[ Go to February 1997 Table of Contents ]|
Companies Just May Kick The PC Habit
Personal computers are expensive; NCs are not. Which do you think businesses will choose?
Lou's First Law of Computers: A good economic incentive beats a royal flush. What's that got to do with whether network computers (NCs) will take the desktop world by storm? Everything.
First, some history. In the bad old days we were slaves to mainframes, and data and processing power were highly centralized. This simplified data backup and recovery, but everyone on the system shared a single CPU, which meant our programs' responsiveness was largely determined by the number of people working at the time. The operative word here is "Ugh."
Then PCs arrived, and we each had our own CPU. But the price was high; we had to manage our own data and system configuration. And we showed far less enthusiasm for this task than we did for playing Doom.
Next, we networked everything, which solved part of the problem because the important data was on servers, where it was more secure and more likely to be backed up regularly. But we were still putting a PC on everyone's desk, which meant endless configuration and virus problems. (Remember how much time you lost the last time you had to mess with an INI file-or worse, the Windows Registry?) If you doubt configuration hassles are a serious issue, ask your local school district what happens when you have hundreds of PCs, each used by several students per day, most of whom consider it their primary challenge to disable the system. One of my clients is a school district, and trust me, it ain't pretty.
NCs promise the best of both worlds-centralizing all data and configuration settings, but giving every user a state-of-the-art CPU. The most-often cited estimate of the savings in switching from PCs to NCs is 80 percent over the lifetime of the system, as juicy an economic incentive as you're likely to see. (Expect the real savings to be about 30 percent, but that's still nothing to sneeze at.) And if we have to ditch our beloved PCs for these sterile NCs to get such savings, will we? You bet, and we'll do it without hesitation, for one simple reason: Businesses consider computers a necessary evil. Give them a cheaper product of equivalent functionality, and their PCs will disappear faster than you can say "CP/M."
Skeptics say a lot has to happen for NCs to be viable. That's true. But what's so often overlooked is that the economic incentives for companies to make the switch are so strong that it would be a mistake to think for an instant they won't. Corel is hot at work on Java versions of WordPerfect, Quattro Pro and the other components of its office suite, and new intranet-related tools and specifications are announced nearly every day. Add to that Microsoft's announcement of its own NC project, plus the formation of an entire NC division at IBM, and it's clear companies of all sizes have joined the NC gold rush.
One detail I hope we get right is local disk storage. There should be none. Nada. Zippo. Except perhaps a Zip, as in a 100MB Iomega Zip Drive. Users need a sneakernet port so they can take work home, and they need the psychological benefit of controlling some aspect of their computing facilities. Floppy disks are far too small and slow. But put a Zip in an NC, and you have an excellent compromise between system security and end-user flexibility. In fact, this configuration would enhance security by allowing you to store the most sensitive data off the servers entirely, far from prying eyes. It would also allow users to carry Zip cartridges to any NC in an installation and have access to all their apps and data.
What will NCs do to the desktop world? Home PCs are safe until content providers such as cable TV companies get their acts together, which could take years. But for any school or business with more than a dozen or so networked computers, things will get very exciting. As soon as a credible selection of applications and tools is available, likely in the third quarter of 1997, hang on to your hard drives, folks. The lack of software will be the last obstacle in the NC's way. And once it's gone, the desktop computing world will change more radically and quickly than it has since IBM introduced the PC.
Lou Grinzo is the author of Zen of Windows 95 Programming (Coriolis Group, 1995). Contact Lou in the "Dialog Box" topic of WINDOWS Magazine's areas on America Online and CompuServe, or care of the editor at the e-mail addresses here. Opinions expressed in this column do not necessarily reflect those held by WINDOWS Magazine.