If you work in what has heretofore been known as "big media," the apocalypse is nigh. Reluctantly tossing aside one of the great wealth-creating machines of the century in favor of door-to-door direct marketing of gizmos and doo-dads, the former purveyors of entertainment to the masses now believe their value now lies mainly in electronic commerce. And it's all Barry Diller's fault.

No, not because of the Lycos deal. We'll get to that in a minute, but trust me, that's just the latest symptom of a long-running trend that spells doom for the big box media companies. No, let's go back to 1993 -- the year the first big media mogul fell in love with electronic commerce. That was, of course, the visionary and opportunistic Diller. Seven years ago, Diller was fresh out of a stint making Fox into the fourth television network and wanted to build his own fortune, instead of playing with others. The collective world of big media gasped when Diller did the equivalent of forsaking the elegance of '21' for a booth a Joe's Diner when he joined QVC.

That year, Diller penned an essay for Wired that ripped the guts out of the old network television model: "We become slaves to demographics, to market research, to focus groups. We produce what the numbers tell us to produce. We chase formulas for ratings or circulation." Visiting the studios of a TV shopping network, Diller had fallen in love for a new kind of media number once consigned to Readers Digest records: direct sales. Even in those days just before the explosion of the Web, Diller recognized that network TV was a dinosaur, that its numbers were flimsy, and that emerging technologies would soon be able to track actual sales, rather than just audience numbers.

Flash forward to this week: Diller essentially completed the assembly of a digital age media network by merging with former Web golden boy Lycos (pending approval by a reluctant David Wetherall of CMGI, whose run of exploiting huge Net upsides may be at an end), and adding a Web portal to a group that includes online community guides, home shopping, local TV news, cable TV entertainment, and ticket sales. In Diller's new world everything works to drive sales. The TV stations and USA Network drive sales through the Web; the Web sites drive sales through Home Shopping Network; TV and the Web drive sales through Ticketmaster. The Diller network reaches 70 million homes via TV, 30 million users via the Web, and takes a million phone calls every day. It's a neat circle, and Diller can skim a buck from every click. It's a plan that takes the revenue-driving power of HSN -- a billion dollars in revenue from 80,000 shipments daily -- and makes it even more friction-free using the Web.

Advertising? That's chump change, compared to the promise of taking a big piece of direct sales -- or at least, that's the way the thinking goes. That's why buying one of the TV networks isn't in Diller's plans, at least for now. He views them as overpriced vestigial media organs, too expensive to bring in millions of undifferentiated viewers sitting in front of 'E.R.' with no compulsion to buy. And indeed, the networks know they're in trouble -- how else to explain the comment by NBC honcho Tom Rogers at the State of New York New Media forum last week when he said that in the new media age, the value of a broadcast network will be to drive users to its other properties?

So far, the marriage of Ticketmaster and Citysearch has produced decent revenues, but nothing earthshaking. The company did $13.6 million in the 4th quarter of 1998, but its prospectus doesn't predict profitability. Adding Lycos and its sub-properties (if the deal goes through) brings in a huge number of users. But look at the greatest Internet e-commerce player, Amazon.Com, which has yet to show a profit and operates under ever-narrowing margins.

Let's face it, the Internet has the potential to turn Barry Diller into Willie Loman. Despite all the hoopla over e-commerce, direct sales is a mature, well-studied, well-documented proposition. Response rates can be predicted down to a tenth of a point. Profit margins are calibrated in advance, and it takes lots of heavy lifting to make the number. Internet retailing throws those margins and calibrations all to hell. Not because they don't work, but because of the Net's "we'll sell you a $10 gizmo at Gizmos.com for $9" model; and the insane valuation of Net-based companies throws modeling out the window.

And that's why Diller's model won't work, at least not for a long time. So far, the Web is bad for business; almost every big e-tailer operates at a loss; competition for users is fierce; and prices drop the second they hit the Net. And there's something else -- the nature of the Internet allows you to capture eyeballs quickly and relatively cheaply, yes. But it also allows users (encourages them, actually) to slip away, like so many grains of sand on the beach.

Selling, even to a brilliant new network of wired users across platforms, may not pay the bills for the first big salesman of the digital age.

LATEST NEWS

Microsoft's Dynamics ERP to Gain New Services
Barnes & Noble's e-Reader Nook Sold Out Already
Memory Market Due for Big Shift in 2010
Microsoft: No 'Back Door' in Windows 7
Tech's H-1B Hiring Faces 'Employ America Act'

* Tom Watson is co-editor and publisher of @NY