our customizable newsletter, sends you security alerts, product updates and software patches on the products you use. Sign up now at www.networkcomputing.com /express/

For the first year of prognostications, I just didn't get what all the e-commerce fuss was about. I had been involved with EDI since 1984. What was the big deal? Finally, one of my media colleagues took me aside and educated me.

EDI was elitist--only the big players with big savings on every transaction were into EDI. By contrast, e-commerce was like the Internet. It was for the high-volume, low-value, ad hoc transactions of small business and the consumer. In the end, my colleague said, e-commerce would sweep away EDI with the force of its simplicity and commonness--just like IP did to OSI.

Sometime later, a colleague at work asked me for some insight into e-commerce. I suggested he imagine an iceberg. E-commerce was visible above the waterline, flashing in the sun and drawing all the attention. But EDI was hidden below the waterline and getting the job done; I added that if we did our job right, e-commerce would grow up to be EDI.

Whether you sell cars, CD players or ice cream bars, the regular customer is the mainstay of any business. Companies have traditionally expended considerable effort maintaining their regular customer base, but it's still less than they spend to attract new patrons. Marketing pros tell me that service, support and value create regular shoppers. E-commerce, for the ad hoc customer, has only a marginal impact on any of these. The value of e-commerce is to create a new group of regular customers who find Internet purchasing more effective for their needs.

It is now standard practice to create and maintain client information to better serve the regular customer. We assign identities so that "regulars" receive better service and value. We tune our business to their purchasing habits to reduce our operational expenses and, in turn, boost profitability. For e-commerce to play a role in our businesses, it must function as a tool in this information-gathering and business-tuning process. It is most effective when it meshes cleanly with our current data systems.

Communication costs and software complexity have been a barrier to the acceptance of EDI. In 1995, RFC 1767 (MIME Encapsulation of EDI Objects) was published and software vendors began standardizing EDI functions over SMTP services. This did not mean that EDI was dependent on unreliable Internet e-mail; rather, EDI was using the SMTP protocol as a message-queuing delivery system. These EDI systems ran their own SMTP servers and were typically configured without mail relays (except when it was necessary to traverse corporate firewalls).