Portals, e-commerce sites, B2B commerce and so on...we're witnessing unprecedented demand for e-business solutions of every stripe as companies rush to put their businesses on the Web. With Y2K now out of the way, this has become the top IT priority.

In the past many of these solutions have been built on top of IRM (Internet Relationship Management) products such as Vignette and BroadVision whose primary specialization has been content management and personalization. For early Web sites, mostly serving up static content, this was sufficient. But once the basic site is up and running, demand quickly shifts to more data-driven applications in which customers can place orders, see inventory, book travel, track packages and the like. This requires the ability to develop highly scalable, enterprise-class Web applications that can access existing corporate data and applications and provide high-end transactional capabilities. This kind of capability is available only from application server and solutions vendors.

Because of the demand for both content management/personalization and back-end data access/transactions, we're about to see a collision between these two markets. In particular, almost every application server vendor has begun developing e-business solutions that include IRM functionality based on top of their application servers to meet the demand from their clients for more prebuilt functionality.

An important backdrop to this upcoming battle is the advent of an extremely important industry standard, J2EE (Java 2 Platform, Enterprise Edition), that standardizes the way application servers should work. This standard has been uniformly adopted by almost every application server vendor, and is being enthusiastically adopted as the standard for Web development by much of corporate America.

J2EE is creating tidal waves in the Web application marketplace just as SQL did when it swamped all other relational databases.

Prior to the standard, Web software companies - including IRM vendors such as Vignette and BroadVision - wrote their own servers to support their applications and frequently spent as much as 70% of their energy on doing so. These were proprietary, with weak architectures missing important features such as distributed objects, transactions, standardized data access and message queuing. Furthermore, they provided proprietary scripting languages that weren't up to the task of writing serious enterprise applications. Today no software company in its right mind would write its own proprietary server. It would simply write to the J2EE standard and support the leading J2EE servers in the market.

Because of the availability of J2EE standardized servers, there's a backlash at customer sites against the proprietary server architectures. Organizations standardizing on J2EE want to use this architecture for all their e-business applications. That way they invest in learning a single skillset that can be used across all projects, and they're easily able to find developers who already know the J2EE standard. They're also able to use a much wider range of development tools since products such as Rational Rose and Macromedia DreamWeaver are adapted to support the J2EE standard. In addition, they can purchase add-on components and prebuilt software modules from third-party vendors that comply with the J2EE standard.

As the content management/personalization vendors collide with the application server vendors, you can see a major battle brewing. At stake is the huge e-business platform market. In the one corner are Vignette, BroadVision and similar products, built on a proprietary architecture with weak scripting capabilities. In the other corner are the application server vendors who are moving into e-business solutions, leveraging their strong architecture.

The obvious move for the existing IRM vendors is to rewrite their products to be J2EE compliant. Companies such as Vignette are describing multiphase plans to do something like this. But it's a difficult task that may well take years to accomplish correctly, during which time it may be close to impossible to simultaneously evolve their solutions functionality at a rapid pace. Probable outcome: loss of momentum and marketshare.

Several application server vendors are already pursuing the strategy of adding e-business solutions to their product line and moving into the complete e-business platform space. Not all will be successful, since the solutions skillset differs from the systems skillset that most possess. This sets the stage for a series of important new battles: it'll shortly become clear that an application server without an IRM layer offering higher-level solutions functionality such as content management, personalization, user profiling and shopping baskets will no longer be considered adequate.

Buyers are placing more and more emphasis on this part of the vendor's offering, and view the J2EE application server as a standardized component that isn't highly differentiated from one player to the next. So the winners will be those who best understand the requirements in this solutions area and who move fastest to create the appropriate product offering.

Ultimately, an interesting battle will start to evolve between the traditional IRM vendors and the new application server-based IRM offerings. As the application server vendors reach critical mass with their functionality, we'll see them begin to take significant market share away from the traditional vendors because their industry-standard application server-based solutions will be more complete and will offer far greater flexibility and customizability.