COVID-19 has been a crisis represented and interpreted through models. Models are metaphors that illustrate one phenomenon in and through another that is better understood or seemingly more transparent. In this article, we consider digitally driven COVID-19 models that draw on the certainty of data from smartphones and social networks to make predictions about a poorly understood virus. Network data normally used to model information spread drive models of an actually existing biological virus. A return to HIV network models of the 1980s helps map the social implications of this latest turn to modeling. These earlier models were used to hone stigmatizing viral metaphors about behavior, risk, and exposure, in the shadow of an emerging digital culture. Thinking across COVID-19 and HIV modeling demonstrates how models can support personal responsibilization, be used to blame “bad” actors, and justify the creep of new surveillance practices under the rubric of “Data for Good” programs. Drawing on critical HIV and queer studies, we argue that the people and behaviors that are opaque to viral models and their methods of capture present potential avenues for speaking back to digital virality’s terms. We highlight these exceptions, which show how certain lives make trouble for models and their sensibilities, telling of queer forms of life, desire, and contact that evade modeling altogether.

“Covid-19 sent you a friend request,” reads the meme’s text (Figure 1).¹ Next to it is the requester’s profile photo: a SARS-CoV-2 viron. The viron, a particle only one hundred nanometers in diameter, is far too small to actually be seen without the aid of an electron microscope. Yet the viron is instantly recognizable after its ubiquitous rendering by public health authorities. Playing on the content and aesthetics of the familiar Facebook “friend request,” the meme shows that COVID-19 and the user already have one mutual friend. It asks us to take action by confirming or denying the addition of the virus to our social network. This meme takes the form of a model-metaphor, presenting in playful vernacular how the spread of a biological virus is like the spread of connections and information over a digital network. This viron-Facebook meme, and millions like it, are generated in a strategic calculation to obtain “viral” currency in the shadow of an actually existing virus. The production and circulation of memes on social media, messaging apps, and other websites makes dark commentary on the absurdity of the virus metaphor amidst this latest global pandemic.

The use of viral digital metaphors to communicate about COVID-19 has gone beyond irreverent memes about social media. Network science and epidemiology explore how connections between people in digital networks might replicate the physical proximities that spread SARS-CoV-2. Similarly, data from network-enabled smartphones drive some contact-tracing protocols, and are used to model the potential of various lockdown measures by inferring how people move and interact through the proxy of their cell phones. The COVID friend request meme is funny in part because it leans into the power of the “viral” metaphor in digital culture: how we are in relation to each other online might model how we mitigate the spread of an actual virus.

The viral metaphor’s incursion into sociotechnical cultures emerges from the intimate, messy entanglement of HIV/AIDS and digital cultures. HIV is a retrovirus exchanged through certain bodily fluids (namely, blood and semen) that attacks cells by multiplying itself to detrimental effect. Starting in the 1970s, computing cultures widely adopted the term virus to describe how pieces of code can copy themselves into other locations, and propagate within an individual computer across a network, to devastating result (Casilli 2010; Gozzi 1990; Lupton 1994). Personal computing and HIV both entered public discourse in the 1980s and 1990s, synchronicity that further embedded this metaphor. As one of us has previously charted with Dylan Mulvin, viral analogies and explanations honed in the context of HIV/AIDS were used to explain “the risks of vulnerability in complex, networked systems” (McKinney and Mulvin 2019, 476).

Computer viruses, then, are routinely and repeatedly equated with biological viruses (Parikka 2007, 120). Conversely, HIV/AIDS also became a powerful means of articulating how digital media and popular culture operate. Viral phenomena, biological or otherwise, describe moments when a few nodes in a network spread information widely and rapidly (Parikka 2007, 131). Douglas Rushkoff (1996) asserts that through the transmission of messages that “attach” themselves to people, viral media shape thoughts and perceptions, reproducing themselves. Virality in digital cultures operates in tandem with capitalism, emerging from viral marketing: the concept of “going viral” first appeared to name how images, memes, videos, or advertisements circulate online rapidly and across vast networks (Cifor and McKinney 2020). Viral in this context points to how network structures might inform economic models.

In this paper, we bring these varied uses of viral metaphor into conversation, to think about how viral metaphors operate as models. Models are manipulable, scalable objects that various disciplines (economics, epidemiology, physics, network science) use to think with (Morgan 2012). Models can be thought of as metaphors because they illustrate one phenomenon in and through another that is better understood or seemingly more transparent, for example, by mapping the “certainty” of Bluetooth proximity onto opaque, potential virus spread (Bailer-Jones 2000; Derman 2010). Understanding models as metaphors can engender a more liberatory science, foregrounding their epistemic as well as social functions (Ravetz 2003, 73; Sismondo 1999).

We consider a deliberately capacious, heterogenous range of viral network models, drawing out the shared tension between what they promise to lay bare and what they must externalize to accomplish that work. First, we consider what might be the formative case of the viral metaphor as model: the Patient Zero sociogram (1984). This model was used to locate blame for the origins of HIV in the network connections of a single French Canadian flight attendant, and is widely understood in AIDS history as an erroneous, harmful example of how models produce stigma for vulnerable populations. Next, we connect lessons that critical HIV studies has drawn from this case to read the turn to digitally driven COVID-19 models. We offer a speculative reading of the standard Alice and Bob figures from cryptography used by Apple and Google to model their contact-tracing API to the public. Then we analyze a small, interpretive selection of well-cited articles in leading journals published as COVID-19 emerged as a global pandemic in 2020, from the intersection of epidemiology and network science, that use social network models or smartphone data. We also consider the tech- and telecom-driven public data projects associated with this type of research. Each of these cases represents an apparent reversal of the viral metaphor’s origins: where biological viruses were once used to understand digital technologies, digital technologies are now being used to understand a virus.

Like metaphors, models do things: they are excitable, anticipatory, “calculative machines of invention” (Rhodes et al. 2020, 4; Rhodes and Lancaster 2021): they both reflect the worlds that their creators work from and create worlds that are imaginable within their term (Morgan 2012, 30–31; Anderson 2021, 177–78). Following from the Patient Zero sociogram, we argue that the latent sexual politics of these COVID-19 models are entries to framing how models understand bad actors, surveillance, and risk. The kinds of people and behaviors that are opaque to these models and their methods of capture present potential avenues for speaking back to digital virality’s terms. We highlight these exceptions through a queer analysis that shows how certain lives make trouble for models and their sensibilities, telling of queer forms of life, desire, and contact that evade modeling altogether. In what follows, we push on the material entanglements of multiple viruses in promiscuous ways. Katherine McKittrick argues in her work on Blackness and science that metaphors risk “fixing social identities in place” and so we must “sit with metaphor(s)” and their materialities, to see how and what they illuminate (2021, 10–11). And so we lean into models as sites where the viral metaphor is stretched thin. Moments when digital networks cannot adequately map on to complex forms of intimate life demonstrate network modeling’s limits in ways that are generative, because they point toward the “critical promiscuity” needed to richly understand and represent crisis (Anderson 2021, 180). With our method we propose that a metaphor we sit with in order to stretch it to its limits might be easier to snap.

Models of COVID-19 always carry the fraught legacies of viral metaphors and meanings honed through the North American HIV/AIDS crises over the last four decades. The case that gave rise to Patient Zero powerfully illustrates how virus modeling creates meanings and effects. In 1984 the infamous sociogram, titled “Sexual contacts among homosexual men with AIDS,” birthed a convenient mythic origin story that relied on claims of sexual deviance, risk-taking, and individual responsibilization. The sociogram (Figure 2) first appeared in an American Journal of Medicine paper published by four Centers for Disease Control and Prevention (CDC) researchers (Auerbach et al. 1984). Their study uncovered that the then-mysterious infectious agent that would eventually be called HIV was probably sexually transmitted. The researchers examined the earliest AIDS cases among a “cluster” of nineteen “homosexual men” residing in Southern California, tracing contacts of thirteen who shared the names of sexual partners (Auerbach et al. 1984, 488). The researchers discovered that in the five years preceding illness onset, nine had sexual linkages. Eventually, this sexual network would grow to map out connections between forty men in ten cities.

Importantly, this model has a center: Gaëtan Dugas (1953–1984), a French Canadian flight attendant. The paper linked four California men who each had sex with one man from outside the state (Dugas), who was also the partner of four men living with AIDS in New York City. The researchers placed an anonymized Dugas at the center of their sociogram. Its caption explained, “Lines connecting the circles represent sexual exposures. Indicated city or state as place of residence of a patient at the time of diagnosis. ‘O’ indicates Patient 0 (described in text)” (Auerbach et al. 1984, 488). They identified the man at the sociogram’s center as “‘Patient 0’—a ‘non-Californian AIDS patient’” (488) and a possible carrier of the infectious agent. The origin of confusion is that the letter “O” stood for “Outside” [of California], but because the researchers also numbered cases in the cluster by the date of AIDS-symptom onset, the letter “O” was misinterpreted as the number zero. From that point forward, the man at the center was misidentified as “Patient Zero.” Dugas was thus framed as individually responsible for the North American AIDS epidemic. The researchers behind the cluster study maintained from the start that the person labeled “Patient 0” was not likely the originator of AIDS for the patient cluster studied, let alone for the wider American epidemic. Yet a misidentification as “Patient Zero,” which in popular parlance denotes an original or primary case of contagion within a network, proved damning.

Dugas was first identified by name in Randy Shilts’s 1987 bestseller And the Band Played On. He appeared as a sociopath who allegedly intentionally infected or, at minimum, recklessly endangered a promiscuous sexual network of gay men in coastal urban meccas. As Ted Kerr (2016) has noted, the popular introduction to Patient Zero was through the media leading up to the book’s premiere. Shilts’s publisher included the following in the press release: “what remains a mystery for most people is where AIDS came from and how it spread so rapidly through America. In the most bizarre story of the epidemic, Shilts also found the man whom the CDC dubbed the ‘Patient Zero'’ of the epidemic” (Crimp 1987; Kerr 2016). Shilts’s claim was widely covered in the media, appearing in news stories under headlines like that published by New York Post, “THE MAN WHO GAVE US AIDS,” and on television, where in November 1987, as Robert McKay (n.d.) reports, “millions of television viewers tuned in to a 60 Minutes news special that profiled Dugas as ‘a central victim and victimizer.’”

Arriving on the scene in 1987, six years into the clinically recognized epidemic, when more than fifty thousand Americans had already died, Dugas’s fashioning as Patient Zero made for a timely, convenient AIDS villain. As activist and epidemiologist Gregg Gonsalves notes, Dugas seemed to be a manifestation of otherness with both a “foreign-sounding name and an out-of-control sexuality” (Belluz 2016). Dugas was not officially cleared of his “Patient Zero” status until 2016, when a team of scientists led by Michael Worobey recovered near full-length genomic sequences of HIV from blood samples of Dugas as well as eight other men taken in 1978 and 1979 (Worobey et al. 2016). Yet correcting the record is not enough to undo harm. It is not the accuracy, or lack thereof, of any given virus model or modeling process that is in question here; rather, we are arguing that models reflect and reproduce a particular vision of the world, one that amplifies sexual politics through sociotechnical means.

The Patient Zero sociogram, like many COVID models, offers an explanation for viral phenomena that seems unexplainable and uncontrollable. The idea of a Patient Zero offered confirmation of the conflation of homosexuality with promiscuity and AIDS, that the virus was perpetuated through individual recklessness and deviant behaviors, and that people with HIV/AIDS got what they deserved, conveniently assuring the straight, white, middle-class public of their safety and superiority. When connected through network paths, “bad” actors model satisfying chains of blame, which some publics desired. Models offer convenient means of detracting guilt from the real and much more frightening villains found in structures of harm: stigma, discrimination, governmental ineptitude, food and housing insecurity, unsafe working conditions, and other structural violence. As we see once again in the age of COVID, pandemics and their viral metaphors exploit extant social fault lines, what Jallicia Jolly (2020) calls “pre-existing conditions,” while shoring up neoliberal individual responsibilization.

From the start, COVID-19 was a crisis represented and interpreted through models. Researchers, governments, and public health have used complex epidemiological models to predict how the airborne SARS-CoV-2 virus will spread through exchange of breath in proximity with others, and represent the potential effects of various non-pharmaceutical interventions (NPIs), such as mask-wearing, or school closures, on that spread. The use of models in these fields begins with the application of mathematical models to epidemiology starting in the 1920s, followed by the move to test social network models of information spread against epidemiological models of virus spread in the early 2000s (Kermack and McKendrick 1927; Newman 2002). What is novel about the reliance on models amidst COVID-19 is that they are not confined to the back rooms of epidemiological research or public health authorities; media outlets such as the New York Times, and tech companies such as Facebook’s (now Meta) Data for Good initiative create high-resolution, data-driven, user-oriented dashboards that reflect profound cultural investments in the visualization of data and mark the domestication of citizen engagement in modeling science (Rhodes et al. 2020, 5; Wernimont 2021). Facebook/Meta’s Data for Good Mobility Dashboard, for example, draws on “aggregated and privacy preserving” data from the social networks’ users to model how people are moving around under various lockdown measures (COVID-19 Mobility Data Network, n.d.). The model is retrospective, using real social network data gathered from mobile phones to show how movement increases or decreases over time. Similarly, another Facebook/Meta Data for Good project called COVID-19 Forecasts leverages AI-driven predictive models, developed for other, non-specified purposes, to make predictions about human behavior related to virus transmission (Meta, n.d.a).

Epidemiological modeling relies on presumptive understandings of human behavior and probability-oriented equations about how contact with others transmits viruses. These SEIR (susceptible–exposed–infectious–removed) models assume random, uniform mixing of nodes in a network, and compartmentalize each node into one of the categories for which the model type is named. Every actor in a model is either susceptible to the virus, already exposed to the virus but not yet contagious, infectious with the virus, or removed from the virus by being already recovered or deceased. Assumptions about the probability of contact and infection have always been driven by data of some kind (census data or surveys, for example) (Ogden et al. 2020). Combining SEIR models with social network models grounds probabilistic assumptions about contact in people’s online behavior. Actual data gathered from cell phones (which claim to be proxies for contact) can also enhance the accuracy of SEIR models by tethering interaction to networked practices. In line with contemporary cultural values, COVID models veer toward the data-driven, using geolocated data from cellular phones, social network models, and even AI, to test or improve upon long established epidemiological models (Chang et al. 2021; Ferretti et al. 2020; Firth et al. 2020; Grantz et al. 2020; Kissler et al. 2020; Klepac, Kissler, and Gog 2018; Wieczorek, Siłka, and Woźniak 2020).² Warwick Anderson argues that these data have often been posited as solutions to SEIR’s models’ simplistic assumptions about behavior and proximity during the COVID-19 pandemic (2021, 176).

Actual data from smartphone users, and data-driven social network models, each normally used to understand information spread or digital virality, are being used to hone models of an actually existing biological virus’s spread. The epistemic structure through which this honing happens is a digital metaphor, where one thing is compared to another to generate knowledge: online social networks are like one’s broader range of contacts, or proximity measured by Bluetooth is like the proximity of virus transmission. In other words, knowledge about how social networks and internet-enabled mobile devices shape our interactions with others drive the creation of models about this virus, and popular understandings of how it spreads. AIDS crisis models such as the Patient Zero sociogram shaped emerging understandings of the promiscuity of (digital) networks and how we are connected to each other in relationships of mutual vulnerability. With COVID-19 models, digital networks are being used to shape emerging understandings of virality, biologically speaking. This phenomenon is not new, but rather a digital rendering of a century-old association Heidi Tworek (2019) identifies between virality, information spread, and globalized systems thinking under modernity.

As in the case of AIDS, models as metaphors do cultural work in this latest crisis, mediating how stigma around certain behaviors that engender intimacy and vulnerability to exposure are represented and interpreted. Black, Indigenous, and people of color (BIPOC), those denied health and economic opportunities, queer and trans people, incarcerated people, immigrants and refugees, and people who use drugs form what Stephen W. Thrasher (2020), calls the “viral underclass” of both AIDS and COVID, people who are harmed not just by biological viruses but by the societal structures that render them and their communities most vulnerable to both viral transmission and inadequate post-transmission care. As studies of the Patient Zero example have shown, viral models engender all kinds of “externalities,” shoring up personal responsibility for risk mitigation over other stories about complex structural violence that are harder to model, solve, and predict. Even with a virus whose spread is not primarily through sexual contact, COVID-19 is still about intimacy: the proximity (and length of proximity) required to share breath and transmit infection, and choices about the formation of intimate social bubbles as a means of protection. COVID models give a noble purpose to algorithmic surveillance and data extraction overreach by both government and industry. They do so, in part, through an explicit as well as a tacit sexual politics about intimacy, proximity, contact, exposure and risk, which carries in it the legacies of viral metaphors honed through HIV/AIDS.

To begin to sketch out the sexual politics embedded in COVID models, we might start with Bob and Alice, familiar standard figures from the ways cryptography problems are modeled and taught to students. Alice and Bob are a fictional couple invented to make the cryptographic exchange of information between people easier to understand. As Alanna Cattapan and Quinn Dupont (n.d.) have shown, Alice and Bob refract the heteronormative gender ideals at the heart of computer sciences, a discipline central to COVID modeling. Often depicted in romantic, straight entanglements to illustrate crytographic scenarios and problems, they show how intimacy and sexuality fundamentally inform how computing disciplines model information exchange.

In April 2020, Google and Apple used Alice and Bob to announce their Exposure Notifications API, which would be used widely by public health authorities in the United States, Canada, and elsewhere to generate their own apps for Bluetooth proximity-enabled contact tracing. Users of iOS or Android who download one of these apps and opt-in can disclose future infections and automatically alert those whose phones have been intimately proximate to theirs during the assumed period of infectiousness. These data can be used by public health authorities in contact tracing, but also as a baseline for understanding contact and spread, extrapolating future infection levels via modeling.

Data-driven contact-tracing apps present crucial questions about the normalization of smartphone-enabled surveillance and the collaboration of public health and university researchers with a tech industry built on extracting profit from user data (Andrejevic 2013; Benjamin 2019; Eubanks 2018). Bob and Alice playfully walk users past these political hurdles, in program announcement illustrations that explain cryptography to potential adopters of the apps (Figure 3). Designed like a comic strip, the first frame features two line drawings of a man and a woman sitting on opposite ends of a public bench, as close to two meters (or six feet) apart as the architecture allows. Alice is wearing a gray shirt and has blond hair, while Bob’s shirt is blue. Both are maskless. His hair, and both figures’ skin, have not been colored in. They are quintessential white defaults (Daniels 2013). The caption reads, “Bob and Alice meet each other for the first time and have a 10-minute conversation.” Bob seems to lean in toward Alice, but she remains resolutely upright. He likes her, she’s not into it. As Alice pulls away from Bob’s subtle advance along the bench, the “anonymous identifier” from her COVID alert app moves silently toward Bob’s phone (Agostinho and Thylstrup 2019; Chun and Friedland 2015; Thylstrup, Waseem, and Agostinho 2020). In the next frame, Bob is alone, leaning over a table, haggard and frowning, a drop of sweat on his brow: “Bob is positively diagnosed for COVID-19.” Alice will be alerted that one of her recent contacts tested positive thanks to the exchange of anonymous cryptographic keys from their cell phones.

Figure 3. Marketing material from Apple’s and Google’s Exposure Notifications API, 2020

Why are Bob and Alice meeting for ten minutes on a park bench “for the first time” during a pandemic? They could be new business associates, or random strangers sharing a bench, but given the ways Alice and Bob are often deployed as a couple, the other subtext here is a first date gone poorly: a safe-ish gander toward intimacy with someone new risked in a moment where dating and hook-ups have become taboo. The sanitized line drawing of Bob’s sick body, hunched over his contact-tracing app a few days after the encounter presents a stark contrast between the sanitized digital sphere and messy, ill bodies. As Deborah Lupton writes, “In a sociocultural context in which interactions with the fleshly and potentially virally contaminated bodies of other people have been continually problematised as life-threatening, the remoteness and hygienic technological imaginaries associated with digitised COVID technologies promise to offer greater safety and security” (2022, 68). Alice and Bob’s encounter also raises a more fundamental problem about what happens when we allow the exchange of cryptographic keys over Bluetooth to stand in for the intimate exchange of touch or wet breath. As the Canadian government’s fact sheet on the Apple-Google alert system explains, Bluetooth does not understand how humans actually live in proximity to others. “If you live in an apartment, condo, or rooming house,” the app might not work too well: “the technology isn’t perfect.” It continues, “Bluetooth signals can be affected by things like metals or even microwave ovens” (Government of Canada, n.d.).

Apple, Google, and Meta are merely the largest tech companies to jump on the COVID Data for Good bandwagon. In Canada, TELUS, one of the country’s largest telecom providers, began offering aggregated and disidentified data from its nine million users (significant given Canada’s population is thirty-eight million) to university researchers in spring 2020. McKinney received an email from a research officer at their university promoting the program in April 2020, forwarding information originally sent from an executive at TELUS. The drive behind this program is TELUS’s assurance that their user data offers something unprecedented for pandemic response, by giving otherwise unattainable transparency to people’s movements and potential interactions with others. This transparency is to be achieved by comparing TELUS network user data with other existing “analog” data about COVID-19, such as the number of positive diagnoses in a region.

Drawing on metaphor’s promise to know through comparison, TELUS’s program marketing explains that comparison can be used “to find correlations in the data that could help our governments and health authorities develop public policy and determine where to allocate much needed resources” (n.d.). Though TELUS’s user data is offered for free to university and public health researchers across disciplines, and contains a sunset clause, the program also includes a provision to share data with commercial entities and charge for the service, for example, to allow “banks to better predict the economic impact of COVID-19” (TELUS, n.d.). “Data for good” models aimed at slowing viral spread are just as “good” when they model the futures of financial capital, which shows how portable viral metaphors can be. The use of mobile data in viral modeling is not a new phenomenon from COVID times, but the crisis made industry offerings of people’s intimate data to researchers into a public relations strategy powered by the urgency of pandemic response (Lupton, 2022, 70). As Facebook’s Data for Good marketing and privacy disclosure material explains, “People are distancing to protect their communities, healthcare workers are saving lives on the front lines, and public health systems are looking to put the right guidelines in place. To do that, they need better information on whether preventive measures are working and how the virus may spread” (Meta, n.d.b). The drive toward modeling as a means of social good, and an urgent and anticipatory response to crisis is key to justifying this transition, where data can be used to validate existing models, or make the variables guiding those models more precise and agile, in order to save lives and prop up economies.

Models promise transparency and resolution to troubling uncertainty, but a closer analysis of their sexual politics, cued to virality as a metaphor born when digital networks and sexual networks are overlaid, points toward what these models cannot capture. As Wendy Hui Kyong Chun shows, network science is an interdisciplinary field that “merges the quantitative social sciences with the physical and computer sciences in order to bypass or eliminate the humanities and media studies, two fields also steeped in theories of representation and networks” (2018, 68). Following this call from Chun and others (Anderson 2021; Rhodes et al. 2020), we bring humanities approaches grounded in queer studies into dialog with network modeling. Learning from the disaster that was the Patient Zero model, we read a selection of COVID modeling projects driven by network data through critical HIV studies and queer theorizations of networks. Together these fields show that illustrations of interdependence always look much simpler than the intimacies they claim to represent. This practice examines how COVID models driven by data from and about digital networks are assembled, what they claim to know, and what they exorcise as remainders.

What actually went down on Alice and Bob’s date is opaque to a Bluetooth proximity network’s way of knowing, flying under the radar of how infection models based on cell phone data imagine the world. These remainders are examples of what Black and queer of color digital studies, drawing on Édouard Glissant, have called opacities, a key modality of refusal toward data extraction and mass surveillance, grounded in the materialities of Black life (Browne and Blas 2017; Cho 2018; McGlotten 2016)³. As Shaka McGlotten has argued, queers and people of color “embrace those forms of darkness in which identity is obscured or rendered opaque. There are no coherent, rational, self-knowing subjects here, just furious refusals. These refusals are a kind of black ops, a form of black data that encrypts without hope of a coherent or positive output” (2016, 278). McGlotten emphasizes both an active stance of refusal toward the ways big data claims to know, but also forms of life, relationality, and interdependence (the embrace) that simply aren’t knowable by these systems. Refusal is a means of ensuring reciprocity and accountability in contrast to the naturalization of normalized extractive data regimes (Cifor et al. 2019; Simpson 2014). Refusal questions the underlying neoliberal assumptions of social “benefit” or “good” and transparency embedded in contemporary data systems.

One of the most highly cited articles from the early days of social network epidemiology is M.E.J. Newman’s “Spread of Epidemic Disease on Networks” (2002). Newman’s work is invoked in many COVID modeling projects. This article is an early example of network topology drawn from the rise of early-aughts internet culture applied to the question of viral spread, research that took shape in a depoliticized, millennial period Kerr (2016) has named the Second Silence of North American AIDS crisis revisitation. The model maps “social,” “technological,” and “biological networks” onto each other, using network topologies to improve the precision of contacts assumed by SEIR models (Newman 2002, 1).

The pages of impressive math equations offered in Newman’s article are far beyond our wheelhouses, but one aspect of his modeling is familiar: it is illustrated through the case of a sexually transmitted infection (STI) moving through a given population. Two decisions are made about the network topology of this population: (1) it is made up by men and women who (2) are “bipartite” in their contacts and presumably their desires (“women” only have sex with “men” and “men” only have sex with “women”) (Newman 2002, 8). Newman centers a Patient Zero at the outset of this model: “Now consider an outbreak that starts at a single individual, who for the moment we take to be male. From that male the disease will spread to some number of females, and from them to some other number of males, so that after those two steps a number of new males will have contracted the disease” (9). The article concludes by showing how the curve might plummet if the most “active members of the network” (those who have the most sex, with the most people) reduce their number of partners: “targeting preventive efforts at changing the behavior of the most active members of the network may be a much better way of limiting the spread of disease than targeting everyone. This suggestion is certainly not new, but our models provide a quantitative basis for assessing its efficacy” (10). Here the model evidences the safe-sex pedagogies of personal risk management in contrast with the willfully promiscuous “bad actors” who emerged directly from the AIDS crisis (Patton 1996).

This is a model without queers, a model where gender is biological sex, and all sexual encounters are created equal (blow jobs, rimming, or other low-risk practices queer communities design to reduce harm are opaque to these equations). As Newman explains, the STI example is deliberately simple in order to illustrate. These simplifications of real-world phenomena are fundamental to network science: they are features, not bugs (Chun 2018, 70). The bipartite STI example is, as Morgan shows of models, small in scale but representative of something larger, a manipulable object that a discipline can use to reason with (2012, 13). It is also, however, indicative of the world this model and the disciplines through which it emerges invokes when it selects the STI as a suitable stand-in for how networks relate to virus spread (Mulvin 2021). This kind of choice in network science, Chun argues, is “performative” because “it puts in place the world it discovers” (2018, 62). And following these choices, and the ways they seem like common sense, helps us to understand both the normative sexual politics of the viral metaphor, and the ways it might be fucked around with. Imagine a bisexual node running rampant through this bipartite universe.

Sexual politics are fundamental to any network form (Barnett et al. 2016; Keeling 2014; McBean 2020). Network diagrams such as sociograms model how we are joined with others in forms of exchange that may or may not be palpable to us without the practice of modeling. Think, for example, of the partner of your partner’s former partner, who you may or may not be able to name (unless you are a lesbian, in which case you can definitely name them and have probably also dated them, see McBean 2020). Even when a network diagram maps a tweet’s impressions as it reverberates through a digital network, the process of tracking unanticipated connections and their fallout is grounded in a libidinal economy: a drive toward making secret connections transparent, in order to predict what intimacy does in and to the world. The sexual politics at the heart of COVID modeling reflect a drive toward several strategies for connection predicated on social network concepts that replicate conservative sexual mores grounded in compulsory, perfectly monogamous heterosexuality.

In the article “Social Network-Based Distancing Strategies to Flatten the COVID-19 Curve in a Post-Lockdown World” (Block et al. 2020), the researchers layer epidemiological curves onto path length differences in social network models, to make three suggestions on how modifications to one’s networked connection can slow viral spread. The strategies are

(1) reduce contact with people whose group memberships are different than yours because of where they work or live;⁴
(2) reduce contact with any of your usual contacts who tend to have lots of varied contacts with others;
(3) build bubbles that repeat contact with the same people.

These strategies are no doubt effective for reducing the overall spread of an airborne illness in a given population. But they can also be recast, crudely as

(1) eliminate difference from your life,
(2) cast out promiscuity,
(3) choose monogamy.

As even the authors acknowledge, “advocating the creation of small communities and contact with mostly similar others on some dimensions could potentially result in the long-term reduction of intergroup contact and an associated rise in inequality” (Block et al. 2020, 594). Their model might look quite different if it took into account other kinds of alterations to network paths, or ways of talking about “intergroup contact” (594) that queer people have developed in practice in response to challenges like HIV, such as network sero-sorting (choosing partners of the same HIV status), disclosing HIV status and viral load, or choosing low-transmission sex practices.⁵ The point is not that these practices might be adapted to a different and airborne virus altogether, but rather that when you begin with a fundamentally different way of understanding sexuality and relation, different metaphors altogether, other kinds of models might emerge for imagining and altering network paths.

Queer COVID memes make reference to these other modalities. For example, “let Alice Pieszecki do the contact tracing” by Taylor Hatmaker (2020) pictures a screen shot of Showtime’s The L Word character Alice typing on an Apple laptop in front of the now-iconic-in-lesbian-culture sociogram of lesbian hook-ups that drives the narrative action in the early series (see McBean 2020) and that launched the short-lived lesbian social networking site OurChart.com. Similarly, the viral YouTube video “Contact Tracys” by The Chaser (2020) features two irreverent teenage girls named Tracy hired by an Australian public health authority to do contact tracing using their smartphone-enabled social media stalking skills. These skills are framed as inherent to digital youth, but also dependent on Tracy’s and Tracy’s queer sexualities, which allows them unique access to networks. As the square, middle-aged white male public health officer asks, trying to navigate his own smartphone, “Wait, do these methods work if you’re trying to find a woman?” The Tracys: “Uh, yeah, we’re bi, obviously.” Alice and the Tracys expand the authority and expertise commonly associated with doing data science as they reimagine and refract viral models to queer ends that are grounded in recognition of “the heterogeneity of collectives and the diversity of relations within and between them” (Anderson 2021, 173). These memes beg the question, what would a COVID model made by Alice and the Tracys look like?

Amongst other fields, critical HIV studies has thoroughly modeled how crisis conditions beget violent transformations in everyday forms of surveillance (Liang, Hutson, and Keyes 2020; McClelland 2019). For example, Stephen Molldrem (2020) shows how in the name of HIV prevention (a presumed social good), we have witnessed the creation and maintenance of surveillant viral load and CD4 T-cell count databases by public health authorities in the US that do not obtain consent from their subjects, engender data creep across contexts, and enact viral hierarchies. The widespread public acceptance of telecom and social media companies offering user data as a public relations strategy “for good,” and network science readily drawing on these datasets with small friction from institutional ethics review boards, may end up being one of the surveillant transformations this pandemic leaves in its wake. As HIV has taught, everyday digital surveillance measures are first tested out on populations made vulnerable, and then expanded to reach everyone else (Eubanks 2018). Once in place under crisis, such surveillance is not rolled back when (and if) the emergency ends (Emerson 2020).

The ways that the COVID-19 pandemic, like AIDS, has disproportionately harmed Black and Brown people is well documented. Digital data-driven models are one way to evidence the connection between these uneven harms and phenomena such as high-density living situations, or continuing to commute for work in front-line jobs. These models can also advance harms related to surveillance (Lupton 2022, 63–64). A COVID model of how people in different census tracts in the United States move to and from geolocated “points of interest” (restaurants, stores, gyms) finds that people from low-income areas “made substantially more visits per capita to grocery stores than did those from higher-income CBGs [census block groups] and consequently experienced more predicted infections” (Chang et al. 2021, 86). The model shows that grocery stores in poor neighborhoods are more crowded, visited more often, and that shoppers spend more time in the store per visit. This model provides data-driven evidence for what Black and Brown people already know about food security in urban areas, in order to make predictive recommendations about the “disparate effects of reopening plans” (86). It does so using two datasets that are worth contextualizing for what they can and cannot see. First, census data about race, for which the model only draws on the categories of white and Black, due to the limits of race-based data collected for the US census. The cities this model analyzes include Los Angeles and Miami, which both have significant Latinx populations who are particularly affected by COVID-19. In a similar vein, the second dataset is SafeGraph (n.d.), which draws on foot traffic data sold for retail analytics. It brings together data from “thousands of diverse sources” that are not named. The SafeGraph data is validated in this study against Google location services data, which is also widely used by law enforcement agencies via Google’s Sensorvault platform. This model exists to benefit vulnerable populations, offering evidence-based recommendations for food distribution programs in economically depressed areas, amongst other non-pharmaceutical interventions. But it does so using data-extractive tools that make these same people more vulnerable to surveillance harms, such as predictive policing or the digital tracking of Electronic Benefits Transfer (EBT) use at grocery stores (Benjamin 2019; Eubanks 2018).

Network science is grappling with these issues of data-driven inequity explicitly. Grantz et al. (2020), a group of epidemiologists concerned about the fast uptake of mobile data for COVID modeling, argue in Nature that there are many things to consider here about these datasets: they do not capture the effects of more people staying at home, where they are more likely to use internet-enabled technologies that don’t generate Call Details Records; older people and very young people don’t use cell phones as often, and aren’t represented in data being used to model a virus that impacts them profoundly; in some households, people share a single phone, or a single SIM card, or one person might use several phones and SIM cards. The assumption here is that accounting for, or even eliminating bias from the data is the goal, rather than a fundamental transformation of the racist, oppressive systems that render extractive datasets recognizable as “solutions” in the first place (Benjamin 2019). As Grantz et al. argue, “The use of mobile phone data, particularly forms such as those proposed through contact tracing applications, must be weighed against the possible infringements of privacy and civil liberties versus the potential public health benefit” (2020, 4). The problem at the heart of Grantz et al.’s article and other such work is that this weighing has predictable results when crisis is used to balance the scales.

These critical epidemiologists are concerned with how their field is using mobile data, and as they lay out these concerns, a queer sort of critique emerges: “People are different” from each other (Sedgwick 2008, 24). Those who share SIM cards or have burner phones aren’t tidy nodes. Shaken here by Grantz et al. are the assumptions network science makes about what it means to be a human who is contained, rather than porous (King 2019, 115), and what proximity means as an erotic relation to someone else. Following from McGlotten (2016), and also earlier work in queer of color critique such as Cathy Cohen’s work on Blackness, queerness, and HIV, we ask what it would mean to think of these scenes of failed homophilyz—spending too much time in grocery stores or sharing SIM cards—as queer, for the ways they take up marginal positions in relation to power that might be the basis for transformative coalition work (Cohen 1997, 438, 462). If the Patient Zero model has done anything useful in spite of its violence, it has cued our attention to what models as heuristics claim to do in the world, and piqued our skepticism about their inadequacies to intimate lives lived in proximity to others.

It is not that the COVID-19 models we have analyzed in this article are not useful, or do not accurately predict the future. Often, they do, as studies that retrospectively compare modeling with testing data show (Chang et al. 2021). As humanist researchers who work on the history and the media of HIV, we do not claim to offer a “transparent” critique of what data and public health scientists are doing: they are, themselves fully aware of and experts on, the benefits but also limitations of their own modeling and the dangers they pose (Grantz et al. 2020). We are not arguing that we ought to do away with models: they are crucial forms of response to uncertainty, and ways of representing how decisions in the present might shape the future (Rhodes and Lancaster 2021). As metaphors that can be scaled and manipulated, models are convincing, and in the COVID-19 pandemic they informed more equitable vaccine roll-outs, and led officials in conservative regions to impose unpopular mask mandates because they were the right thing to do.

But models always represent a particular view of things as they are and might be. They are “excitable matter” in that they offer certainty in precarious times, generating forceful anticipatory affect (Rhodes and Lancaster 2021). As objects that we use to think with, they make possible knowing about the world and its horizons through the terms that they set (Morgan 2012). In this case, these terms understand a biological virus as being like a digital network, which justifies data extractive regimes and future surveillance measures, and stigmatizes “bad” actors. We have offered a perspective, informed by sexuality and critical HIV studies, on what understanding the sexual politics behind viral models and metaphors can do. This is a queer imperative, because we only sort of know what stigmas flow from COVID-19 modeling (not to mention from emerging virus models such as monkeypox): how they will be used against people who party, do drugs, and hook up, and how they will place blame on poor, Black, Brown, queer and trans people’s necessary proximities to each other. They are in danger once again of advancing individual responsibilization, rather than dismantling the structural violence that actually drives infection, death, and harm. How these COVID-19 models and the viral metaphors they embody are used to imagine a better future will shape the actual future worlds we all occupy.

¹ The creator of the “Covid-19 sent you a friend request” was not readily discoverable. Rather, the meme, in characteristic meme fashion, circulated virally by users across social media platforms.

² Louise Amoore argues that these data-driven developments represent an epistemic age of correlation that is beginning to challenge the usefulness of theoretical models in the first place (2020, 44).

³ Glissant is briefly critical of modeling science’s audacious claim to capture the dynamics of Black relation as projections of the future (“futurology”) or “energized structures” (1997, 173), and this is closely connected with his concept of opacity.

⁴ This is called homophily in network science; see Chun 2018.

⁵ These are examples of “local expertise” and “alternative cultural logics” that might generate more adaptive and expansive modeling (Rhodes et al. 2020, 6).

Amoore, Louise. 2020. Cloud Ethics: Algorithms and the Attributes of Ourselves and Others. Durham, NC: Duke University Press.

Anderson, Warwick. 2021. “The Model Crisis, or How to Have Critical Promiscuity in the Time of Covid-19.” Social Studies of Science 51 (2): 167–88. https://doi.org/10.1177/0306312721996053.

Andrejevic, Mark. 2013. Infoglut: How Too Much Information Is Changing the Way We Think and Know. New York: Routledge. https://doi.org/10.4324/9780203075319.

Auerbach, David. M., William W. Darrow, Harold W. Jaffe, and James W. Curran. 1984. “Cluster of Cases of the Acquired Immune Deficiency Syndrome. Patients Linked by Sexual Contact.” The American Journal of Medicine 76 (3): 487–92. https://doi.org/10.1016/0002-9343(84)90668-5.

Bailer-Jones, Daniela M. 2000. “Scientific Models as Metaphors.” In Metaphor and Analogy in the Sciences, edited by Fernand Hallyn, 181–98. Dordrecht: Springer Netherlands. https://doi.org/10.1007/978-94-015-9442-4_11.

Barnett, Fiona, Zach Blas, Micha Cárdenas, Jacob Gaboury, Jessica Marie Johnson, and Margaret Rhee. 2016. “QueerOS: A User’s Manual.” In Debates in the Digital Humanities, edited by Matthew K. Gold and Lauren F. Klein. Minneapolis: University of Minnesota Press. https://dhdebates.gc.cuny.edu/read/untitled/section/e246e073-9e27-4bb2-88b2-af1676cb4a94.

Benjamin, Ruha. 2019. Race after Technology: Abolitionist Tools for the New Jim Code. Cambridge: Polity.

Block, Per, Marion Hoffman, Isabel J. Raabe, Jennifer Beam Dowd, Charles Rahal, Ridhi Kashyap, and Melinda C. Mills. 2020. “Social Network-Based Distancing Strategies to Flatten the COVID-19 Curve in a Post-Lockdown World.” Nature Human Behaviour 4 (6): 588–96. https://doi.org/10.1038/s41562-020-0898-6.

Casilli, Antonio. 2010. “A History of Virulence: The Body and Computer Culture in the 1980s.” Body and Society 16 (4): 1–30. https://core.ac.uk/reader/47722430.

Cattapan, Alana, and Quinn DuPont. n.d. “Alice and Bob: The World’s Most Famous Cryptographic Couple.” Accessed March 25, 2021. http://cryptocouple.com/.

Chang, Serina, Emma Pierson, Pang Wei Koh, Jaline Gerardin, Beth Redbird, David Grusky, and Jure Leskovec. 2021. “Mobility Network Models of COVID-19 Explain Inequities and Inform Reopening.” Nature 589 (7840): 82–87. https://doi.org/10.1038/s41586-020-2923-3.

Cho, Alexander. 2018. “Default Publicness: Queer Youth of Color, Social Media, and Being Outed by the Machine.” New Media & Society 20 (9): 3183–3200. https://doi.org/10.1177/1461444817744784.

Chun, Wendy Hui Kyong. 2018. “Queerying Homophily.” In Pattern Discrimination, edited by Clemens Apprich, Wendy Hui Kyong Chun, Florian Cramer, and Hito Steyerl. 59-97. Lüneberg: meson press. https://doi.org/10.25969/MEDIAREP/12350.

Chun, Wendy Hui Kyong, and Sarah Friedland. 2015. “Habits of Leaking: Of Sluts and Network Cards.” differences 26 (2): 1–28. https://doi.org/10.1215/10407391-3145937.

Cifor, Marika, Patricia Garcia, T.L. Cowan, Jasmine Rault, Tonia Sutherland, Anita Say Chan, Jennifer Rode, Anna Lauren Hoffmann, Niloufar Salehi, and Lisa Nakamura. 2019. Feminist Data Manifest-No. https://www.manifestno.com/.

Cohen, Cathy J. 1997. “Punks, Bulldaggers, and Welfare Queens: The Radical Potential of Queer Politics?” GLQ 3 (4): 437–65. https://doi.org/10.1215/10642684-3-4-437.

Crimp, Douglas. 1987. “Introduction: AIDS: Cultural Analysis/Cultural Activism.” October 43 (Winter): 3–16. https://www.jstor.org/stable/3397562.

Daniels, Jessie. 2013. “Race and Racism in Internet Studies: A Review and Critique.” New Media & Society 15 (5): 695–719. https://doi.org/10.1177/1461444812462849.

Eubanks, Virginia. 2018. Automating Inequality: How High-Tech Tools Profile, Police and Punish the Poor. New York: St. Martin’s Press.

Ferretti, Luca, Chris Wymant, Michelle Kendall, Lele Zhao, Anel Nurtay, Lucie Abeler-Dörner, Michael Parker, David Bonsall, and Christophe Fraser. 2020. “Quantifying SARS-CoV-2 Transmission Suggests Epidemic Control with Digital Contact Tracing.” Science 368 (6491): 1–7. https://doi.org/10.1126/science.abb6936.

Firth, Josh A., Joel Hellewell, Petra Klepac, Stephen Kissler, Adam J. Kucharski, and Lewis G. Spurgin. 2020. “Using a Real-World Network to Model Localized COVID-19 Control Strategies.” Nature Medicine 26 (10): 1616–22. https://doi.org/10.1038/s41591-020-1036-8.

Glissant, Édouard. 1997. The Poetics of Relation, translated by Betsy Wing. Ann Arbor, MI: University of Michigan Press.

Gozzi, Raymond. 1990. “The Computer ‘Virus’ as Metaphor.” ETC: A Review of General Semantics 47 (2): 177–80. https://www.jstor.org/stable/42577199.

Grantz, Kyra H., Hannah R. Meredith, Derek A. T. Cummings, C. Jessica E. Metcalf, Bryan T. Grenfell, John R. Giles, Shruti Mehta, Sunil Solomon, Alain Labrique, Nishant Kishore, Caroline O. Buckee, and Amy Wesolowski. 2020. “The Use of Mobile Phone Data to Inform Analysis of COVID-19 Pandemic Epidemiology.” Nature Communications 11 (4961). https://doi.org/10.1038/s41467-020-18190-5.

Keeling, Kara. 2014. “Queer OS.” Cinema Journal 53 (2): 152–57. doi:10.1353/cj.2014.0004.

Kermack, William Ogilvy, and Anderson G. McKendrick. 1927. “A Contribution to the Mathematical Theory of Epidemics.” Proceedings of the Royal Society of London. Series A, Containing Papers of a Mathematical and Physical Character 115 (772): 700–21. https://doi.org/10.1098/rspa.1927.0118.

King, Tiffany Lethabo. 2019. The Black Shoals: Offshore Formations of Black and Native Studies. Durham, NC: Duke University Press.

Kissler, Stephen M., Petra Klepac, Maria Tang, Andrew J. K. Conlan, and Julia R. Gog. 2020. “Sparking ‘The BBC Four Pandemic’: Leveraging Citizen Science and Mobile Phones to Model the Spread of Disease.” BioRxiv, May, 479154. https://doi.org/10.1101/479154.

Klepac, Petra, Stephen Kissler, and Julia Gog. 2018. “Contagion! The BBC Four Pandemic – The Model behind the Documentary.” Epidemics 24 (September): 49–59. https://doi.org/10.1016/j.epidem.2018.03.003.

Liang, Calvin, Jevan Alexander Hutson, and Os Keyes. 2020. “Surveillance, Stigma & Sociotechnical Design for HIV.” First Monday 25 (10). https://doi.org/10.5210/fm.v25i10.10274.

Lupton, Deborah. 1994. “Panic Computing: The Viral Metaphor and Computer Technology.” Cultural Studies 8 (3): 556–68. https://doi.org/10.1080/09502389400490361.

———. 2022. “The Quantified Pandemic: Digitised Surveillance, Containment and Care in Response to the COVID-19 Crisis.” In Everyday Automation: Experiencing and Anticipating Emerging Technologies, edited by Sarah Pink, Martin Berg, Deborah Lupton, and Minna Ruckenstein, 59–72. https://doi.org/DOI: 10.4324/9781003170884.

McClelland, Alexander. 2019. “‘Lock This Whore Up’: Legal Violence and Flows of Information Precipitating Personal Violence against People Criminalised for HIV-Related Crimes in Canada.” European Journal of Risk Regulation 10 (1): 132–47. https://doi.org/10.1017/err.2019.20.

McGlotten, Shaka. 2016. “Black Data.” In No Tea, No Shade: New Writings in Black Queer Studies, edited by E. Patrick Johnson, 262–86. Durham, NC: Duke University Press.

McKinney, Cait, and Dylan Mulvin. 2019. “Bugs: Rethinking the History of Computing.” Communication, Culture and Critique 12 (4): 476–98. https://doi.org/10.1093/ccc/tcz039.

McKittrick, Katherine. 2021. Dear Science and Other Stories. Durham, NC: Duke University Press.

Molldrem, Stephen. 2020. “How to Build an HIV out of Care Watch List: Remaking HIV Surveillance in the Era of Treatment as Prevention.” First Monday 25 (10). https://doi.org/10.5210/fm.v25i10.10295.

Morgan, Mary S. 2012. The World in the Model: How Economists Work and Think. Cambridge: Cambridge University Press.

Mulvin, Dylan. 2021. Proxies: The Cultural Work of Standing In. Cambridge, MA: MIT Press.

Ogden, Nick H, Aamir Fazil, Julien Arino, Philippe Berthiaume, David N Fisman, Amy L Greer, Antoinette Ludwig, Victoria Ng, Ashleigh R Tuite,
Patricia Turgeon, Lisa A Waddell, Jianhong Wu. “PHAC SEIR Model on COVID-19.” 2020. National Collaborating Centre for Infectious Diseases (blog). October 19, 2020. https://nccid.ca/phac-seir-model-on-covid-19/.

Parikka, Jussi. 2007. Digital Contagions: A Media Archaeology of Computer Viruses. New York: Peter Lang.

Patton, Cindy. 1996. Fatal Advice: How Safe-Sex Education Went Wrong. Durham, NC: Duke University Press.

Ravetz, Jerry. 2003. “Models as Metaphors.” In Public Participation in Sustainability Science: A Handbook, edited by Bernd Kasemir, Jill Jager, Carlo C Jaeger, and Matthew T Gardner, 62–78. Cambridge: Cambridge University Press.

Rhodes, Tim, and Kari Lancaster. 2021. “Excitable Models: Projections, Targets, and the Making of Futures without Disease.” Sociology of Health & Illness 43 (4): 859–80. https://doi.org/10.1111/1467-9566.13263.

Rhodes, Tim, Kari Lancaster, Shelley Lees, and Melissa Parker. 2020. “Modelling the Pandemic: Attuning Models to Their Contexts.” BMJ Global Health 5 (6): e002914. https://doi.org/10.1136/bmjgh-2020-002914.

Rushkoff, Douglas. 1996. Media Virus!: Hidden Agendas in Popular Culture. New York: Ballantine Books.

SafeGraph. n.d. “The Source of Truth for POI Data & Business Listings.” Accessed March 29, 2021. https://www.safegraph.com/.

Sedgwick, Eve Kosofsky. 2008. Epistemology of the Closet. ACLS Humanities E-Book. Berkeley: University of California Press.

Simpson, Audra. 2014. Mohawk Interruptus: Political Life across the Borders of Settler States. Durham, NC: Duke University Press.

Tworek, Heidi. 2019. “Communicable Disease: Information, Health, and Globalization in the Interwar Period.” American Historical Review 124 (3): 813–42. https://doi.org/10.1093/ahr/rhz577.

Wieczorek, Michał, Jakub Siłka, and Marcin Woźniak. 2020. “Neural Network Powered COVID-19 Spread Forecasting Model.” Chaos, Solitons & Fractals 140 (November): 110203. https://doi.org/10.1016/j.chaos.2020.110203.

Worobey, Michael, Thomas D. Watts, Richard A. McKay, Marc A. Suchard, Timothy Granade, Dirk E. Teuwen, Beryl A. Koblin, Walid Heneine, Philippe Lemey, and Harold W. Jaffe. 2016. “1970s and ‘Patient 0’ HIV-1 Genomes Illuminate Early HIV/AIDS History in North America.” Nature 539 (7627): 98–101. https://doi.org/10.1038/nature19827.

Cait McKinney is Assistant Professor of Communication at Simon Fraser University and the author of Information Activism: A Queer History of Lesbian Media Technologies (Duke, 2020).

Marika Cifor is Assistant Professor in the Information School and adjunct faculty in the Department of Gender, Women and Sexuality Studies at the University of Washington. She is the author of Viral Cultures: Activist Archiving in the Age of AIDS (Minnesota, 2022).

Cait and Marika co-edited a 2020 issue of First Monday on Reclaiming HIV/AIDS in Digital Media Studies.