Meanwhile, Boy 3, Microsoft, which is richer than God, talks to both sides and also makes its own pass at MediaOne, but decides it would rather be buds with Boys 1 and 2 than fight them. So it buys a $5 billion piece of AT&T, which agrees to seed part of the cable world with Microsoft software. And all three players get in bed together to cooperate against their common enemy: the tax man. Everybody gets something, except for America Online. AOL, invited into the game by Comcast’s Brian Roberts, says in public that it walked away because the deal price was way too high, but complains in Washington’s corridors of power that it’s being cut out of the action unfairly. What fun! And we’re not even going into the side stories, which are a trip and a half just by themselves.

The AT&T-Comcast-Microsoft entente over MediaOne, announced last week, represents a second giant gamble by AT&T that it can use cable wires to deliver new “broadband” services to millions of people and businesses. The total price for MediaOne and the previously purchased Tele-Communications, Inc.: $115 billion, though AT&T expects to recoup $25 billion or so by selling pieces of MediaOne to Comcast, Microsoft and other players to be named later. In addition to the fat purchase price, AT&T will spend billions to upgrade its newly acquired cable facilities, then pray that the whole thing works when it presses the button, probably late next year.

AT&T chairman C. Michael Armstrong, who’s committed AT&T to a record $130 billion of deals since taking office barely 18 months ago, says the company has no choice but to roll the dice. Its lucrative long-distance business is gradually losing market share and declining in profitability, he argues, and he had to do something to get AT&T into growth businesses. Because AT&T couldn’t make deals with cable companies to get access to their systems, it decided to buy them outright. “I’m doing this because I have to do it for the future of the business,” he says. “I’m not buying cable systems. I’m buying a distribution system to be able to implement new businesses.” To wit: local telephone service, video and Internet access.

As for Microsoft’s role in all this, AT&T agreed to put Windows CE software in some of the cable-TV set-top boxes it will use to pump the Internet into homes. That gives Bill Gates a leg up on competitors, even though AT&T insists that Microsoft isn’t getting an exclusive deal. In addition, Microsoft, which has apparently decided it’s too late to get into the cable business in the United States but has designs on the rest of the world, will buy MediaOne’s 29.9 percent stake in a European cable operator, TeleWest. AT&T cut Microsoft in on the deal in return for Microsoft’s agreement not to bid for MediaOne and to invest $5 billion into AT&T. (“I didn’t need $5 billion,” says Armstrong. “What I wanted was the endorsement by a technical leader.”) It’s another example of how Microsoft, with more than $20 billion of cash on hand, uses money not only as an investment medium but also as a tool to buy its way into markets. The AT&T investment resembles Microsoft’s $150 million investment in Apple Computer and its $1 billion investment in Comcast, which got Microsoft a preferred position as a vendor to both companies. “Unlike a venture capitalist who will invest wherever he or she can make a profit, we invest only when we can further our strategies and do well financially,” says Greg Maffei, Microsoft’s chief financial officer.

OK, almost everyone’s happy, so it’s time to roll the credits, right? Well, not just yet. The show isn’t remotely over. This is, after all, the era of dealmaking frenzy among the converging cable, phone and Internet companies. Just as AT&T is trying to use cable wires to bypass local phone companies’ phone lines, phone companies are trying to upgrade their phone systems to carry the broadband services that AT&T wants to carry over cable wires.

What’s more, because MediaOne owns a 25 percent stake in Time Warner Entertainment, which owns most of the cable systems of Time Warner, the nation’s biggest cable company, you can argue that the TCI and MediaOne acquisitions give AT&T total or partial ownership of about 60 percent of the cabled homes in the United States. AT&T says the right number to use is about 26 percent, because it has no control over the partly owned systems. This is a crucial difference, because the Federal Communications Commission once adopted (and has since suspended) rules allowing a maximum national market share of 30 percent. In fact, AT&T is turning into such a large cable player that some punsters, invoking AT&T’s old Ma Bell telephone-monopoly days, have taken to calling AT&T “Ma Kay Bell.” With the emphasis on Kay. Get it? Nevermind.

AT&T’s huge expansion into cable TV is an unintended consequence of the Telecommunications Act of 1996, which was designed to spur competition for local and long-distance phone service. In return for opening their systems, the Baby Bell local phone companies would be allowed to enter the long-distance business, from which they are currently barred. But local competition has been far slower to develop than lawmakers expected. AT&T, the nation’s biggest long-distance company, was allowed by regulators to buy the No. 2 cable company, TCI, in the name of local competition, since it will use TCI’s cables to provide local phone service. Similarly SBC, the former Southwestern Bell, is making headway with its argument that it should be allowed to buy fellow Baby Bell Ameritech to attain the size necessary to compete in local phone markets nationwide.

Critics say such megamergers were hardly what Congress had in mind. “AT&T is now asking that they be allowed to put together a monopoly unparalleled in the telecommunications world–all in the name of local [telephone] competition,” says Gregory Simon, director of OpenNet, a coalition of Internet service providers. “They’re asking everybody to look the other way while they do it, using local phone competition as a Trojan horse.” Sid Boren, executive vice president and chief strategist of BellSouth, says it’s totally unfair for regulators to force phone companies to make their lines accessible to competitors while not forcing cable operators like AT&T to do the same.

Because AT&T will own the pipes into homes, competitors fear that it will push its own Internet services. AOL president Bob Pittman predicts that regulators will press cable operators to open up their systems to the likes of AOL. “They can’t discriminate in favor of their own companies or back the consumer into a corner,” he says.

And now, to the numbers. AT&T, which is paying $60 billion in cash, stock and assumed debt for MediaOne, says it will raise $18 billion to $20 billion, mostly free of capital-gains taxes, by selling pieces of MediaOne, such as its foreign cable and wireless properties and its stake in AirTouch Communications. That’s in addition to up to $9 billion that AT&T will get from Comcast, which has the right to buy as many as 2 million MediaOne customers. Rather than paying AT&T $9 billion in cash money–which it doesn’t have–Comcast has the right to make its purchases with some $4 billion of AT&T and AtHome stock that it owns, and with $5 billion of securities it would issue to AT&T. Swapping the AT&T stock in return for cable systems would be tax-free. Swapping AtHome stock would be taxable. Comcast has already gotten a $1.5 billion breakup fee because MediaOne succumbed to another bidder. That fee, alas for Comcast, is taxable.

One of the parlor games here is to figure out how much Microsoft is actually paying for its AT&T stake. On the surface, it’s $75 a share–the price at which it can convert its $5 billion of new AT&T securities into AT&T stock. That’s almost 25 percent above Friday’s closing price. But look closer. The $5 billion of securities carry an interest rate of 5 percent a year, far higher than the dividend on AT&T’s stock. And, Microsoft gets 40 million stock-purchase warrants, valued at $300 million to $350 million, as part of the deal. And, Lord only knows what kind of games there are in the Windows CE sale and in Microsoft’s aforementioned purchase of MediaOne’s TeleWest stake. AT&T claims it’s getting an effective price of more than $60 a share. Microsoft isn’t talking price, but its partisans claim the web of transactions produces an effective price in the $50 range.

AT&T, which originally said that buying MediaOne would reduce its earnings per share by 25 percent, now says the dilution is down to 10 percent. Some three quarters of the reduction consists of cutting costs $2 billion a year. Read: “firing people.” Which AT&T could have done even without buying MediaOne. The rest of the reduction comes from the Microsoft deals and other financial factors.

AT&T says that despite the high price it’s paying for MediaOne, the deal will pay off because the AT&T brand name will add about 8 percent to MediaOne’s customer base, and local phone service, Internet and video will be very popular. In addition, AT&T will save major moolah if it connects AT&T long-distance customers to AT&T-owned local cablelines. That’s because bypassing the local phone company will allow AT&T to save so-called access charges of about 1.5 to two cents a minute that it has to pay when one of its customers makes or receives a phone call over wires owned by a local phone company. (The total access charge, on both ends of the call, is three to four cents.) Will this actually work? No one knows. If AT&T can produce what it’s promising, chairman Mike Armstrong will go down in history as one of the biggest business geniuses ever. If it doesn’t work, he may well split AT&T into separate long-distance and broadband operations, and declare victory. He didn’t get where he is by not having fallback plans.

So there you have it. The Wall Street chapter of this financial soap opera is winding down. But the Washington part, consisting of trying to keep federal regulators at bay while setting them on the other guy, is just starting up. As is the tax avoidance. As is a financial smoke-and-mirrors game. Moral: “Melrose Place” may be going off the air, but soap operas are still going strong.