Product Options, Alternatives and Risks

Voter Parallel

There's a parallel outside the sphere of information technology. To explain falling voter turnout in national elections in the 1960s, social scientists advanced the theory that voting is an irrational act. Since the probability of a single vote affecting a national election is effectively zero, it makes little sense for a rational person to take a couple of hours out of a busy day to head for the polls.

Ah, but what if everyone thought this way? Then voting really could make a difference. That's true, to a point, but the fact remains that an individual's decision to vote or not vote won't impact the outcome of a national election. A decision by an individual to persuade others to vote for a particular candidate could tip the scales, but your one vote won't make a difference. Scholars tested this "rational voter" theory and found that it didn't explain low turnout.

IT decision-makers face a similar dilemma. In their hearts, many would love to buy switches from someone other than Cisco or run operating systems from someone other than Microsoft. They understand that if enough of their colleagues did the same, it might pressure Cisco to lower prices or Microsoft to get its security act together. But when it comes time to sign the purchase order, they inevitably settle on Cisco or Microsoft because they conclude that such a decision is in the best interest of their employers, regardless of whether it's good for the buying community as a whole. Let someone else speak up for what's right. I want to keep my job.

Opportunities and ObligationsThis herd instinct is a powerful one, and it's unlikely to change until IT pros begin to feel some obligation to the larger community. People vote in national elections not because they think their vote will decide the outcome, but because they appreciate the negative consequences to the nation if everyone took the easy way out and stayed home.

The same thing needs to happen in our neck of the woods, but let's be realistic: Few are willing to make decisions that jeopardize their job simply to stand up for what is in the broader interest. But you can position your organization in a way that keeps incumbent vendors on their toes while minimizing your exposure to risky alternatives.

Take Ethernet switching, where Cisco dominates. Although Cisco's enterprise gear is solid, it's hardly superior to products from the likes of Extreme, Foundry and Hewlett-Packard, whose offerings can cost half as much. So why not build a network that incorporates products from at least two vendors? As long as you're willing to work within the context of standards--and that covers most of us--the interoperability issues are minimal.

Will it cost you a little more to manage two vendors' products rather than one? Probably. But the money you save up front usually will offset the increased support costs. You'll also be sending a powerful message to Cisco that it needs to compete for your business. And like the voter expressing support for democracy, even when his or her vote is unlikely to affect the election, you'll be contributing to the broader good.

Dave Molta is Network Computing's senior technology editor. Write to him at [email protected].Post a comment or question on this story.