Full-text audio version of this essay.

Every few months I accidentally catch a glimpse of my computer desktop (which I usually keep obscured with eight overlapping windows) and realize it’s time to deal with the damn screenshots. This time there are 168 of them, captured at a rate of about five per day, the unevenly sized png files piling up across the desktop because I never bothered to change the folder. What do people do with their old screenshots? Just bin them? Not me. I review each one in case there’s a snippet of essential documentation or a glimmer of genius captured in the pile of weird fragments.

Here we go: A small bit of clipped text that reads “creative class chore-coats” in a serif font (keep). An awkwardly cropped photo of Daniel Day Lewis riding the subway, wearing head-to-toe Carhartt and texting on a flip phone (keep). A picture of a Skype chat log where I or the other person did a funny typo (toss). A square of saturated red with a hex code underneath (toss). An image of a bad tweet from an annoying venture capitalist (keep, unfortunately). Most of the files go in the trash. A few of them go into a folder on the desktop labelled “Screencaps to keep.”

The making of screenshots still feels like a slightly piratic and makeshift activity

This review process is similar to that of looking through photos on my phone — of which there are also a million, good and bad and ephemeral — but these days I have less and less reason to look through my phone’s photo reel. Why bother, when Apple assesses each photo automatically, analyzing the images and categorizing them so I can call up a date or a place, or choose to look only at selfies or photos of food. The phone and the iCloud account it’s connected to can hold great volumes, and the logic of “keep ’em all, and let god (the algorithm) sort it out” has superseded the sorting and assessing of years past.

Compared with other kinds of digital output, a screenshot feels “homemade” — a piece of digital DIY confirming that screen spaces contain as much mundaneness as the “analog” or “offline” world. In a time when so much communication and production is mediated by a set of increasingly powerful corporate platforms and absorbed into the internet’s big cloud files (aka data centers and fiber optic cables), the making of screenshots still feels like a slightly piratic and makeshift activity, not yet subsumed into the seamless logic of other kinds of digital image making and distribution. The screen consumes so much of me — time, labor, attention. I screenshot to lay claim to the act of seeing, which remains mine alone.


Photography’s function, and its mode of production, have changed dramatically in the past quarter century. The rise of the consumer digital camera and then the phone camera, the ease of distribution offered by the internet, and the ever-expanding archival capacity afforded by cloud storage services have all shaped how we make and think about images. In 2008 (pre Instagram, pre non-rubbish phone cameras), artist Joachim Schmid was struck by the preponderance of food photos being uploaded to Flickr, the online photo sharing site founded in 2004. People were photographing and uploading the prosaic moments of their lives with what seemed to him like a prodigious fervor, so much so that “the exponentially growing quantity of images has turned into the foremost quality of photography itself.”

Of course, the number of images produced and shared online has increased exponentially since then. Most of these photographs are not intended as art objects or as formal documentation. They’re pieces of casual communication tossed into the stream of group chats and Discord shit-talk and Instagram’s hungry maw. This is the “social photograph,” as theorist and writer Nathan Jurgenson describes it — “everyday images taken to be shared.” As a cultural practice, the social photo is a close descendent of what Schmid described a decade earlier: “its existence as a stand-alone media object is subordinate to its existence as a unit of communication,” says Jurgenson. The impulse of the social photographer is to frame and capture a moment from the flow of life and to share it out, in real time, with an imagined community. Or, if not to share immediately, then to capture it as shareable, a pre-social act. A croissant poised insouciant on a plate, a selfie in the vintage store’s fitting-room mirror, an underwhelming sunset. Daily life becomes photographable, and photography becomes a practice of everyday life: a moment, a breath, a social event, a marking of time. To photograph is to digest the world.

My life today is a life lived near screens. For 8-12 hours a day, my visual field is dominated by a large screen (the bad screen) and punctuated by long, glazed minutes looking at a smaller screen (…also the bad screen, if I’m being honest).

If smartphone photos are a catalogue of moments, images, nota benes, and visual questions captured as I move through the world, then so are screenshots. “Screenshots are the snapshots of our computers,” writes digital media theorist Jacob Gaboury in his essay series Screens Shot. “They capture the movement and forms of everyday life as it is lived through the interface of a computational device.” Gaboury uses the screenshot as a way to chart the changing ways that computational processes have been visually represented, the methods used to documented these changing images, and the cultures that coalesced around them. My screenshots are still a document of computation, but often only incidentally, the way a snapshot of a nice tree is also a document of forest management. They are an archive of a life lived in the vicinity of screens.

This screen-centric visual experience has a lot going on: photographs and videos and vector graphics and overlapping chat windows pressed up together with spreadsheets and reminders that updates are ready to install if only I’d let them. It’s the everything-screen! And so I feel called to frame and claim and digest it all. I have limited control over the on-screen textures, architectures, and environments I spend time with, but still, my experience of screen-life is my own. No one else has the exact same dead pixel in the upper left corner coupled with my specific desktop file arrangement, and perhaps no one else had the same news story juxtaposed beside an ironically hilarious ad. As with casual smartphone photography of non-screen scenes, screenshotting is a small assertion of control and customization, asserting my sensibility as a reminder that I was here, in front of this screen, in this singular moment.

Like the snapshot, the screenshot is a technique I use to process and frame the world, to keep or to share with others. The act of capture is a simultaneously archival and communicative act. I screenshot an image from a blog roundup of “comfiest face masks” for posterity, with the hopeful sense that this image will feel historic and defamiliarized in the near future. My sister screenshots goofy stock photos of robots typing on computer keyboards to send to me. I screenshot a meme to post on Instagram, because it takes two fewer clicks than actually downloading the jpg. I know my experience of screens is different from yours, and that we notice and enjoy different things. And so, I frame, capture, and share.


Around the time Joachim Schmid was noting the proliferation of food pics on Flickr, I was worrying about Tumblr. Tumblr, the micro-blogging site launched in 2007, was (and is) defined by its “reblog” capability. That is, any image or gif or audio file that a user posted to their Tumblr page could be seamlessly reposted onto other users’ pages. Harry Potter fan art, William Eggleston photos and hentai gifs alike were all equally shareable, equally appropriable. However, to reblog an image wasn’t just a repetition but an excision that ripped it out of its genealogical lineage, removing it from the context where it was first posted and leaving behind any text that may have accompanied it. For artists that posted their work, reblogging meant an image could quickly become detached from its attribution and used by others with no regard to its creator or uploader.

My screenshots are an archive of a life lived in the vicinity of screens. They are a small assertion of my sensibility as a reminder that I was here, in front of this screen, in this singular moment

Back then, it still felt like online images came from somewhere — somewhere offline. By 2008 I was no longer a web naif, but I still assumed that an image of a painting (be it Buffy fan art or a Renaissance portrait) was only a couple of steps removed from its offscreen referent, whether the Met’s newly digitized collection, or Spike_lover44’s latest and greatest, uploaded to DeviantArt. The endless reblogs of the Tumblr universe felt decentering, Baudrillardian, a series of shifting rootless images unmoored from “a referential being or a substance” as he wrote in Simulacra and Simulations, which seemed like an appropriate text for the dizzying moment. Over the next couple of years, Tumblr wrangled internally with the reblog attribution question, and in 2010 moved to a system that shifted attribution information to a series of metadata fields in the CMS, aiming to ensure “the entire reblog lineage” rode along with images as they moved through the web. Still, attribution continued to get lost, lineages got severed, and users didn’t always bother to create good data when they uploaded content.

What has changed? For one, any expectation I had that online spaces might act as a digital corollary to offline spaces has entirely dissolved. The online/offline distinction was always fatuous, but for a time the metaphors were necessary to frame our understanding of an evolving internet. Wikipedia as an encyclopedia. Email as a digital letter. The Oh No They Didn’t Livejournal as a gossip mag. Each in their separate sphere, with seemingly little intertextuality. Over time, especially with the transition to search-based navigation and Google Images, the “reblogging” and regurgitations of photos and other web content became an integral part of the online mundane. And so, while I was lamenting the loss of authorship and the ease of misappropriation in 2008, in 2021 I grab whatever catches my eye, screenshotting film stills from streaming sites and Dutch golden age paintings from museum websites and many, many memes. The act of screenshotting an image (rather than downloading it) strips out the metadata from the original photograph, and I don’t necessarily attribute the artist if I share. They’re certainly not “traceable” images. And I don’t feel guilty about it (although writing this out, perhaps I should).

In 2008, put-a-bird-on-it radical crafters worried that other, less scrupulous crafters would steal ideas or pass off ideas as their own. Photographers worried their photos would be reblogged without attribution by others who would be lauded for their good taste, with no clout reflected back to the photographer. These scenarios are still widespread. But the things that perturbed me about 2008 Tumblr — endless circulation, appropriation, customization — have become the logic of the internet. Multiple ownership claims pile up on a single image. The cozy ’70s serif titles on direct to consumer skincare websites blur and meld. Squarespace templates proliferate like minimalist fractals. Today, the parties doing the most to accumulate images and use them for personal gain aren’t some kids on Tumblr. They’re the platforms, the big dogs that run the place, like Facebook, Apple, and, especially, Google.


My phone is in cahoots with FAANG and friends. What was once a discrete object is now at one with the platforms, and as long as it’s receiving a wifi or data signal, nothing I feed into it stays private for long. Increasingly, anything I put into my phone has the sense of being always-already shared. My notes app full of shopping lists and bad midnight ideas is fully connected to “my” iCloud, so I can access the useless scribblings from any device. My snapshots don’t wait until I upload them to Instagram or the family chat. As soon as the fake shutter opens, iPhone is busy at work, synching with the iCloud as well as analyzing, sorting and learning from the faces, food and locations in the images I make.

What has changed since early Tumblr? For one, any expectation I had that online spaces might act as a digital corollary to offline spaces has entirely dissolved

The act of sharing — the “social” part of the social photo or the social screenshot — isn’t so innocent, either. I’m just trying to participate in a simple communicative act, trade a little gossip between friends and share a couple of snarky screenshots, but there’s so many other people getting in the way. Or rather, so many intermediaries. As with the “sharing economy” — Airbnb, etc. — when you share something with a friend via digital platform, the middlemen are getting a share too. In the case of most sharing economy platforms, it’s a 10–25 percent cut of the transaction and the ability to define the parameters of the relationship. In the case of a photo I share online, the platform scrapes what data it can. Google and other intermediaries digest my image into the computational churn and use it to learn, understand, and optimize. The goal of these intermediaries (sorry, “service providers”) is to wriggle themselves into the middle of just about any interaction so they can extract rent, keep users trapped in a walled garden of privatized access, and siphon up valuable data. Sure, Whatsapp (Facebook) or Gmail (Google) enable the interaction. But they get more out of it than we can ever fully agree to.

In her experimental visual essay Memory Metadata, Livia Foldes describes what happens after a photo is taken. “My iPhone’s photo app has trained itself to recognize the people I photograph most often,” she writes. “According to Apple, it does this by executing one trillion operations. Many of these operations involve facial detection and analysis. These algorithms are a way for our machines to look at — and, more importantly, categorize — us.” Apple uses this photographic data for a number of purposes, finding ways to add convenience and optimize our experience of using the phone, while building powerful data sets and advancing its own computer vision capability. Our faces, and the large-scale data sets into which they get aggregated, are, at times, given and sold to other companies and used to market to us, train facial-recognition technology, and surveil us after protests and political actions.

I’ve come to think of the networks and infrastructures connected to the phone as active parties in the photographic process. I take pictures of the non-computational world and I give them to the phone so it can understand my world better. The phone’s understanding is extractive, not empathic, but it’s the tradeoff I accept in order to store and share my endless photo reel. I process the world through image-making, and in doing so, I transform the world into something the computer can process and use. I share the image to communicate, and in doing so it becomes something that tech companies can accumulate and commodify.

When I take a screenshot, it feels like a tiny rejection of the logic of the contemporary corporate internet. Instead of offering up fragments of my photographic life to the computer gods, the screenshot feels like I’m stealing something back from the computational world for my own uses, removing it from the networked flow (sure, some of these snippets are eventually sucked into my iCloud and otherwise absorbed back into the digital abyss, but I don’t mind too much, as it feels a bit like making the computer eat its own vomit). The screenshot is a gesture that lays claim to the act of seeing, and turns the framing and capturing of an image that often contains a mixture of windows and picture fields (a web page, a Microsoft Word window, all of it) into an act of creating anew. The old metadata: Gone! In its place, my image, with fresh data and the pleasingly prosaic file name “Screen Shot 2020-12-01 at 9.54.48 PM.”

While the ability to screenshot is an affordance offered by the computer operating system, often times it feels like a semi-illicit act. When faced with a complex process of export and rendering, or a proprietary piece of software that won’t let you download without paying a subscription, the screenshot yoinks the image from the plane of the screen and says “this will do.” Sometimes I don’t want to download the jpg, save the metadata, or adjust a bunch of settings so the software can export a high resolution pdf for me. And why bother, when the image in front of me is already what I want, and has more than enough fidelity for my purposes. Even bad images have something to say. So many of the memes that float around message boards and social media feeds are a complete mess, edges fuzzy and pixels popping out all over the place. But the poor quality becomes part of the point: It’s a marker of virality, a signal that the image has been shared and stolen by multiple viewer-artists who labor across networks to produce this parade of shitty jpgs, captured and reshaped and blown up and shared and recaptured, unstoppably.


Perversely, the chaotic and anti-attributional act of screenshotting has become a way to protect myself from the more unreliable and adversarial parts of screen-bound life. With so much of our socializing, organizing, and life administration routed through screens and networks, we face the contradictory risks of things disappearing or things staying findable forever. Nothing on our computers stays the same for very long. Software updates, websites disappear, and newspapers edit their copy and hope to get away with it. Terminally online Twitter users rush to screenshot controversial tweets so they can be revealed as proof of transgression in case they’re later deleted. More prosaically, I screenshot the page that confirms the IRS has received my tax return, because I don’t trust the IRS and I don’t trust their crummy website. The screenshot fixes a moment in time, acting as a form of vernacular truth, as receipts. “When words are not to be believed, screenshots may be offered up as proof,” writes Jacob Gaboury.

When I take a screenshot, it feels like a tiny rejection of the logic of the contemporary corporate internet

Screenshots are also proof of memory — proof that I was there, online at a moment in time. Some people photograph the same cityscape over and over, a witnessing of the slow evolution of skylines and property development as the material space is altered by powerful forces and slow decay. Digital spaces change more quickly, and leave less evidence behind, so I make screenshots to capture the moment. A screenshot of my Facebook feed from 2008 is as powerful as any punctum Barthes could describe. The shape of the icons and the very slightly different shade of blue give me a visceral shudder, and remind me of things that Facebook would prefer we forgot.

The inverse to the screenshot’s role as a memory tool is its powerful ability to omit, to capture a photograph while stripping out the original metadata. While there are robust frameworks and methodologies for documenting human rights abuses in a way that ensures traceable metadata and visible action, last summer’s protests against police violence following the police murder of George Floyd highlighted the risk of sharing images of protest actions online at all. Protesters were tracked, identified, and, in some cases arrested using evidence from uploaded, unredacted photos and video. In response, activists and indie media organizations urged people — if they must share images at all — to obscure protesters’ faces and share a screenshot of the original image rather than the image itself. This scrubs the metadata, making the image less traceable and less likely to identify and incriminate the photographer. Screenshotting produces a protest image that is visually legible as a document of revolt and action, while making it less legible and less useful to police and the state.

The conflicting imperatives of the screenshot-as-archive and the screenshot-as-anonymizer perhaps aren’t in conflict after all. In both cases, the screenshot serves the needs of the image maker and the computer user as they work to enlist the computer’s tools for their own purposes, eluding the surveillance eye and walled gardens of corporations and the state. It’s a small gesture, of course, and I risk exaggerating the ability of screenshots to do anything other than help us cope in a tiny way with the frustrations and complexity of computer life. Still, I’ll take what I can.

“Every film is a documentary” wrote André Bazin, film critic and founder of the Cahiers du Cinéma. Actually, it turns out I’d misremembered the passage — Bazin used the phrase “a social documentary,” but Godard and then Rivette honed the sentiment into the aphorism I remember (or at least, versions of the quote are attributed to both of them). Regardless, they’re right. Every film is a documentary, in the most literal sense. Films capture and document people, objects, and shards of light as they existed in the moment of filming. No matter how thorough the auteur, or how outlandish the plot, the material reality of the present seeps through. The way a background extra has curled her hair. Anna Karina’s gait: These are historic artifacts. The film is a documentary of how that year’s film stock reacted to light, the specific timbre produced by the microphones, and the cutting rhythms favored by the editor.

Looking through my old screenshots, I think the same thing. Every screenshot is a documentary. A documentary about what I found funny or shareable in the moment, a record about what the internet or a computer application looked like at the second of capture, and a document of file type, metadata, and naming conventions. My screenshots remember that Google Maps renamed San Francisco’s South of Market neighborhood to “the East Cut” after the tech companies moved in. They remember what the Firefox browser window looked like in 2014, before the search bar and address bar were folded into one. Against the logic of the always-already-online photograph and the ever-more-enclosed internet, the pile of unsorted screenshots on my desktop make up a furtive little archive of my own.