The 1990s hadn’t gone as expected. A bad recession kicked off Gen X’s adulthood, along with a war in the Middle East and the fall of communism. Boomers came to power in earnest in America, and then the lead Boomer got impeached for lying about getting a blow job from an intern in the Oval Office. Grunge had come and gone, along with clove cigarettes and bangs. The taste of the ’90s still lingers, for those of us who lived it as young adults rather than as Kenny G listeners or Pokémon-card collectors, but the decade also ingrained a sense that expressing that taste would be banal, a fate that the writer David Foster Wallace had made worse than death (I swear he was cool once, along with U2).

Such was the crucible in which the computers were forged. Not the original computers—come on, give me some credit—but the computers whose reign still haunts us. Windows 3.0 arrived in 1990; the Mosaic web browser, Netscape’s precursor, in 1993; and Hotmail in 1996. I’m too tired to tell you the rest of the story, even though you’re probably too young or too old to fill in the blanks.

By 1999, the commercial internet had engorged into the dot-com bubble. The brick-and-mortar business world was going online—for information, for communication, for shopping, for utility billing. This moment had its own vocabulary: the information superhighway, the apostrophized ’net, and so on. E-business, we called it. (In a recent telephone call with a fellow old-timer, I used the word e-business tongue in cheek, and my interlocutor immediately dated my human origin to the early-to-mid-1970s.) The part of the internet that would persist got planted here, but we overharvested its fruits. Pets.com and Webvan, an early Instacart-type site, and so many more fell apart following the dot-com crash of 2000, ushering in a downturn that had turned further downward by 9/11 the following year.

People, trends, companies, culture—they live, and then they die. They come and go, and when they depart, it’s not by choice. Habituation breeds solace, but too much of that solace flips it into folly. The pillars of life became computational, and then their service providers—Facebook, Twitter, Gmail, iPhone—accrued so much wealth and power that they began to seem permanent, unstoppable, infrastructural, divine. But everything ends. Count on it.

We didn’t consider this much back then. We were still partying like it was 1999, because literally it was. Everyone had an Aeron chair and free bagels every morning. One such day, sitting in front of the big, heavy cathode-ray-tube monitor at which I developed websites that helped people do things in the world rather than helping them do things with websites, I could easily have read this press release, about a partnership between Bluelight.com, Kmart’s nascent e-commerce brand, and Yahoo, the biggest, baddest, coolest internet company of all time (at the time).

“‘This is an unprecedented offer,’” Bluelight.com CEO Mark H. Goldstein said in the release, “‘bringing together Kmart—one of the nation’s strongest retail brands—with Yahoo!, the leading Internet brand worldwide.’”

That line would have seemed corny at the time—it’s PR preening, after all—but now it reeks of digital mothballs. Imagine, as you read this today, a collab more impotent than Kmart, an inventor of big-box retail that failed spectacularly just as that format became widespread, and Yahoo, the internet company that failed to buy Google for $1 million (1998) and $3 billion (2002), but happily handed over $3.6 billion for Geocities and $1.1 billion for Tumblr, both of which it destroyed.

Kmart and Yahoo still persist, of course, in the way Werther’s Original butterscotch candies do: as withered husks that make it difficult to ponder their former heroism or tragedy. But these two organizations also mark two heydays separated by a quarter century, and two endings that occurred roughly that long ago. Those reigns and ruins marked my generation in profound, if (forgive me, DFW) banal ways. They recorded beginnings and also—more important—ends.

I started thinking about Kmart thanks to a tweet posted by the Super 70s Sports account: a handwritten sales check dated December 20, 1981, for the purchase of an Atari Video Computer System and three games (Casino, Asteroids, and Space Invaders). The Atari was already on my mind, because I still make games for the 1977 console; I’m teaching a class on Atari programming this term, and I keep turning away from this document to troubleshoot my students’ assembly code. Prior to this course, they had never played an Atari—nor had they seen a cathode-ray-tube television, either.

But Kmart was a place where you could buy either one—or anything else, it seemed. Lego kits, Tupperware sets, Wrangler jeans (in slim, regular, or husky fit), automobile tires, bedding, Preparation H. You could even sit down for a tuna melt on the Naugahyde seats of a full-service diner inside.

The idea of excess was hardly born with Kmart, but its rise tracks retail consumerism’s ascent in the American mid-century. Later known as big-box stores, they were then called general-merchandise (as opposed to specialty) stores, a novelty beyond the megalodonic Sears (which Kmart would absorb decades later). By 1981, when a lucky kid got an Atari for Christmas, more than 2,000 Kmart stores dotted the whole nation.

And so, for me and my generational kin, Kmart became a symbol of commercial surplus. The department store was antiquated by then (a shop for your parents), and the mall was mostly spatial (a place to be rather than to shop). But Kmart contained all possibilities, under one roof, a walk or a bike ride from home. Looking back, the first hit of spangle and sloth that would become native to internet life was doled out at Kmart.

Read: I wrote this on a 30-year-old computer

A decade and a little later, it became clear that consumerist gluttony could apply to information. It had been possible to take home computers online since the Carter administration, first through BBSes, then via dial-up services such as Prodigy and Compuserve. But the internet was quantitatively—and therefore qualitatively—different. A BBS was local—some of all the computer-nerd maladroits in your town—and a dial-up service was a walled garden, composed from whatever materials the service provider deigned worthy. But the internet was a network of networks, all of them fused into a traversable whole. That idea is so old, it too seems banal, but it once felt new.

The World Wide Web, as we still styled it in the early 1990s, offered the best-yet way to use the ’net. Email had been around for a long time, and newsgroups, and others that didn’t live long enough to earn a widely recognizable shorthand (one was Gopher, a text-based protocol that the web suffocated).

I struggle to explain the scarcity of information at the time. Nobody noticed what they couldn’t yet imagine otherwise, as is the case with all historical change. Desires and needs were shaped not by online recommendations but by retail shelves, at venues like Kmart. Knowledge was different: Faced with an information problem, where could one shop for solutions? The library, or the bookstore, or the museum, or some other archive perhaps, but only if you already knew enough about the information you sought to know where to look.

Nowadays, too much information is on offer, most of it bad or wrong, and we spend our time either sifting for gold in the filth or mistaking the filth for gold. But things were simpler back then. In 1994, I was able to pilot Gopher via telnet from my personal computer through the burrows of song-lyrics annals, a secret lair that’s since become, like everything else, ordinary. I extracted the lyrics for the Pearl Jam song “Yellow Ledbetter”—notoriously incomprehensible thanks to Eddie Vedder’s mumbling—and became a minor hero when I sent my find off to my friends, many not yet online. It was akin to becoming the kid on the block with an Atari, except I was flaunting my access to information rather than goods.

Yahoo was the first company that tried to organize information deliberately. It did so by means of directories—categorizing websites into groups by topic, such as movies and politics, which were broken down into subcategories. Jerry Yang and David Filo, two Stanford engineering students, created the site around the same time I mined Gopher for Pearl Jam lyrics. I have strong memories of the first time I loaded up Yahoo, still from Stanford’s server, on an expensive Sun workstation in the university computer lab.

There it was. I didn’t know what, exactly, but just as Kmart’s aisles created expectations and desires, so did Yahoo. “Art,” “Business,” “Events,” “Science”—the categories weren’t novel; they represented human life and its interests. But what people were doing with those interests on the web—this was a new idea. Circa-1994 Yahoo had as many entries for “Art > Erotica” as “Art > Architecture,” suggesting that the WWW was going to be a horny place more than a spatial one (yup). “Society” and “Culture” broke down into “People,” “Religion,” and related categories. No one asked then why computer engineers were categorizing human knowledge, though they should have.

The marriage of Kmart and Yahoo was short-lived. Bluelight.com, named after the retailer’s famous blue-siren in-store specials, was meant to host the store’s e-commerce business. Its customers might not even have had home internet access at the time, so Bluelight.com also provided free internet access as an incentive to shop by subsidizing a private-labeled offering from Spinway.com. Kmart handed out CD-ROMs that provided software to access the service, AOL-style; it also sold branded PCs that came preloaded with Bluelight.com internet. To lure its retail customers online, first it would have to get them online.

But Kmart mistook the internet as a way to shop for goods rather than a means to swim in information. That’s why Goldstein’s line about the strongest retail brand and the leading internet brand was justified when he said it almost 25 years ago. “Yahoo was cool!” Goldstein told me, and he’s right. Yahoo had tamed the information space, as Kmart—which was still the third-largest merchant in the country at the time—had done the retail space. It just made sense.

Bluelight.com had intended to launch in mid-2000, but by that time the market had crashed and the dot-com era had ended. Spinway.com went out of business, and Bluelight.com was forced to buy up some of its assets over the next year to keep its in-house ISP running. Hungry for revenue as the economy faltered, it started charging for the service, too. Goldstein left in mid-2001, and Kmart itself filed for bankruptcy several months later. Today, only three Kmart stores are left in the contiguous United States. Want to know how I know that? I read it on Yahoo, which has mostly devolved into a weird content website that sometimes shows up in Google searches.

Goldstein, now a venture capitalist who invests in health-care companies, also knew Kmart was everywhere. The ideas at work at the time were correct but early, and deployed under unfortunate circumstances, among them Kmart’s inevitable decline. For example, Bluelight.com championed a buy online, pick up in-store program it honestly called “sticky bricks”—a model that wouldn’t become ubiquitous for another two decades, during the coronavirus pandemic.

Today, the collapse of a big technology or retail company is almost unthinkable. Just look at the pearl-clutching over Twitter’s recent shambles: The public can’t fathom the idea that it might decline, let alone possibly die, for real. But the certainty of death, rather than the hubris of assumed eternity, was the salient cosmic feeling of the 1990s internet. Its creators had learned that sentiment from the Cold War, tapping out time on Atari games about the apocalypse while awaiting its real-world counterpart. Of course Kmart died, and Yahoo too. What else could have happened? “We’re all going to be absorbed; we’re all going to be consolidated,” Goldstein said. “At the end of the day, we just hope to end up as a button that survives.”