As we get ready to embark on what promises to be one of the most challenging eras in IT history, it's becoming clear that we are not nearly as prepared as we should be. No matter what side of the electronic-business equation you line up on, the odds are pretty good that you will completely redesign your IT architecture this year.
This holiday season showed that just about every dot-com start-up company is stretched to its technical limits. Most of the dot-com operations were not really designed to handle the kinds of volumes we're now seeing. Most folks believed it would be years before we saw this kind of traffic, which means they thought they could rely on a basic client/server infrastructure for a long time. However, it's becoming apparent that client/server applications aren't going to do the job. And although it is possible to extend these architectures to an n-tier model, managing that many tiers becomes very complex.
So, many companies will need to start rewriting their applications using native Web technologies such as Enterprise JavaBeans and Extensible Markup Language or face the consequences.
First, the cost of maintaining n-tier systems is prohibitively high, which in turn leads to the second problem. If your competitors are able to leverage a competitive edge using Web technologies, you won't be able to compete over the long haul because the success of any e-business is tied to the efficiency of its IT architecture. If your competition has a more efficient architecture, they will be able to rack up more revenue and greater profits. Eventually, you won't be able to compete on pricing, your market share will fall, and Wall Street will punish you by lowering your stock value, making you an attractive takeover target.
Another one of today's great myths concerns brick-and-mortar companies that plan to "Webify" their legacy systems to take advantage of new e-business opportunities. These folks argue that they have unique, valuable business logic running on legacy systems. Once this business logic is made available to customers on the Web, they expect to become e-business juggernauts overnight.
But the truth is that most of this business logic is sadly outdated, and the business models on the Web change multiple times a year. Legacy systems are just not flexible enough to meet the requirements of e-business. Dot-com companies more often than not already run more efficiently than their brick-and-mortar counterparts -- a fact that hasn't escaped the notice of investors. Wall Street is valuing start-up companies higher than brick-and-mortar companies, because in the long term the business models of the dot-com companies yield higher profits.
Companies using client/server must move to n-tier architectures before going to architectures based on Web protocols. And companies dependent on legacy systems must "Webify" their architectures along a similar n-tier strategy.
These efforts represent mere way stations on the way to incorporating the Web into the core of the business. Anything short of that will ultimately prove futile.
Got a different view? Write to me at email@example.com.
Michael Vizard is editor in chief of InfoWorld.