You'd have enjoyed it. I wish you'd been there.

A couple of weeks ago, I was at dinner with a number of friends who, like me, have been known to do some prognosticating, and who have,

umm

... shall we say, a few opinions about things.

About halfway through the meal, one of our number suggested that we do a little dinner-table opining for one another: What, he asked, was for each of us the most outrageous, unexpected thing each of us was willing to forecast? Something way out on the edge, but at the same time, something we really believed probably lie ahead. Say, 70% or better odds that it really would come true.

I love this kind of stuff -- both the insane suggestions my friends in this business make, and also the chance to test on them some of the farther-out thoughts I've been harboring about the directions tech, and tech companies, may take.

I didn't go first -- hey, the smart guy always takes a dive for a while in these things to see what else others are going to put on the table! -- but I jumped in somewhere around fourth or fifth. And, true to the spirit of this game, the table fell silent after I said my piece:

"Dell's going to stop making PCs."

A few giggles, a few blank stares, a few whispered conversations. I could imagine their whispers: "He's gone overboard this time." And "Nuts, just nuts."

And then someone said, "So ...

explain

!"

So I did.

I think this one, which I see as 80% probable sometime over the next three years or so -- and maybe, much sooner -- arises from the confluence of several rivers of opportunity for

Dell

(DELL) - Get Report

, and also from the fundamental changes taking place in the computer business.

My reasoning:

    We all know that boxmaking is increasingly a low-profit, even profitless, activity. That isn't going to change. The real money is, increasingly, in servers and services - the "S&S" you're going to hear me harping on a lot in the future. But selling ordinary desktop PCs has real advantages for boxmakers, including increasing gross revenues, driving down component costs through greater purchasing volumes, and locking in customers to a single-source supplier. Laptops already are a big and increasingly important part of Dell's business and of its reputation -- but it doesn't make them. It helps design them, then buys them, complete, from original equipment manufacturers in Asia (as do virtually all of its competitors). Dell is increasingly having trouble growing fast enough, physically, to stay on top. It has created a second major U.S. manufacturing center, for example, in Nashville, Tenn., with 2,500 employees, because it simply can't get enough new employees in the Austin, Texas, labor market, where it already has absorbed something like more than 22,000 of the best workers. Dell dollars allocated to space and people in management, marketing, planning, investments, design, services and other nonmanufacturing jobs have a far bigger payoff for the company than do dollars invested in production lines and assembly workers. And now, for the really big one: Dell may have reached the limit on its build-to-order, inverse-working-capital, no-finished-goods-in-inventory model.

Nearly every observer agrees that Dell's brilliant and obsessive focus on build-to-order, including forcing its suppliers to stockpile uncharged-for components just outside Dell's factory doors, or in vendor-run "stores" right on the factory floor, has been the key to its explosive success.

Compaq and others have struggled to match Dell's success here, but none has yet matched it. Dell's got the cycle down to about five to six days now. That's the interval between the time it receives your order over the phone or Web and the time it ships the finished computer to you.

Which is impressive.

Until you look at

Apple

(AAPL) - Get Report

. Whatever you may know about Apple's Mac hardware and System X operating system, or think of Steve Jobs' role in cranking Apple's stock price up from 16 bucks at the end of 1997 to the 130s this week, what you probably

don't

know is that Apple has virtually stopped making its PCs, and has farmed production out -- in the process cutting that "Dell cycle" to between two and three days.

In other words, twice as good as Dell.

Could Dell match that? And what was the reaction of my skeptical colleagues? Click here to see the rest of this column.

Jim Seymour is president of Seymour Group, an information-strategies consulting firm working with corporate clients in the U.S., Europe and Asia, and a longtime columnist for PC Magazine. Under no circumstances does the information in this column represent a recommendation to buy or sell stocks. At time of publication, neither Seymour nor Seymour Group held positions in any securities mentioned in this column, although holdings can change at any time. Seymour does not write about companies that are current or recent consulting clients of Seymour Group. While Seymour cannot provide investment advice or recommendations, he invites your feedback at

jseymour@thestreet.com.