Buy a Packard Bell television, circa 1960, at a garage sale. Take it home, plug it in, watch "The X Files."

Visit your grandmother. Use her telephone, built in 1965 by Western Electric, to check your bank balance without ever talking to a human.

Load a Cobol inventory program from 1967 in your Pentium computer. Better still, load Windows 95 on your Compaq computer from 1984. In each case, watch nothing happen on the screen.

In these three examples, we have the essence of the dilemma facing the computer, communication and entertainment industries as they come to terms with the digital convergence. In television, where content has always been king, backward compatibility is vital. Your digital TV in 2003 will carry "I Love Lucy." For the telephone system, where the network is already a computer, customer equipment is so trivial it can be bought for $20 at Wal-Mart. Yet the personal computer business is based on expensive hardware that dies every three years and is replaced by new hardware that is just as expensive and is typically incompatible with the previous generation.

These are unlikely partners, yet digital convergence requires the talents of all three, despite the fact they hate each other and each wishes the others would die.

What's driving the convergence is market maturity, not customer demand. There have been no petition drives from viewers asking for High Definition Television. The television makers have just run out of new customers and want to sell everyone a new TV. The telephone companies already have every potential customer signed-up and need new services to fuel continued earnings growth. Broadcast and network equipment manufacturers want to sell new stuff to their existing customers. And the PC industry, having already stuck a computer on every desk and in the easiest 40 percent of American homes, needs new customers, even if those customers can't imagine what they'd ever do with a PC.

Enter digital television, the holy grail and hula hoop of digital convergence. It's a workstation-class computer disguised as a TV. It's a global communication device. And it will carry "I Love Lucy." But only if the TV, phone and PC people can get along.

Right now, the fight is over technical standards and philosophies: Will digital television be a computer pretending to be a TV or a TV acting like a computer? This is Microsoft against the world. If Bill Gates is to take his control of the system software standard into digital television (and he is determined to), it requires that future equipment be computer-based. But television technical standards have never had revenue attached to them. There has never been a $50 fee to the National Television Standards Committee (NTSC) in exchange for the right to build each television set compatible with U.S. broadcasts. Microsoft would like there to be such a fee payable to Redmond, and in fact, needs there to be such a fee to stay on track for global domination.

This whole idea would be ludicrous if Microsoft didn't have the support of Intel, Compaq and the rest of the PC industry, which has long relied on Bill Gates to provide any napalm needed to burn their way into new markets. A $200 billion PC industry that reinvents itself twice per decade has to scare the bejeezus out of the sleepy television business, which still looks to General Sarnoff and Desi Arnaz as its twin deities.

The only thing standing in Bill Gates' way, in fact, is his own greed. Microsoft could own the digital TV software market except that the kids in Redmond are requiring software licensees to sign onerous patent cross-licensing agreements as a condition for getting to use Microsoft software. Imagine Sony, with its thousands of patents, having to indemnify Microsoft for infringing any of those patents. This is the only condition that's keeping the patent-crazy and Gates-fearing Japanese TV industry, which is to say the world TV industry, from giving Microsoft carte blanche.

And if the TV business is sleepy, the phone business is comatose, planning each new move more a decade in advance. The phone companies have spent 20 years building digital networks so the TV industry could spend 10 years arriving at a digital televison standard so the PC industry could decide a year ago to steal the whole thing.

The phone companies are in especially bad shape and don't even know it. By the end of this year, data traffic on the U.S. telephone network will for the first time exceed that of voice traffic. There will be more bits carrying Internet packets than phone conversations. And the phone companies like this, because optical fiber is cheap and plentiful and they'll fill it up with any old photons as long as they are paying photons. What the phone companies haven't done, however, is make a bit-for-bit comparison of digital technologies.

Here's a word problem. As AT&T showed years ago when it asked to be broken apart to get the Feds off its back, the real money is in long distance service. Phone companies and Internet Service Providers both use optical networks. In fact, they use the same optical networks. But phone companies and Internet Service Providers take different approaches to how they use their digital networks. ISPs switch packets while phone companies switch circuits, the result being that while a phone company can cram exactly 768 phone conversations on a DS-3 line, an ISP can carry more than 6000 conversations on the same line. If a bunch of ISPs declared themselves to be long distance carriers and got in a price war with AT&T, MCI and Sprint, who would win?

But the television and cable TV and telephone people still have the advantage of real estate. They own the broadcast licenses, the cable systems, and the millions of miles of glass fiber, coaxial cable, and copper wire. Control of this infrastructure makes them confident, even giddy. Every TV broadcaster, for example, sees himself suddenly in control of a 20 megabit-per-second digital datastream that can be used for TV, video games, web surfing, whatever. Every telephone company sees their familiar and reliable billing relationship with millions of customers as the key to electronic commerce. Every cable TV magnate expects to use his system to steal phone and Internet customers from the others.

Yet the television, telephone and cable folks have no designs on Microsoft other than a vague concept of diskless network computers -- a concept they didn't even invent. This will be their undoing. While he plots to subvert their infrastructure advantage, Gates keeps waiting for the counterattack that will never come. It's already too late.