CNET también está disponible en español.

Ir a español

Don't show this again

Christmas Gift Guide
Tech Industry

CIOs unplug from wireless apps

Chuck Phillips explains how tougher economic times means taking on tough-minded decisions about which projects stay and which wind up on the chopping block.

If wireless is about to take over the world, no one has informed the CIO community about the coming revolution.

In fact, surveys of chief information officers over the last five months suggest they have pushed wireless applications down on their priority lists. As a BlackBerry pager addict--and long-time gadget head--that has left me a bit puzzled. But despite a heavy push from the telecommunications industry, the killer wireless application clearly hasn't yet made itself known to CIOs.

In searching for an explanation, I've spoken with a few CIOs. Some issues surrounding wireless applications following these discussions became clear.

• As the economy slows and cash becomes scarce, new-age experiments and nice-to-have projects are first to go on the chopping block. Companies often get conservative, and make only incremental investments to existing infrastructure. New projects must be more closely linked to cost reduction or revenue growth. Offering Joe Middlemanger more frequent access to his calendar, for example, just isn't going to make the budget.

• Enterprise applications are so complex and heavy that pushing something meaningful down a fairly thin pipe is often impossible. The bandwidth needed to derive value from an enterprise application simply overwhelms current wireless networks.

Remember that it took years of tuning to get these applications to perform adequately in local and wide area network environments. For years, the knock against client/server applications was that its excess weight on the network resulted in sluggish performance.

An entire industry was formed just to optimize the performance of these applications for the network--and companies such as Citrix made a killing slimming these applications back down for a corporate network. Now, moving these applications to a thin client environment attempts to shove an even wider load through already clogged pipes.

• Text Expectations are high. People have grown accustomed to instant response time, unlimited information and a customized experience on a color, high-resolution screen. To revert to a 4800 baud connection on a small, dim, black and green screen means the application had better be important.

Some are but many aren't.

And don't even mention trying to interact with a real application from a cell phone keypad--just slit my wrists. Yes, the flurry of new devices is always interesting and I've tried them all--at least for a month, anyway. My graveyard of devices is quite gruesome looking, half of which I'm still getting charged for--I just don't know which half.

The construction of broadband wireless data networks will take a long time. The infrastructure will have to be substantially upgraded for broadband to bend around buildings and provide global coverage. It's a different gig to get wireless to the next level of throughput.

Problem is, the money isn't there right now. No one wants to fund a big telecom investment right now given the low returns and intense competition. In fact, many of the existing wireless networks look questionable from a financial viability perspective.

Sure, low bandwidth connections work for some enterprise applications. But coverage is still an issue. The problem is that corporate enterprises are moving in the opposite direction.

Businesses need more data in real time and they want analytical insight to come with it. No wonder e-mail remains the most popular wireless application. That shouldn't be a surprise since it was the original killer application for the consumer market as well. But e-mail is still a powerful application and can accomplish a lot for some companies with well-profiled alerts.

Companies are still holding out for the one-pound notebook computer with a built in global wireless connection--including on a plane, simple VPN, Windows burned into firmware for "instant-on," a three-day battery life, and remote diagnostics and maintenance. Besides those small details, we're almost there. I'm still a believer in the potential for wireless data to transform computing.

But it's going to take years and maybe a new approach.

Enterprise requirements are predictably more robust. Consumer wireless applications are easier to conceptualize and often simpler. I'd still like New York City to hit my pager if there's a water-main break on my street or my car registration is up for renewal.

Several sites already will inform me when B.B. King or Wilson Pickett comes to town. No security problems, and the information doesn't have to be fed in real time. But for the enterprise to pull similar tricks, it might take a while.

In the meantime, my BlackBerry will remain glued to my hip 24-7.