I use a cable modem for my home Internet service these days, and, quite frankly, I'd be angry if I had to part with it. It's pretty reliable, it's on all the time (no more screeching dial-up), and it's blazingly fast.

With this cable connection, I have to use Time Warner Cable as my Internet service provider, or ISP. My Internet access is bundled with my cable modem connection, which means there's no evidence of an extra charge for Internet access. But if I wanted to buy service from another ISP and use my cable modem, I couldn't.

That's because cable companies are not regulated by the government as common carriers, a phrase and a rule that does apply to telephone companies. Common carriage means that telephone companies are required to carry over their lines anything the telephone system supports and without their intervention between the caller and who is being called. The telephone company can't favor one grocery store chain over another, for example, and route customers' calls to the preferred store.

In other words, a common carrier cannot control both the wires and the content on those wires.

But the cable industry has escaped this regulation because, first, the delivery of TV channels through a cable wire was considered a natural monopoly, because no community wants many different cable companies snaking their multiple wires on poles or under the ground; and, second, because being able to sell the full range of TV in a specific community was viewed as the economic incentive to invest in and build a cable network in the first place.

Now comes the complicated part: the Internet. Everyone wants the Internet to be faster and less congested. The solution is broad-band telecommunication connections, a technical and industry term for technologies that are much faster than what you can get with a modem and a telephone wire. The two leading broad-band media right now, for consumers, are cable and DSL (for digital subscriber line), a service that allows the telephone companies to get more speed out of the copper wire that's already in your house.

The reason these two particular technologies are leading right now is fairly obvious: They don't require rewiring large parts of the U.S. Most homes have telephone wire that can handle DSL, and more than 70% of U.S. households can be served by cable. This means that the battle is on between big telephone companies and big cable companies for an unbelievably huge market.

But what we're seeing is telecom's version of the old African saying, "When elephants fight, the grass gets trampled."

First, we're seeing a wave of mergers, characterized by the recently proposed acquisition of the cable company Tele-Communications Inc. by AT&T, and big deals like the pending partnership between America Online and Bell Atlantic. The TCI-AT&T deal is further complicated by the fact that TCI owns a large share of the Internet service company carried on some cable systems, called @Home, and @Home itself just agreed to pay more than $6 billion for the Internet portal and search engine company Excite.

A few heads started shaking in Portland, Ore., over the implications of these kinds of deals. Portland and surrounding Multnomah County are served by TCI, which asked for a license transfer to AT&T, pending the two companies' merger (which still has to be approved by the Federal Communications Commission). Both the city and county governments insisted that for the license transfer to go through, they would need assurances that AT&T would open its cable system's Internet access to competitive ISPs. AT&T said no way, so Portland refused the license transfer. TCI and AT&T promptly sued the city of Portland and Multnomah County.

The city of Los Angeles recently faced the same dilemma. TCI is not a big cable carrier in L.A., but its license, and the question of open access on TCI networks in Los Angeles, would be important precedents as the city begins reviewing all its cable franchises later this year.

The city took a Solomonic approach 10 days ago, opting for a resolution that will require AT&T to abide by any rules adopted by the FCC. The city also began a 90-day study on the future of open broad-band access in Los Angeles, a study that will be conducted by the city's Information Technology Agency.

Last Thursday, the FCC punted the whole issue by saying that no changes in broad-band access are required at this time, a conclusion that must certainly be related to the agency's no-win, high-pressure, nail-biting and head-whirling situation.

Open access to broad-band services is not only threatened by cable companies. Telephone companies offering DSL connections are also trying to lock customers into their own information services, and hence their own partnerships with content providers, like AOL. Not a single independent ISP in Texas can offer DSL yet because Southwestern Bell makes the wholesale price for DSL higher than what they charge their own individual customers.

Why should anyone care about this? There are several issues at stake. First is that the Internet doesn't have to work this way, and in fact shouldn't work this way. We already have about 6,500 ISPs in the United States, which must be the definition of competition. They offer a wide variety of services, prices, levels of support, etc. But most of them could be wiped out in a few years if present trends continue.

Second, if you have to buy an information service provider when you select what wire you want in your house, you're going to be looking at the user interface of a huge, monolithic, vertically integrated corporation. Your first encounter with the Internet is likely to look a lot like walking into a shopping mall -- boring, redundant and absolutely writhing with advertising. It is true that you can buy e-mail from someone else, or change the home page in your browser, but most users don't know this.

Third, this competition -- really just a battle among a few titans -- means the U.S. will overbuild its telecommunications infrastructure. We'll wind up investing tens of billions of dollars in redundant, underused or even never-used network wires. It's as if we'll be building highways on top of highways, 10 deep.

That means the cost to consumers to pay off this redundant investment will be unnecessarily high, simply because we refuse to think of telecommunications as public infrastructure. We could have one wire into our homes or under our streets -- a fiber-optic cable -- that could do everything. Then we'd have competition among a million different services, the way things should be, instead of having to choose between two or three leviathan telecom companies and their lucky partners.

Of course, given the way things are going in the U.S. these days, bigness and blandness is probably what we'll get. Unless, of course, citizens get mad and offer to help the FCC and local and state governments fight better, smarter and harder.

Gary Chapman is director of The 21st Century Project at the LBJ School of Public Affairs at the University of Texas at Austin. The 21st Century Project focuses on increasing public participation in science and technology policymaking.

This column, first published in The Los Angeles Times is reproduced here with the author's permission.