hard NetWare fan would be hard-pressed to deny that Novell's relevance is nothing compared to what it used to be. At one time, the reach from Provo, Utah, was far beyond Novell's actual grasp. Novell's products affected such industries as NICs, printers, hard drives and controllers and a horde of independent software developers. Its products even spawned a number of software industries, such as software metering and network-based backup.

Wisely, Novell chose not to compete with the fledgling companies that clung to its coattails. This strategy kept the company growing and moved the industry along at a breakneck pace. Stratification helped Novell reach into every corner of computing--from dentist offices to virtually all of the Fortune 1,000.

The key element that always seemed to escape Novell was the ability to diversify. Its failed purchases are legendary. From Unix Systems Laboratory to WordPerfect, very little good came of Novell's purchases. Even when the company did make good purchases, new divisions were rarely exploited to the fullest and most profitable extent. Take the purchase of Excelan, for example. Excelan brought Novell great TCP/IP software, but Novell was never clever enough to really put the standards-based TCP/IP on an equal footing with its proprietary IPX/SPX.

In fact, Novell has only recently seriously talked about IP being a primary protocol. Given that Novell now calls NetWare, its flagship network operating system, IntranetWare, it seems logical that IP should be as usable as IPX. But this is, in fact, still not the case and won't be until some future release of IntranetWare.

While purchases within the scope of NetWare were often merely problematic, purchases intended to diversify the company were true disasters. Neither UnixWare nor AppWare were of any use to Novell and probably caused real damage. Those of us who were fans of Novell saw almost immediately that its UnixWare strategy would never amount to anything. From the product's inception, Novell showed that it knew nothing about the Unix market, making almost every feature of UnixWare offensive to the existing Unix user.

AppWare, for its part, attempted to oversimplify a very complex task. Point-and-click application development appeals to almost no one. If you felt you must point and click, you probably didn't have the sophistication to create a real enterprise application. Those who were up to the task preferred other tools.

So how does all this relate to Oracle?

A One-Product Kind of Company? First, to say that Oracle is a one-product company is to overstate the obvious. Database sales still account for approximately 75 pe rcent of its total revenue, with tools and services to support those database sales making up the rest. Although it has built onto its core database technology in various intriguing ways, the heart of almost everything that Oracle sells is an Oracle Database.

Second, attempts to diversify, at least so far, have been slow to take off. Most of Oracle's new sales opportunities come in the form of business application software, where it competes for market share with SAP and PeopleSoft. Oracle may do well, but as Novell learned with WordPerfect, running third in a three-way competition is rarely profitable.