Let's start with pencils. What are the attributes that have made the pencil so successful? Simply put, it's easy to use and it's cheap. The problem is, the personal computer is no pencil. It's not easy to use, it's not easy to program ... it is extremely expensive. It's time for a new device -- the network computer, or NC. Think of it as an electronic pencil.

-- Excerpts from a speech by Larry Ellison at the Commonwealth Club of California, March 1996

In the beginning there was the pencil. Then, at last, the PC was born -- and so, too, the need for a software platform to release its full potential. Enter Microsoft Windows. The Windows operating system transformed the PC into the ultimate network-computing device.

-- Excerpts from a full-page Microsoft ad in The Wall Street Journal, March 1997

Microsoft is infamous for reinventing technology and presenting it as innovation. So I suppose it shouldn't be surprising that Microsoft has stooped to co-opting speeches, as well. But, please. "The Windows operating system transformed the PC into the ultimate network-computing device?" Is this the same Windows that barely understood NetWare until Windows 95? The same Windows -- with its ActiveX and Internet Explorer -- that has such sloppy security that it couldn't protect your computer from harsh language?

The ad, which accompanied Microsoft's announcement of its NetPC, goes on to credit Microsoft with the innovation of multimedia and mobile computing. I can forgive The Wall Street Journal for ignorance on technical issues. But if, like InfoWorld, the paper has a policy to pull ads that it knows are false or misleading, the Journal should have remembered enough PC history to kill this one.

Dazed and confused

By using the phrase "ultimate network-computing device," Microsoft is attempting to co-opt the definition of the Network Computer rather than build one.

Let's clear up some confusion right now. Here are some of the features the Zero Administration NetPC will include: automatic pre-boot configuration; nightly maintenance; automatic upgrades, scans, and policy enforcement; remote problem resolution; security restrictions; diagnostics for monitoring and predicting failures; a remote wake utility, Web-based management, messaging-based management, and developer kits. Here are the administration features built into the Network Computer: a network connector. You tell me which one is the "zero-administration," ultimate network-computing device.

The fact is, Microsoft can morph the PC, but you will always need to manage those clients at the desktops and server. In a centralized network-computing environment, the server never has to know anything about the clients. It spits out standard protocols and standard, platform-independent applications, and you can connect any client that understands them.

From a management standpoint, this makes network computing the ideal migration strategy. You can install a bona fide Java-based NC right next to a 286 or Pentium Pro. As long as the PC is running TCP/IP, HTTP, and Java, the server won't have a clue which is which. In contrast, plop a 486 running Windows 3.1 next to a NetPC and see what the dozen or so Zero Administration management features on your server think about that.

The problem with the NetPC is that Microsoft can't respond to the threat of the Network Computer by reproducing the competition's technology. A NetPC can serve as an NC, but the Windows portion adds a proprietary management burden and security risk without adding any value to the network-computing model. To truly compete on the NC's turf, Microsoft literally has to remove what we now know as Windows from whatever comparable client it "invents."

The tables have finally turned. As long as Microsoft owned the desktop, no other company could compete on a level playing field. Consequently, Microsoft created products and strategies to crush competition, not to serve the customer. The customer has always been secondary. Witness the fact that it only owned up to the expense of managing Windows PCs when the NC threatened to kick the PC's butt.

In the coming era of network computing, however, vendors can only compete on a level playing field. Quality, features, and price will be the differentiating factors among products that will otherwise be completely interchangeable. If they're not, they'll lose.

I'm not saying Microsoft has never delivered anything of quality, but until now, it has never really had to. Unfortunately, judging by the NetPC, Microsoft isn't about to change. Instead, Microsoft is simply going to confuse the issue with marketing. Larry "Edison" invented the light bulb and launched a new era, leaving Microsoft stuck with a warehouse full of candle wax. Rather than work on a better light, Microsoft is shaping its wax like a bulb and hoping you don't notice that nothing happens when you flip the switch.

Enlighten me with your views. Send me e-mail at nicholas_petreley@infoworld.com, and visit my forum on InfoWorld Electric at http://www.infoworld.com.

Source: InfoWorld