The Internet is destined to be the dominant form of media, but until it becomes mainstream several years from now it will be tough for publishers to make money on the Web. Too much information is being given away to make selling content viable, and the audience is still too small to captivate advertisers.

"Content is where I expect much of the real money will be made on the Internet, just as it was in broadcasting," I declared at the beginning of 1996. "The television revolution that began half a century ago spawned a number of industries, including the manufacturing of TV sets, but the long-term winners were those who used the medium to deliver information and entertainment."I still think that's true, but it certainly hasn't happened yet. Fortunately, I warned that it would take time.

"For the Internet to thrive, content providers must be paid for their work," I wrote. "The long-term prospects are good, but I expect a lot of disappointment in the short-term as content companies struggle to make money through advertising or subscriptions. It isn't working yet, and it may not for some time."

Two years later, I have an even keener appreciation for just how long it may take companies to find financial success in interactive publishing. I realize now that even when profits become possible, it isn't clear whether more than a comparative handful of Web sites will make much money by selling either content or advertising to accompany free content.

CNN, USA Today and MSNBC operate very high-volume sites, but they don't turn much of a profit yet. Eventually they will make money because rising water will float some boats.

But Internet advertising will probably have to increase by another factor of four before even the busiest sites make much from ads.

Doing well will be even tougher for sites that have modest audiences. Some will succeed in a big way, but not soon.

After offering the online magazine Slate free on the Web for two years, my company started charging a subscription in the spring of 1998. Before long we'd reached our initial target of 20,000 paying subscribers, but that's only about a tenth the number we need to make the magazine viable.

New print publications often take years to break even. Michael Kinsley, the founder and editor of Slate, and I both anticipate that profits are years away. We're in this for the long haul.

There are reasons to operate Web sites other than immediate profits. Companies publishing on the Web are staking out their turf. They are learning. They are using the Web to promote brand names and products.

Some sites are succeeding financially already. These are sites that conduct transactions and use the power of the Internet to eliminate middlemen and drive down costs. My company's Expedia site sells a lot of airline seats and hotel stays, for example.

Some Web sites have done well, at least among investors, by distributing content or guiding people to content. Yahoo and AOL are successfully building brand names and more or less breaking even financially. This puts them in a much stronger position than pure content providers.

Content has never been a particularly profitable business, except for a few leading companies and individuals. Not much money is made in books, for example. Some authors break through with their "brand" and make a lot of money but the vast majority of authors do not.

The book business is tremendously important. It makes it possible for anybody to spend a few dollars for an immense amount of information. The book industry greases the world, but few authors, publishers, distributors or retailers ever figure out how to get much out of it financially.

Some Web content companies will make serious money, but most won't.

One of the relatively successful content areas so far has been the online delivery of information about the computer industry. The Internet is an ideal way to reach consumers interested in computers, so advertisers have been willing to support some of these sites.

But the computer trade press, and to some extent the rest of the print media, have a problem. As Web publications improve, subscriptions and advertisements will begin to fall in print publications. Publishers who treat a Web edition as a marginal cost covered by marginal income will be disappointed once advertisers start to shift dollars away from print.

Online publications that don't have print editions aren't immune. They still have substantial cost structures because consumers expect news to be up-to-date seven days a week, 24 hours a day - even though the news is free.

For the biggest and best of the Web's content providers, the situation will improve a lot as the Web becomes mainstream.

Digital television will help. People will start thinking about going to Web sites once they're easily available on their television sets. Sites may capture some of the revenue going to television networks today.

Computers will get easier to use. They'll turn on instantly. Dialing up an Internet service provider and downloading pages will become unnecessary once Internet connections persist 24 hours a day. New approaches will make it unnecessary for people to remember more than one password, regardless of how many sites they connect to.

The Web content business will really get exciting when you're able to carry an inexpensive electronic tablet with you that connects wirelessly to the Internet. You'll be able to look up anything: "Let's go to a movie tonight. Oh, OK. Let's go to a restaurant."

At that point, Internet content will trump everything else. Until then, it's a waiting game.