enjoyed the InformationWeek Spring Conference. The speakers were good and the topic, Radical Simplicity: Conquering Complexity and Enhancing Customer Value, was important. So when on the airplane coming home I read a newspaper article on the latest brouhaha over domain name registration, it immediately occurred to me that here was a perfect example of the need to figure out a better solution.

In the early '90s Jon Postel created the present domain-name system in which suffixes to the Internet address describe the type of site. Thus, .com refers to a commercial venture, .gov means a government function, .edu is an educational facility and so on. The Internet Corporation for Assigned Names and Numbers was created in 1998 to set the rules for domain names.

It has been struggling ever since to figure out what new suffixes are worthy additions to the original ones. ICANN expects to provide a final list of additional top-level domains sometime this fall. Among the seven chosen are .biz, .aero, .info, and .name.

While I can understand the desire to codify the rather unstructured Internet, I have to wonder exactly what problem we are trying to solve. It appears the objective is to assign a lot of new site names because we're running out of easy-to-remember ones such as www.informationweek.com . There is also the desire to tell us more clearly the type of institution just by looking at the suffix. While the thought is noble, I doubt the efficacy of the solution. Here's why:

First, I can't imagine that just because we have .aero available that American Airlines and United Airlines are going to decide they no longer need www.americanair.com and www.ual.com. Nor do I expect Nike to conclude it doesn't want www.nike.biz simply because it already has www.nike.com. After all, it already has enough trouble trying to register its site address in every country so that if you type www.nike.fr you get the real swoosh people as opposed to some Gallic imitator. These firms will rush to reserve their treasured names with the relevant new extensions.

Equally involved in the scramble will be cyber-squatters who will seek to make a fast buck by tying up the same valuable names for later resale. Of course, the companies involved could sue, but in many cases it will be a lot easier and cheaper to pay a squatter some money to go away.

No matter how you look at it there's a great deal of money riding on what happens with the domain-name system. VeriSign, reportedly gets about $180 million annually for managing the master lists of .com, .net, and .org names. Imagine the money at stake for the various companies involved in the new schema.

To avoid a lot of expensive confusion, we should make sure we've defined the real problem. We aren't running out of potential Internet names, just ones that are meaningful and easy to remember. So what's the solution?

First, leave alone what we have. Don't open up any more domain names unless we're eager to see another version of the gold rush as people stake out their claims. Instead, build a database that contains the URL and a brief standardized description of each registered site. Integrate into browsers a rudimentary "where" feature that queries that database only. Nothing fancy. When used, if I type in United Airlines or Nike it takes me where I want, or gives me the fewest possible choices. My guess is it will replace a lot of the heavy lifting done by tools such as Google as well as go a long way to eliminating the need for more domain names.

I suspect it's too plain a concept to be adopted and probably contains flaws; but after all, I was really bitten by that simplification bug at the conference.

--Robert M. Rubin is CEO of Valley Management Consultants, which specializes in E-business and IT strategy, organizational design, and evaluation. Before joining VMC, he was senior VP and CIO of Elf Atochem North America, a $2 billion diversified chemical company. He can be reached at valleymc@home.com.