Chris Gulker column

Is there really a market for low-utility computers tied to (probably) proprietary and (no doubt) expensive servers?

Imagine a computer that's real cheap, say pounds 295. Good, you say. We like cheap computers. Now imagine that the computer in question has no hard drive, no floppy drive, limited RAM and comes in a sealed case so you can't change it.

Furthermore, these cheapies are useless unless they're hooked up to a network and a server, and only the server administrator dictates the software you can use. Worse, each particular kind of computer will only work with a particular kind of server, sort of like TV sets that only receive stations using a certain brand of transmitter.

These machines take 20 minutes to an hour or longer to boot. They require a network connection and their performance is almost entirely dependent on available bandwidth - a miserable situation, since networking is the only computer subsystem (CPU, RAM, hard disk et al) that isn't surging wildly in power as it simultaneously plummets in price.

In fact, networks everywhere are choking on exponentially-growing traffic. Just ask any Internet surfer.

Not surprising, you say. After all, you get what you pay for.

Indeed. But the value proposition above is precisely what's proposed by one or another of the many devices referred to as Network Computers, or NCs.

If you hadn't noticed, the Network Computers techno-hype is rocketing in exponential, and inverse, relation to its price and utility.

Help me, here. Is there really a market for glorified terminals, low- utility computers tied to (probably) proprietary and (no doubt) expensive servers?

So why have no fewer than 12 companies signed up to build one, including Hewlett-Packard, Compaq, IBM, Oracle, Sun, Acer, RCA, Zenith, Intel and Microsoft? Wasn't it the severe limitations of the old mainframe/terminal paradigm that drove us to embrace the PC in the first place?

Do computer users really want slow, feeble hardware that puts them and their data (memos, spreadsheets, e-mail and, more importantly, love letters and computer games) at the mercy of some shadowy Big Brother server out there somewhere?

Probably not. But computer users aren't the reason for all the interest in NCs. The real reason is accountants. More precisely, the reason is that accountants (aka bean counters, pencil-pushers) regard computer ownership about the same way as they regard expense account items like business conferences in Tahiti.

True, PCs are relatively cheap to buy. A basic business PC doesn't cost, in constant terms, a whole lot more than a sturdy typewriter once cost. And the same PCs will readily do chores that once required expensive mechanical machines.

But PCs are very, very expensive to maintain - even Bill Gates admits to that. Studies show a $2,000 computer can easily cost $6,000 annually to support. But modern, competitive businesses can't afford non-productive workers. Businesses have to keep their PCs going.

Business tools were once a lot cheaper to run. There were no typewriter support persons in the old days, no upgrading to Ribbon 95, Ribbon 98 or Ribbon NT.

Owning complex PC systems today means hiring legions of expert, expensive people who possess the singular skill of keeping devilishly complicated machines from blowing up and taking days, weeks or months of work with them.

They're MIS people (short for Management Information Systems), also called nerds, geeks, dweebs, propeller-heads and sometimes much worse. But not usually to their face - not by anyone, especially any accountant, who values the contents of their hard drive, anyway.

Bean counters don't like 'em because MIS people don't bring in sales, don't add to revenue, don't push up the stock price - in short, MIS people are a complete drag on the bottom line.

Adding insult to fiscal injury, MIS earns a lot of money relative to other, more visibly productive workers, like, for instance, accountants.

Bean counters can't understand the lingo MIS speaks, and have a hard time arguing with their budgets.

Bean counter: "Can't we make do with a less expensive, er, object-oriented software upgrade module for our multiplexed, switched, TCP/IP WAN adaptor?"

MIS person: "No."

Bean counter: "Oh."

In Bean Land, MIS is regarded as the techno tail that wags the corporate dog. From Accountings perspective, a really good MIS director would reduce their department size to as close to zero as possible in the pursuit of lower costs.

In the Real World, corporate directors strive mightily to command as large an army, and budget, as possible. Hence, MIS usually proposes reducing costs by making large investments in new technology, all of which requires expanded support, of course.

Enter the Network Computer. It's cheap, which pleases the bean counters. It's new technology, and will require lots of complicated systems and, of course, support, which pleases MIS.

Memo to stockbroker: I think we have a winner heren

cg@gulker.com

News Page: http://ww2.gulker.com/