Lost to the Sands of Time

Ephemerality in the Digital Age


a shapeless mass; a network of times, by Berny Tan

On a Monday afternoon this spring, a movie that was never supposed to be shown to an audience screened for the first time in a basement theater at the Museum of Modern Art in New York. Shot in 2013, it took seven years for the original version of Peter Bogdanovich’s last film to surface, unexpectedly emerging thanks to an avid buyer of abandoned storage units and a scholar who evaded grading papers by browsing eBay. The likely only copy of Squirrels to the Nuts—a rom-com starring Owen Wilson, Imogen Poots, and Kathryn Hahn that the studio requested be reshot, reedited, and renamed—would have been left to rot as an industrial cassette tape. (The studio’s cut was released, to mixed reviews, as She’s Funny That Way.) Its conversion to a Blu-ray disc and digital file allowed me and everyone else in that theater to feel a treasure hunter’s giddiness secondhand. It was a fine watch. More than anything it made me wonder what else there is that we have lost, what else needs saving.

Leaving the museum, I rode the escalator that looks out at the sculpture garden once described by a parody-philistine character in John Cassavetes’s Shadows as “a place for a bunch of sexless women who don’t have any love in their life, a lot of big-deal professors, a lot of creeps trying to show off how much they know.” I was a perfect target, I guess, for the man a few steps below me to ask, “What’d you think? Of the film?”

I liked it. What about him? (Classic deflection.) He felt it was all a bit much, a shade too improbable. Then he said he’d known Bogdanovich personally; my interest was piqued. We stood on the corner of Fifty-Third Street and Fifth Avenue for almost an hour, and in the cold he told me as much of his life story as he could. He was an NYU film student in the 1960s, a Vietnam war vet, a puppeteer, a friend of Albert Maysles, and the subject of a retrospective at MoMA in 2006. His name is Joseph “Joe” Jacoby. Virtually no one knows who he is.


Jacoby stuck on my brain like an email left unanswered because I can’t find the words. He still identifies as a filmmaker, and introduced himself as one. But what he’s made has faded into obscurity, if not oblivion. When he described his work to me, I knew it wouldn’t be my thing (One synopsis from Letterboxd of a film with eight logs on the app: “A pretty young girl arrives in New York City to make it in ‘the Big Apple,’ but winds up getting involved with lesbians, an escort service and the underground bondage scene.”) Yet our contact gave me a sense of investment in him. Hearing him describe his work was like listening to a radio with a human DJ instead of a playlist generated with an algorithm. I was reminded of people making things with their hands and minds, and the luxury of how easy it is to access the product of their labor on my laptop screen.

Jacoby did not become my Moby Dick. Almost nothing is interesting enough on its own, the difference between a single-minded obsession and an expansive passion for a theme. Few artists are remembered past their prime. But all the bits of cultural minutiae taken together create an era, an aesthetic, a patchwork of what people thought and felt. I think of what television critic Emily Nussbaum said of “canceled” artists and the prospect of never again mentioning work made by bad men: “It would be nuts for me to behave as if I had never seen Bananas, Manhattan, The Purple Rose of Cairo . . . never to refer to those movies, not to place the world in their context, or myself in their context . . . you can’t scrub history off your skin.” Removing canceled content from the public sphere demands deliberate amnesia. But that an act of deliberation occurred at all cements the content in public consciousness and, often, leads to reactionary protection measures, as downloads of the four 30 Rock episodes pulled from syndication and streaming for instances of blackface circulate on Reddit and bad 2000s songs with off-color lyrics play loudly at house parties.

What is gradually forgotten rarely has people rooting for it. Libraries throw away books that don’t get borrowed and nobody notices. Online, site owners neglect to pay their domain name fees and writers lose access to their own essays. It would be nuts for us to lose art that was made by now-unknowns because we’re running out of space in the cloud. But who is “us,” who is invested in the cause when a select few companies own the sites that so much contemporary art is on? How much is already irretrievable? The politics of preservation aren’t new. Archiving as a conscious process has been around since antiquity, and has always required resources doled out by the powerful.

Film provides an early example of the modern storage problem. A statistic often cited by one of Jacoby’s former classmates, Martin Scorsese, is that half of American films made before 1950 are lost forever. Film stock’s acetate bases dissolve, indie film houses’ storage units get humid, soundtracks get separated. In 1993 a Library of Congress staffer appealed for federal funding with the plea that “motion pictures of all types are deteriorating faster than archives can preserve them . . . the United States is fighting a losing battle to save its film heritage.” Still, preserved non-digital materials have proved in practice to be more resilient than digital sources. Historians can read hieroglyphs worn away by centuries of sand, but a strike of lightning can corrupt a file.


Many people have the false impression that things on the Internet last forever, traumatized by warnings not to post drunk photos for future employers to find. The reality is that it will probably all go away. The Voyager Golden Record will survive longer floating through interstellar space. In 2019 the journalist Sarah Ditum put it succinctly: We are headed for a “digital dark age.” We create an incredible amount but what will there be for the next generations to see? Every day we take photos, maintain constant correspondences, and build an impossibly long and complex data trail just by scrolling around, looking at memes, participating.

Jason Scott, likely the most visible figure of the digital archiving scene, is a deep believer that “the digital record is very reliable until it isn’t.” He said, “People put their lives online and then one day wake up and realize it’s not there anymore. They are keeping their memories on spinning magnetic pieces of metal.” I like all the English phrases we have for the experience of being in a world that was written in another language: cookies, clouds, bugs. At the gaping mouth of the Internet rabbit hole, everyone is Alice. Though some of us have better night vision down there: Scott is known for his successful missions excavating lost content fuzzily recalled by strangers.

The Internet Archive staffer was called upon recently to help identify a mysterious cartoon character that appears on a TV set in an old family photo. The elfin figure was familiar enough to captivate tens of thousands who felt they’d seen him somewhere before, yet vague enough for his identity to evade discovery for years. When the saga finally concluded, it became a lesson in why people like Scott do what they do at all. The Canadian Christmas special the character is from means nothing to most of us. But to a few people, it was formative. It has to be. One of the laws of culture is that someone has cried, fucked, or died to every song ever written. Those B- and C-list teen movies I rented at the video store are part of what made me me.

Scott’s project TEXTFILES collects the content of dial-up bulletin board systems (BBSes) from their mid-1980s “golden age”: 8-bit GIFs of scanned Playboy centerfolds, entire sci-fi books written by amateurs, a bibliography of known Trojan horses, pages and pages and pages of writing about the Internet that range from the prophetic—“We are the cyber-beatniks…the DANGEROUS NEW ARTISTS….. the TECHNICIANS OF ECSTASY and we are all ENMESHED IN THE NET”—to the practical—a dictionary of possible keyboard smiley faces. He started the archive because he couldn’t stand the thought of a botched retelling of his own history. There’s evidence of what it was really like, and he believes in the right to examine it.

Today, a piece of media can be accessed by millions around the globe in seconds, the promise of the dot-com bubble made good. The original textfiles’ readership was limited to those nerdy enough to tinker with emergent technology; plus, servers skewed hyperlocal due to long-distance phone fees. One particularly touching piece of TEXTFILES is a history section where people who experienced the era can write about it. At 4 a.m. in January 2003, Steve Reeves submitted his account of BBSing in suburban Philadelphia in the late 1980s, a Linklater-esque scene of two best friends and next-door neighbors downloading pirate games onto floppy discs and memorizing recipes for homemade LSD that Reeves concedes he never tried but doesn’t think would have worked favorably. Then he grew up and so did the culture, quickly. By the mid 1990s, small-scale systems were traded in for central platforms. Sysops disconnected the phone lines, and Reeves’s mother no longer worried about her son burning down the house by keeping their “IBM clone” computer running all night.

“From 1993 until probably about 2001 I really didn’t give a crap about BBSes and didn’t care that they’d died. But now I find myself getting nostalgic for it. Now I wish I’d held onto some of the stuff from then,” he wrote. “I’m now an internet junkie and am wondering if soon something better than the Internet might come along? Everything has a life cycle, you know, even the mighty Internet.”

I tried to find Reeves, to ask if he ever found a secret, better thing. I think I wanted confirmation that the Internet hadn’t outlived him, or that he hadn’t outlived his optimistic impression of it. I did my best but was overwhelmed by the specter of a 1950s bodybuilder who shared his name and effectively gave up when I closed the window displaying so many variations on the same chiseled chest. I am Alice and I am lost because I don’t know where I want to go.


A web page is an art form, but it is different from a book or movie or album or game. I stay for only as long as a passing whimsy and then I leap away. I am not forced to finish anything; I return to endlessness again. I have forty-six tabs open and they stay there even when it all comes crashing down. Then I get into a mood and slam the window shut, and I don’t remember what it once contained.

As senior citizen creatives seem wont to do, Jacoby lamented the Internet again and again on that street corner. It was the reason for his cultural decay, in essence. I sympathize with him; he has watched the entire playbook of where and how art is made change in the span of his working years. And then realize I empathize. That’s going to happen to me too.

A web page is an unstable work. It is durational art, collective art, immersive art. My first pieces were my nights spent on my (shitty Dell) laptop and the record of my contributions to the collage of uploads, downloads, and reactions. I can’t retrace it myself. Most of it was on Tumblr, which is riddled with holes since its notorious porn ban in 2018, which censored even fairly innocuous content. My defunct, half-gone blog is a buildup of references that constructed my taste.

On the last day of 2020, Adobe Flash Player stopped being supported. It was recommended that we uninstall the tool for security reasons, and we were told that it would barely affect our browsing experience—what uses Flash Player these days anyway? The insinuation was that Flash Player is a breeding ground for computer viruses at worst, and at best nostalgia bait for people now in their twenties who grew up playing mini games on Newgrounds. Cultural change leaves everyone behind, sure, but this graveyard of dead links is uniquely frustrating. Flash animations cannot be viewed without a Flash player; such is the nature of native content formats. If one day we woke up and all the VCRs in the world disappeared, how many movies—homemade or otherwise—would be unwatchable? Similar to the appeal for salvaging American film history, the archivists rescuing tens of thousands of Flash games—using an emulator and automated web crawlers—believe in these animators and their work as crucial to the history of gaming, of youth culture, of technology.

The preservation of aughts content feels uniquely precarious because it exists in the uncomfortable space between physical archiving practices—which, however imperfectly conducted, are at least understood as something that should be attempted—and “the cloud.” The idea that our files could be stored in a way that made them accessible to us anytime and anywhere, with no need to rely on our personal devices to feel secure in the eternality of a recording of Baby’s First Step, was quickly adopted. The prospect is too good to question. Only the data isn’t actually suspended in the sky; it’s running along the Atlantic Ocean floor. Or it’s in a warehouse in the Badlands. It’s expansive and “everywhere,” the opposite of BBSes, while still adhering to the laws of physics. Even as I’m writing about this on a word processing platform that uses cloud computing, I can’t totally conceive of it—the ultimate sign/signifier/signified conundrum.

As a bizarre April Fool’s Day joke in 2017, Google made a blog post announcing their cloud infrastructure’s expansion to Mars: “Even if Earth experiences another asteroid strike like the one that wiped out the dinosaurs, your cat videos, selfies and other data will still be safe. Of course, we’ll also store all public domain scientific data, history and arts free of charge so that the next global catastrophe doesn’t send humanity back into the dark ages.” Are they mocking the fear of obsolescence because they think it’s an inevitability, or are they so confident that they have found the solution? On the blog is a link to “visit Mars with Google Street View.” Click anywhere on the image capture of the Red Planet and you’re dropped into the hallway of a corporate office complete with foosball tables and orange bean bag chairs. It certainly feels mocking.

Big Tech has sold us on the idea that the physical world is passé, that the experiences to be had in the metaverse will supersede whatever there is left to do out there. That’s an opinion; what’s fact is that the digital world can only exist because of physical infrastructure. Ninety-nine percent of international data is transmitted by fiber-optic submarine cables and the biggest content providers own or lease more than half of them. The closest thing to a total Internet shutdown, which has never happened, would be if Amazon turned off their cloud computing services. Nearly a third of the Internet would stop functioning properly. My oh so noble New Year’s resolution to never buy anything from the monolithic company suddenly feels utterly vain. Suddenly my USB flash drive that doesn’t even plug into my current computer feels safer than my undersea files smaller than a speck of sand.


The culture of tech and media consumption is moving so fast there isn’t time to question how it operates. The inventor of the World Wide Web is only sixty-seven years old, a Unitarian Universalist in England who sold a “signed” copy of the original source code as a non-fungible token, or NFT, last year. (An NFT is a digital identifier that represents ownership of an actual thing but is not the thing itself.) He made $5.4 million off the protocol he famously gave to the world for free, winnings that will go toward a data decentralization project to allow people to own their information and license it out to companies. (One proposed use case is an app that suggests optimized work commutes.)

“Decentralization” is the magic word for the new Internet phase we are entering. Those who grew up on the Internet tend to vaguely challenge the nature of its governance and the politics of ownership, but may not have conceived of any other past, present, or future. Web3 is some kind of answer. Website content stored on the blockchain—a database maintained communally by a network of contributors, as opposed to one privately owned server—is theoretically incapable of disappearing because even if the site goes dark, the raw data remains accessible to all the computers in the network. That’s strictly in theory. In practice, at least in this admittedly nascent stage, Web3 is very entangled with Web2, the current iteration of the Internet.

My foray into the new landscape has led me to conference calls hosted on Google Meet and debates on Twitter Spaces. Same old, same old. I am open-minded, attracted to the idea of agency for creators. I am just skeptical that this is what’s really going on. Digital life wasn’t created by any one person, and now we’re scampering inside it trying to stake out corners. Knowing how easy it is to amputate the cultural past, whether an accidental slip of the knife or a purposeful severing, I don’t feel secure in making a lasting impression.


The art that people make matters—all art, and all people. Yet the only constant in creation is ephemerality. Everything will fade away, whatever hasn’t already. It often comes down to chance what gets pulled out of the darkness, which artists get to feel what Bogdanovich did when he realized his director’s cut had been discovered: “It is like a lost child being found.” It could have been a Jacoby film shown in that basement. Or could it, realistically? He was never famous enough for a scholar to have set up news alerts for his name. He gave me his business card when we parted, me to the M stop across the street and him to his apartment at the bottom of Central Park. I got on the train, and searched my tote bag and all my pockets. The card was already gone.

Greta Rainbow