From algorithmic identity to community co-liberation
The second Kernel piece we’re pre-releasing is an essay by Matthew Sun on transcending the early internet’s naive post-racial dream. It’s equal parts touching and analytical, effortlessly crossing between personal narrative, critical theory, and culture critique.
Kernel Magazine costs $30 for the full print and digital edition — plus access to monthly events and audio recordings of posts like this one — until October 8th. All surplus revenue supports Reboot’s future work and a shift toward an organizational model that is created and funded by our community.
If you’ve already subscribed, get the audio version of this essay here.
By Matthew Sun (he/him)
In 1970, the economist Thomas Schelling produced a profound insight into racial segregation using nothing more than some colored pennies and a chessboard. With red and green pennies representing members of different racial groups, he placed each penny on a random spot on the board. He then created a simple rule to govern individual behavior: if at least 30% of a penny’s neighbors shared its color, it would be “happy” and remain in its spot; otherwise, the penny moved to a random empty square. Schelling applied this rule to the pennies one-by-one until each one was satisfied in its position.
As the simulation progressed, highly homogenous “neighborhoods” of same-colored pennies emerged and remained stable, despite the fact that each individual would have been satisfied with 70% of its neighbors being members of a different group. The experiment revealed a hard-to-swallow truth: mild same-group preferences at the individual level could result in stark, persistent divisions at the system level.
Of course, there’s plenty that Schelling’s theoretical exercise fails to consider. The fraction of Asian families in my childhood neighborhood was certainly not above 30%, and yet these families rarely moved out for a variety of reasons: the local job market, the location of well-funded public schools, the fantasy of an American suburban life. Schelling himself noted that the scope of his work was limited to the study of “individual action,” rather than “organized” or “economically induced” segregation. Since the publication of his study, scholars have debated the limits of what his experiment can tell us about the real world, often enriching his model with additional social context and mathematical complexity.
Revisited today, perhaps Schelling’s study is best interpreted not as an explanation for observed patterns of residential segregation, but as a cautionary tale for a new realm that would emerge in the following decades: the internet. Indeed, the appeal of the internet is often sold to us through stark contrast to the difficulty of achieving a sense of belonging offline: on the internet, we are free — and increasingly, algorithmically guided — to join meaningful communities and make new social connections, unbound by geography or capital or backwards institutions. Describing the process of physical relocation might require far more than colors and pennies, but moving from one digital community to another asks little more of us than following a new set of accounts, creating a new profile, shedding an old username. What could possibly be less frictionless?
I’ve always been a bit envious of people who have strong feelings about the neighborhoods where they grew up. To have either contempt or nostalgia for a place is to at least be connected to it, to situate your memories in a story of opposition against your environment or nurtured by it. Even ambivalence is too strong a descriptor: it implies the presence of contradictory feelings, opposite forces acting on one’s psychic center of gravity.
In contrast, I never formed much of an emotional attachment to the suburbs of my childhood at all. The streets where I grew up evoke familiarity, but not much else. Perhaps the reason I hesitate when people ask me how I feel about my hometown is that, as a minority, I never felt like it was mine to begin with. My family was a lone red penny in a sea of green — separated from the neighborhood, despite being within it.
Instead, the drama of childhood was strictly bounded by the walls of my family home. It’s inaccurate to chalk this up to the trope of the insular Oriental clan; if we retreated into our home, we retreated from racist encounters and microaggressions. Stories of passersby shouting slurs at us — and once, bruises from physical aggression — loom specter-like in my family’s memory. Outside the home, I was simultaneously uncomfortably visible and constantly overlooked, particularly in predominantly white activities (Boy Scouts, sports camps, swim club). At home, at least, the cognitive burden of managing how others perceived me disappeared.
But the dull ache of failing to be seen again and again exacts a psychic toll, no matter how much comfort one’s home provides. I wonder if that’s why I found so much comfort in fantasy worlds, where I would devour stories about children who had adventures in magical lands and stay up playing Pokemon on my Nintendo DS long after my parents fell asleep. But the characters I obsessed over were imaginary, and escape was always temporary. They never filled the void where a vibrant community beyond the family should have been.
Just as Chinatowns became sites of collective struggle, mutual aid, and social change, so too might we reappropriate the communities occupying digital platforms. Like the tenants who fought together for the I-Hotel, these communities would be strengthened through collective action rather than shared passive consumption.
When a penny in Schelling’s simulation decides to move, where does it move to? Intuitively, Schelling’s methodology of random assignment seems unrealistic. In general, families might experience greater costs the further that they move; consider the difference between moving to a house a few blocks away and moving to a different state across the country.
Unlike physical reality, however, where traveling between locations involves real costs, users in the digital world essentially teleport from location to location when they close a tab and open a new one. Indeed, since William Gibson, seen by many as the father of cyberpunk, first coined the term “cyberspace” in 1982, myriad internet manifestos have breathlessly extolled the emergence of a digital society that would transcend borders, liberating individuals from geography itself. In 1996, John Perry Barlow, a founder of the Electronic Frontier Foundation, wrote a document titled “A Declaration of the Independence of Cyberspace,” and within just nine months of its publication, an estimated 40,000 sites on the still-nascent web displayed a copy of the declaration. In it, he proclaimed the internet would be the “home of Mind” beyond the reach of governments, those “weary giants of flesh and steel.” Immediately after stating that “ours is a world that is... not where bodies live,” Barlow declares that “we are creating a world that all may enter without privilege or prejudice accorded by race, economic power, military force, or station of birth.”
Barlow never makes clear how exactly individual “minds” aggregate to form the collective “Mind” of his fantasy internet. As media studies scholar Lisa Nakamura notes in Cybertypes, the tension in this utopian vision is that the internet will usher in a new age of diversity while also simultaneously erasing all differences between individuals. It was not subjugation on the basis of difference that techno-optimists hoped the internet would eliminate, but difference itself. Idealists hoped that the internet would circumvent, rather than directly dismantle, existing hierarchies of domination, by allowing users to shed the visual signifiers of race, gender, and class. But as philosopher George Yancy writes in “Whiteness and the Return of the ‘Black Body’,” what is whiteness if not an all-consuming command into sameness, an implication that anything besides itself is to be pitied or conquered?
Setting aside whether the paradoxical erasure of difference in order to promote diversity was problematic in and of itself, Barlow’s vision of an internet without “prejudice accorded by race” was doomed to fail. As Schelling’s experiment makes clear, even in a setting where one attempts to disregard power dynamics or history, social categories are stubborn: when individuals are marked by their group status and act accordingly, social segregation arises. And in fact, if the costs of travel are essentially zero on the internet, we should expect to see the digital world become not Barlow’s post-racial utopia, but rather devolve into segregation shockingly quickly.
Sometime during elementary school, I discovered YouTube. The hints of sociality were tantalizing: watching the same video that millions of other people were watching felt like participating in a massive shared cultural experience. YouTubers spoke to me when they looked directly to the camera, and actively cultivated (parasocial) relationships with their fans. In retrospect, I wonder if I subconsciously gravitated towards the popular creators who looked like me: KevJumba, Ryan Higa, Natalie Tran, Freddie Wong, Wong Fu Productions, AJ Rafael, Sam Tsui. They became the neighbors I didn’t have growing up, successful Asians for whom increased visibility seemed to lead to money and fame, not the discomfort of being perceived as other.
YouTube seemed to transcend the whiteness of Hollywood. It was a medium where identity seemed irrelevant to your success. Except, as I realized years later, that wasn’t quite true: there seemed to be nothing more profitable than Asians making fun of themselves. Much of the humor was nuanced — perhaps even subversive — but in my elementary school days, the lesson I absorbed was that talking about race was okay, so long as it avoided seriousness at all costs. I picked up the language of racial humor, retelling jokes from YouTube on the playground to win laughs from kids who weren’t as online as I was. Looking back, it’s hard not to be embarrassed by the subtext of those interactions: You don’t have to feel anxious around me — I get the joke, I think Asians are weird, too!
Ryan Higa’s most infamous video is “How to Be Ninja,” uploaded on July 25, 2007. Rewatching the video, I can’t help but wince at Higa’s exaggerated East Asian accent or the instruction that ninjas should be able to act like a “fag.” The video has garnered 55 million views to date. Today, there are about 19 million Asian Americans in the United States. Had Higa’s viewers been gathering in person, I wonder if I would’ve realized sooner that the community of funny, popular Asians I imagined existing online was not so much a neighborhood as it was a circus, performing for an audience whose members almost certainly consisted mostly of non-Asians. Like the pennies in Schelling’s simulation, circuses are designed to hop from one community to the next, welcome only for as long as their audiences can sustain a laugh.
Segregation is not merely the result of individual decisions, devoid of history or power relations. Schelling himself concedes that individual decision-making might be less pertinent than de jure or economically induced segregation and, furthermore, that “the lines dividing the individually motivated, the collectively enforced, and economically induced segregation are not clear lines at all.” Take, for example, the role of real estate agents in perpetuating residential segregation today, more than fifty years after the passage of the Fair Housing Act. Real estate brokers regularly make assumptions about their clients’ racial preferences before steering them away or towards certain neighborhoods, arguing that it “facilitates the sales process.” Under market incentives, the responsibility for segregated outcomes is thus deflected onto individual consumers.
Like real estate agents who curate the possibilities their clients can access, today’s algorithms selectively recommend online content in the name of increased convenience for users and, of course, larger profits for tech companies. Increasingly, behind every webpage and application is a massive network of third party trackers, logging devices, and cookies designed to capture and mine user behavior, a process Shoshana Zuboff describes in The Age of Surveillance Capitalism. Making assumptions about what users prefer and how they will behave is fundamental, rather than incidental, to the profit engine of surveillance capitalism. The vision of an internet that offered unparalleled freedom to its users has given way to one that surveils, predicts, and steers them instead.
Digital platforms have the ability to intervene in users’ lives far more than real estate agents ever could: they can suggest content and nudge users towards particular communities, without users ever having expressed a desire to find something new. If real estate agents make assumptions about individual preferences, algorithms assert the power to reshape our preferences in the image of profit. The idea that these algorithms perform a kind of racial steering is not unheard of. Black creators on YouTube and TikTok have pointed out ways in which the algorithmic systems invisibilize their work by promoting white influencers following trends over the Black creators who invented them. So long as tech companies claim to be simply following market incentives — incentives in markets they themselves are responsible for creating — they must be held accountable for reproducing and intensifying racial fault lines.
The other day, I logged on to Twitter and read a few tweets by economists and activists debating optimal tax policy. After a year of quarantine-induced scrolling, I felt exhausted, but I reminded myself that it was important to know about these things, even if they felt so far away from being relevant to me. I kept scrolling. I read a few tweets about the rising number of AAPI hate crimes. I felt exhausted, but I reminded myself that it was important to know about these things, even if they hit too close to home. I closed Twitter, and lay motionless on my bed, thinking about what to do next.
After picking myself up off the bed, I walked downstairs to the kitchen, cracked open a few eggs, and stirred them into a bowl. I steamed them gently, attempting to follow along with both my memory and a video from a Cantonese cooking channel on YouTube. When I finished, I sent a picture to my mom to show her my attempt at steamed egg custard. It looks even better than mine, she replied, and suddenly a lump began to rise in my throat. As the tears fell, a few moments passed while I figured out why exactly I was crying. When it hit me, I texted her back with the simple explanation: It tastes like home.
Moving from one digital community to another asks little more of us than following a new set of accounts, creating a new profile, shedding an old username. What could possibly be less frictionless?
My feeds today look nothing like Barlow could have imagined, and more like what Schelling might have predicted. Instead of an egalitarian utopia of anonymous sameness, the digital world is hyper-fragmented, with a dizzying multiplicity of trends, memes, and tribes of varying exclusivity. Anyone who has felt isolated by the offline world and its crumbling institutions is usually just a few clicks away from discovering that they’re not as alone as they thought they once were; however niche we think our interests might be, there’s always a subreddit, or YouTube channel, or TikTok hashtag that somehow suits us perfectly. It is likely that as long as society continues to fail to meaningfully include people of color, so too will we navigate the internet to find alternative narratives that affirm us, so too will the tech capitalists surveil us as we do so to classify and target us for profit.
From real estate brokers to tech companies, our desire to find a sense of home has been hijacked by the agents of racial capitalism. As renowned political theorist Cedric Robinson wrote in Black Marxism, the “tendency of European civilization through capitalism was thus not to homogenize but to differentiate — to exaggerate regional, subcultural, and dialectical differences into ‘racial’ ones.” If the internet enlists us in an endless march towards gleaming, new homes, each more beautiful and individually tailored than the last, we should be skeptical. It is too easy, too frictionless to believe that we will all find our perfect communities in an algorithmically curated set of recommendations. I see these as communities of consumption, in which the soothing feeling of a shared identity is hijacked as a tool for maximizing engagement. To be clear, I am not opposed to the flowering of communities organized around identity on digital platforms; I myself have found comfort in the digital milieu of Asian American and other minority creators. But the tech industry’s constant repetition of platitudes about “connection,” “community,” and “engagement” risks forgetting that these words have historical meanings and political implications, too.
In 1968, the owners of the International Hotel near San Francisco’s Chinatown attempted to evict its elderly Filipino tenants. In what became a historic milestone in housing advocacy history, 150 elderly Filipino and Chinese tenants responded by beginning a nearly decade-long fight to protect their community, eventually joined by a coalition of thousands of people, spanning students, LGBTQ+ activists, and trade unions. If I can find the familiar taste of community in egg custard tarts in Chinatown, I can also find there a history of community that was not taken for granted, nor solely inward-looking. Instead, forming coalitions and finding solidarity with other groups was essential to preserving the existence of the community itself through a process of co-liberation.
I wonder about what it looks like to move beyond communities of consumption and towards communities of co-liberation in the digital realm. Just as urban ethnic enclaves like Chinatowns, carved out by enclosures of legal and economic segregation, also became sites of collective struggle, mutual aid, and social change, so too might we reappropriate the communities occupying digital platforms designed for profit to instead serve a vision of a more just future. Like the tenants who fought together for the I-Hotel, these communities would be strengthened through collective action rather than shared passive consumption. They would become lively spaces for dissent and discourse, as members weighed different strategies for achieving shared goals. And through participating and contributing to them, individuals might find not only a feeling of belonging, but also a newfound sense of agency.
I myself am searching for these communities of co-liberation, online and offline. But perhaps it is not enough simply to search, to take for granted that they already exist, waiting to be discovered by the perfect permutation of keywords entered into Google. Instead, to orient myself homeward, I must also orient myself towards an attitude of active construction, rather than simply passive consumption. The idealized communities we yearn for will not simply be found through freedom of movement on the internet and ever-improving algorithmic curation. Instead, our collective imagination must serve as the blueprint to build them.
This is a pre-release. The digital home for this piece, as well as for all other Kernel pieces, is at kernelmag.io.
Matthew Sun (he/him) is a second-year graduate student in Computer Science at Princeton University. He enjoys building and learning with kind people. Find more of him on his website.
What is your post-tech dream?
What would I be doing if the world weren't burning? I'd be cooking meals and baking desserts that would bring me together with friends and family.
Why did you write this piece?
Writing this essay helped me develop the mental and emotional equipment to situate my racial identity formation process within a larger societal narrative about technology, race, and community.
Thank you to everyone at the Reboot Studio who provided feedback and encouragement when I came up with the concept for this piece: Jasmine S., Hugh, Saffron, Ry, Theresa, Jess Z., Emily, and especially the members of my small group, Jasmine W., Chris, and Archana. Special thanks to Michelle, Emily, and Jessica D. for editing. My gratitude goes out to the entire Reboot community for being a source of motivation and stimulating conversation throughout the past year. Finally, shout-out to Dan and Kat for listening to me brainstorm ideas during our summer writing club.
💝 closing note
If you liked this essay, consider purchasing the magazine with a paid subscription to Reboot.
In addition to a print and digital copy of Kernel, annual subscribers will receive audio recordings of posts, invitations to monthly discussion seminars, and our immense gratitude from the nearly 30 volunteers who made Kernel happen. Read our editor’s note to learn more about the magazine’s mission.
Tomorrow, we’ll be publishing Riley Wong’s speculative fiction piece “hypnopompia,” which traces the journey of a dream researcher in surreal, glimmering prose.
Jasmine & Reboot team