Application designers need to keep in mind the volume of data, any time-bounded constraints (critical in the case of voice or multimedia), the capacity and responsiveness of the selected wireless network, and, of course, such details as the display size, processor and memory, and user interface of the subscriber unit. Today's wireless networks and mobile devices are indeed limited, and mobile middleware bridges this gap nicely.

But we now count 200 mobile middleware vendors, all selling much the same vision as outlined above, and clearly the world doesn't need all of them. I recently heard the CEO of one well-capitalized middleware firm say that in 18 months there would be only five competitors left here. While he believed that how much money one had would be the determining factor, one could also argue that lack of effective product differentiation might be a better reason for a shakeout. After all, most of these products look much the same. But it's worse than that.

While a major reduction in middleware products and firms is inevitable, I think so for a different reason. The evolution of middleware fits a pattern that has been a key element of the business of information systems almost forever. You see, the functionality of middleware will eventually be defined in standards and then embodied in systems software, most typically the operating system or equivalent functionality. We are going to see the standardization of wireless communications protocols, and the standardization of subscriber-unit characteristics, most likely as some variant of HTML/XML/XHTML (Japan's wildly successful iMode, for example, is based on CHTML). These two factors virtually eliminate the need for middleware. Manage the data with the DBMS, send it using TCP/IP, and display and interact with it using HTML. Sound familiar?

Thoreau and Occam were right-simplify. Complexity is a byproduct of evolution, as Nature searches for an approach that works. But, in the end, complexity and proprietary solutions need to be replaced for an industry to move forward.

Craig Mathias is an analyst with the Farpoint Group (Ashland, Mass.).