America Online's purchase of Time Warner is supposed to represent the victory of the Internet--a "triumph of new media over old," chirped the San Francisco Chronicle. "The nerds have won," it said. A front-page story in the New York Times portrayed Stephen Case, AOL's chief executive, as a stubborn dreamer who prevailed through a decade when no one believed in his "grand vision."

These independent news organs (not yet owned by AOL or Time Warner) are right: Something has indeed triumphed and prevailed. But it's not the Internet of lore that has succeeded, that scrappy place where everyone was supposed to be a publisher. That Internet is dead. More precisely, it has been dead for a while. The AOL Time Warner mega-deal is simply its very public funeral; and this, I suppose, is an obituary.

The Internet was not Stephen Case's vision; it was not created by venture capitalists in Palo Alto, Calif. It was conceived by the U.S. government--born in the 1960s as a project of the Department of Defense, to be a network for sharing computing power. The network was soon opened to university researchers, then computer industry engineers, and by the 1980s it was poised to become something altogether new: a government-created and -funded infrastructure, run day-to-day by an unofficial coalition of dedicated volunteer programmers and made available to the general public. It was here that I first encountered the Net, working as a programmer in the 1980s, sending e-mail to colleagues at other companies, learning of news and research projects, overhearing engineers more senior than myself imagining the future of this amazing resource that our government had turned over to us.

What, indeed, would we do with the Internet? For a very few years, anything seemed possible. Whatever we were going to do, however, certain principles evolved. The Internet would be a multipolar world, we thought; no one group or organization would control it. Oversight, where necessary for technical or political reasons, would devolve to the smallest, most local body, composed of the people most affected by the outcome. It was to be a plasma through which all manner of ideas could flow freely. And it would be a global plasma: Tim Berners-Lee, inventor of the Web protocols still in use today, named his project the "World Wide Web" when there were only three connections to it--a testament to his grandiosity, perhaps, but also to a brave kind of hopefulness.

The Internet was to be a vast level field, where all ideas and ventures were equally in play, and its emblem became the search engine: the means by which anyone was supposed to be able to find anything. Here, theoretically, the empowered individual could search the planet, all its wares and peoples, for what he or she uniquely desired.

Of course it couldn't last. The very ideas that fueled the growth of the Net--its libertarian ethos, inimical to regulation--also led, perhaps inevitably, to the AOL Time Warner deal. For the Internet was let loose without oversight or protection of its early principles, and what followed was nothing less than the frenzied, unregulated privatization of a valuable public resource. By 1994, the Internet had become a sales medium, its ownership ceded by default to commercial interests.

You only have to look at AOL's home page (or Yahoo's, or one of Microsoft's) to see the change. Every square pixel of the screen has been sold off. It's not just those terrible migraine-inducing blinking ads. It's also the deals that decide which online stores will be available where, what suppliers will even be surveyed by those "impartial" price-comparison sites--the whole range of product-placement and payola that determines the content of a Web page, the new and invisible techno-middlemen hiding in the code.

Even the search engines are not what they pretend to be. They do not search the entire planet; they will not find that home page where each person becomes a publisher. A recent study by Steve Lawrence and C. Lee Giles of the NEC Research Institute found that the most far-seeking search engine examined only 16 percent of the searchable Web. Yahoo covered only 7.4 percent, for example, Lycos a miserable 2.5 percent. And the methods used by the engines are biased in favor of larger sites: The more links a page has to it from other sites, the more likely it is to be found. This might argue that the sites covered are the ones of the highest quality, since other sites find them interesting. But another way of reading the study is to say that big sites get bigger yet; and new, small ones stay invisible for months, if they are found at all. Yes, you can still say almost anything on the Web. The question is: Will anyone hear you?

By this logic--by the logic of the new, big, commercial Web--AOL's move to acquire Time Warner is the right thing for the company. It means that AOL can build a universe of pages all linked to one another, ensuring their route to the top of the search results page. The merged company can control a nice vertical slab of the business: from desktop software, to cable lines, to content on the server, a vast culture conglomerate. By the principles of today's Internet, AOL has to compete with Microsoft, a behemoth, and therefore must be a bigger behemoth yet.

But let's not confuse AOL's perhaps-inevitable business decision with the success of the original Internet. That Net was supposed to be a world where power was diffused widely and finely. No one was supposed to control it. Let everyone debate AOL's cleverness, if they wish. But I'm going to at least remember the lost opportunity of the public Internet. I'm going to mourn its passing, and keep its memory clear in my imagination, where neither AOL nor Time Warner nor Microsoft can own it.

Ellen Ullman was a programmer for 20 years. Her essays about the profession of programming and its social implications have appeared in Harper's, Wired, Salon and the New York Times, as well as in several collections. She is the author of "Close to the Machine: Technophilia and Its Discontents" (City Lights).