One afternoon this summer, Catherine Olsson was relaxing with a friend in San Francisco’s Mission Dolores Park when she had an idea, a little spontaneous flicker from pre-pandemic life. Why not grab dinner? She called in an order for two veggie burritos at a nearby taqueria.

Olsson popped over to the restaurant and, as she waited for her order number to be called, she looked around. She counted 10 workers standing along the open prep station assembling tacos and burritos behind a plexiglass barrier. Each worker wore a mask, but some wore them hanging off their chins or noses. Also, they were standing shoulder to shoulder. What were the odds that any of those workers had the virus that causes Covid-19, she wondered? The prevalence in San Francisco, she knew, was something like one in 300 people, but a recent antibody survey in this very neighborhood suggested frontline workers were about six times as likely to be infected as those who could stay home, which was awful to consider. And what about the five other customers shifting impatiently in the small enclosed space? At least their masks fit. She considered the steam rising from the beans and beef and wished she could crack a window and let in some air.

Olsson was beginning to regret this adventure, which was unusual because her decisionmaking process is exquisitely calculated to avoid regret. Olsson thinks about risk for a living—she works for a Silicon Valley foundation on projects that seek to mitigate the potentially catastrophic effects of advanced AI—and is in the habit of assessing her daily life with data and models. A few years ago, after a close friend told her about a scare she’d had while cycling, Olsson decided to reevaluate her own bike commute. Was her life span more likely to be cut short by a fatal crash biking to work or by the increased chance of heart disease from sitting idly on the train? She was happier riding her bike than squeezing in with fellow passengers, but sometimes feelings need a fact check. She did the math and was pleased that it validated her choice to cycle.

Olsson had begun applying this approach to living with the new coronavirus. The task was far more comprehensive. Unlike the risk of a bike accident, the risks posed by the virus radiated off of everything, turning the littlest things—a burrito!—into a gamble. At first, managing those risks was easy, if unpleasant. When the pandemic arrived in March, lockdowns constrained life and therefore made decisions simple. It was all of us together, in the interest of keeping hospitals from becoming overrun. But then, gradually, the world reopened, and life got more confusing. Perhaps tired, perhaps led astray by a government that wanted to believe the pandemic did not exist, much of the country fell into a collective delusion. If it was OK to play frisbee in the open air, then maybe sitting around inside, in a bar with those same frisbee-playing friends, wasn’t so bad, either. Or maybe the virus wasn’t that dangerous. Maybe it was even a hoax. So the number of daily cases kept rising: 150,000. 170,000. 200,000. As did the deaths.

We have vaccines now, and an end is in sight. But even optimistic projections put us at least six months from widespread inoculation. In the meantime, the pandemic is as bad as ever, and people still need to make decisions about how to behave. Even the clearest advice—wear a mask, stay 6 feet apart, avoid indoor gatherings—doesn’t address many of the subtle situations in which we find ourselves. Olsson’s response was to calculate her way through our collective apathy and disillusionment, treating the virus not as an abstract and unknowable risk but one that could be measured and tamed until a vaccine eliminated it.

Relentless tabulators often come off as zealous, maybe a little paranoid, and certainly no fun. Luckily, Olsson shares a house with fellow tabulators. She and her five housemates needed to find a way to live safely together. So they decided to adhere to a collective risk model of their own design. Any model is only as good as the data that goes into it, and the virus was too new for anyone, even experts, to have perfect information. Olsson and her housemates knew this, but they weren’t going to make the perfect the enemy of the good. They wanted to protect themselves, and by extension others, by making responsible choices. But they also wanted to be more free to actually live. Maybe math would make that possible.

That day at the taqueria, as the minutes ticked by and her risk tally rose, Olsson abandoned her burritos.

Olsson’s friends call her Catherio, after the email address she was given while studying computational neuroscience at MIT. Two and a half years ago, at 28, she was living with her partner but missing the days when she could step out of her bedroom and instantly encounter a variety of other minds. It so happened that a friend from college, Stephanie Bachar, was in the process of “forking,” like incompatible software, from a communal living situation that no longer felt homey. So one June day, they and four friends decided to join forces and move into a beige, hacienda-style townhouse in San Francisco’s Mission District. Their new home, they decided, would strike a better balance. It would be like a bash’—a type of chosen family described in Ada Palmer’s science fiction novel Too Like the Lightning as a radical “haven for discourse.” They named it Ibasho, the Japanese word from which bash’ is derived, which means “a place where you can feel like yourself.”

“Being yourself” in Ibasho meant being “slightly alternative, but professional,” says Rhys Lindmark, one of the residents. He had founded an online school for “world-class systems thinkers” after a stint researching blockchain ethics. The household was “high IQ, high EQ,” as Sarah Dobro, a primary care doctor who wears a septum ring and fauxhawk, describes it. Nerds, proudly, but socially aware nerds. They were well networked within a larger community of similar group houses around the Bay Area. It was like belonging to a more grown-up version of MIT dorms. Everyone seemed to know everyone from some salon or startup or quirky coding project. The social graph was dense.

From the start, the friends had choreographed a sense of independent togetherness. They had a communal fridge and a private one. Everyone had a different diet: paleo, vegan, gluten-free, bread lover. Every two weeks they gathered for a house meeting around a big wood-slab table, made by one of Olsson’s friends, in the room they called “the hearth.” They made decisions by consensus, following a detailed agenda with minutes and a time limit, lest the debate wear on too long. When things got a little raw—say, after two housemates moved Dobro’s pottery and Olsson’s trinkets from the fireplace mantel into a box and texted the two about the “clutter”—the group would move over to a big couch and bean bag chairs, where they could better speak with feelings, rather than logic.

Logic, however, usually ruled the day. The residents of the house were all, to varying degrees, adherents to rationalist modes of thinking and sought to reduce human biases in their day-to-day lives. As Olsson put it, the emotions they discussed on the couch provided important data, but they would return to the table to make any final decisions.

They were certainly people who could easily grasp the implications of exponential growth. So last winter, as the novel coronavirus hit far-off places, the residents of Ibasho girded themselves. In late February, at their biweekly Tuesday night open house called Macwac (milk and cookies/wine and cheese), visitors cycled through a sanitizing station by the front door, and Olsson’s party trick was a roving demonstration of proper handwashing technique, using ultraviolet gel. After that, Ibasho hunkered down. The following week, so did the rest of San Francisco.

Living was simple at first. The government had ordered everyone to stay home, so the housemates stayed home. Macwac made a brief online appearance on experimental software that took the form of a virtual living room, where people could gather in separate corners to chat. (“It was sufficiently weird and depressing that we never did it again,” Dobro says.)

Then the world started to open back up. This was accompanied by the reemergence of a thing called “desire.” In its wake came strain. Olsson describes what happened next as the “everyone-needing-exceptions problem.” The underlying issue was that any one person’s desires affected everyone else in the household. That wasn’t an unfamiliar concept. A similar principle applied to the matter of the ceramics, and they had resolved that by compromising on the number and prominence of the pots. But the prospect of one person’s actions exposing the others to the virus was more harrowing, and the questions were more intimate, more fundamental to their happiness. Questions, primarily, of love. Could Rhys kiss that girl? The committee would hear the matter next week. Polyamorous relationships made for dilemmas: Could Dobro’s partner’s spouse’s partner go on a bike ride with a friend? The answer was yes. What about visits to Mom? Well, they couldn’t veto that, they thought—unless maybe, actually, they could.

A breaking point came during the Black Lives Matter protests in late spring. Could a few of them join? The pod convened an emergency meeting and decided that waving signs from a distance of 30 feet with a mask was fine, but if anyone entered the crowd, they would have to wear a mask around the house for 11 days to reduce the risk to the others. “Some people thought we were discouraging civic participation,” Olsson says. But as she saw it, they were issuing cautious public health guidance to their pod based on the little data they had.

The meetings were growing tiresome, the decisions more contentious. “We were watching group houses fall apart around us,” Dobro says. What was the point of living as a chosen family in your thirties if doing so meant doing less? “At some point, they’re just like, ‘I can’t.’”

It was a minor universe to be concerned with in a moment of global sickness and death. But then again, for most of us, our own small daily choices are all we have. To be free is to pursue your own choices. But freedom also means being responsible to the wider community; it is freeing to know that the people around you are not a threat to your health and well-being, and that you are not a harm to them. As the housemates tried to decide on the correct amount of freedom, they were in agreement that they could not rely on the government to delineate it. After all, in some places, indoor dining opened before playgrounds, and basic guidance about masks and disinfection had been bungled. And governments did not seem interested in questions of love and friendship. Olsson had started tweeting at San Francisco mayor London Breed. Where was the data to justify reopening this and not that? What was the city’s guidance on hugging a friend? If Ibasho was going to survive the pandemic intact, its members would have to figure out a better way to evaluate risk. They couldn’t convene the house parliament every time Rhys wanted to kiss a girl.

In the late 1970s, a Stanford engineering professor named Ronald Howard became preoccupied with the risks of life. Every activity, he wrote in a research brief for the US military in 1979, involves hazards:

• Walking: dog attacks, motor vehicles, falling
• Horseback riding: falling, being kicked, struck by branches
• Staying in bed: fires, burglars, falling airplanes, meteorites, earthquakes

Even the least consequential risks could be quantified with enough data, but often they were so small that they were hard to grasp. So Howard proposed a subunit that he called a micromort: a one-in-a-million chance of death. The advantage of this measure was that it could be used to compare the dangers of apparently dissimilar activities. The risk of a scuba dive, 5 micromorts, could be shown as roughly equivalent to the risk of driving a car from New York City to Cincinnati and back. In this way, Olsson might evaluate the risks of her bicycle commute versus a train ride, or a person might ease the dread of an upcoming surgery by comparing it to something they like to do, like downhill skiing. Howard’s belief was that we willingly invite the possibility of death all the time in order to live our lives more fully. So why not optimize the benefits we get out of this gamble?

Howard’s work was part of a wave of research interested in righting human wrongness. Previous orthodoxy had held that human decisionmaking—whether in the stock market or war—could be described by rational models. So why do humans sometimes make decisions that don’t bring them maximal benefit? Psychologists like Daniel Kahneman proposed that the mind takes shortcuts—heuristics, he called them—which are beset by biases. It’s perfectly human, for example, to fear a plane crash more than a car crash—plane crashes are out of your control and kill many people at once. But a car crash is more likely to kill you, mile for mile. Sometimes these shortcuts lead us straight to perfectly good outcomes. Often they do not. Even giving people clear evidence that they are acting against their own interest doesn’t help them change course. The biases seem to be hardwired in our brains.

Everything You Need to Know About the Coronavirus

Here's all the WIRED coverage in one place, from how to keep your children entertained to how this outbreak is affecting the economy. 

By Eve Sneider

Howard’s answer to such irrationality was “installing a new operating system on your brain,” as he wrote. He had first started thinking about the complexities of decisionmaking in the 1960s, when he was called in from academia to help General Electric ponder installing a new component on its nuclear reactors—a question that involved minute uncertainties as well as dire risks and great expenses. The process took eight months. His method, which he called decision analysis, involved reducing a decision to a series of inputs: probabilities in place of uncertainties, using the best possible data to properly weigh each cost and benefit. Essentially, he wanted to clear away the biases and red herrings and get to what actually mattered to the final outcome. It was an iterative process, designed to consider the range of all possible outcomes and inputs. It didn’t erase the risk of a nuclear meltdown. But if one happened, the company could look back and know that it had made a sound decision.

Ibasho had engaged in a similar process each time the residents sat down to discuss exceptions. But since this was becoming agonizing to do over and over again for each little thing, Olsson wondered if they could find a way to agree on the costs of everything in a more systematic fashion. That way, they could budget their risk until there was a vaccine. It would be a little like counting calories. Like consuming a sliver of chocolate cake, it was easy to rationalize a single risky decision. Soon enough you’ll consume the whole thing. But if you have to record that sliver and put its calories on display, you’ll think about each bite. The decision about how to spend your budget would be yours—not chosen by the committee—but it would require weighing today’s indulgence against any future snacks.

When she proposed this idea, a few of her housemates thought it sounded wonderful. They also happened to be the ones asking for the most exceptions. They had romantic partners living outside the house or they were looking for someone new. Or else, like Olsson, they were numerical enthusiasts. But to Bachar, the housemate Olsson knew from college, it sounded overwhelming. “Thinking about Covid on a daily basis freaked me out,” she says. As the world opened up again, she had chosen to live with minimal exceptions. She wanted life to be settled and safe, not optimized. More data wasn’t going to make her life any easier. It was going to make her crazy. So she and her like-minded fiancé, Nick Breen, asked to be excused from the virus talk. This wasn’t normal house procedure, but the others agreed that they would keep Bachar and her partner looped in. Then, when it was ready, they would all stick to the budget.

Beginning in May, the four members of Ibasho’s new Covid subcommittee began to develop a system for weighing and budgeting viral risk. Olsson called their risk points microcovids, in a tip of the hat to Howard, and one microcovid equaled a one-in-a-million chance of catching the virus. They pulled epidemiology papers from Google Scholar and gathered around the table in the hearth to go through the data. The first step was to impose a top-line risk budget that would anchor all of their calculations. They debated this question at length. Olsson floated the idea of 10,000 microcovids per person per year—the equivalent of a 1 percent chance of catching Covid. But what was the actual cost of 10,000 microcovids? By their estimations, for people their age, a 1 percent chance of getting sick was about as risky as driving, which was something they did without thinking. And besides, they figured, if other people who could stay home kept to a similar budget, the hospitals would not overflow. The virus might even disappear.

It was clear from the start that the evidence they were looking for did not exist. Unlike the risk of dying in a scuba accident, it was not well understood how people exposed to the virus in specific situations actually catch Covid-19. You couldn’t attach a number to the risk of visiting a grocery store or taking the bus. “The data is not designed for people doing math to make their lives better,” Olsson says.

But there was another way to approach the exercise. Any situation could be broken down into two parts: the risk that the people around you were infected and the risk that any infected person would give you the virus. There might not be data on the spread of the virus inside a particular restaurant, or even restaurants generally, but you could attempt to calculate the risk of being in a room with, say, 10 masked workers and 20 unmasked diners for one hour. And then you could tweak the calculation depending on whether there was ventilation or loud talking or the tables were distanced 12 feet apart. Such a model would be a Frankenstein’s monster hodge-podge of estimations. But it would be a place to start comparing one disparate situation to another.

So, first, the risk of the people. One metric that Ibasho could determine with some accuracy was the local prevalence of the virus—a function of the number of cases reported and the rate of positive Covid-19 tests.

The second part was trickier: the likelihood of an infected person spreading the virus to you. At first, like everyone, the residents of Ibasho had freaked out about contaminated surfaces; hence the sanitizing at Macwac. But then experts they trusted began to believe that the virus was spread in the air. As they scoured the rapidly expanding canon of Covid research, they found the factors that made a situation more or less risky came down to mask quality, ventilation, distance from other people, and—a particular surprise to them—the volume at which people spoke, because loud talking meant spewing more virus. The complication was determining how much weight to give each factor.

One of the researchers Olsson followed on Twitter was Jose-Luis Jimenez, an aerosol scientist at the University of Colorado Boulder who had made an attempt at modeling the rate of transmission between people within a closed space for a given duration of time. His calculations took into account humidity and airflow and breathing rates, and he sourced them from dozens of papers on masks and ventilation. But much of the evidence had been gathered in studies of flu and other viruses, not the new coronavirus, and the true rate of spread would depend largely on the specifics of any given space. “We don’t know this disease very well,” Jimenez says. Still, the Ibasho residents raided the cupboard, supplementing his sources with other studies they found.

There were also special situations to consider. They added in the risks of housemates and partners based on contact-tracing studies that estimated how likely the virus spread within homes and at work. If new data on airplane filtration or infection rates from indoor dining became available, they would revisit the evidence and update accordingly. They would never be exact, Olsson knew. Every data point they had was uncertain, and the evidence for everything from masks to ventilation was under intense public litigation. The numbers they came up with were not any an epidemiologist or government entity would endorse. But as they checked their intuitions, the numbers were starting to feel right.

In July, after a few weeks of testing and demos with the more apprehensive housemates, the house met and agreed that the Microcovid system was ready for use. The risk assessment tool was really just a Google spreadsheet. Each person had a tab, and they established protocols. The Microcovid creators had allocated each house member 10,000 points for the year, but they had only 3,000 points to spend. Because they lived in a house of six people plus two quasi-live-in partners, just being at home would cost them each about two-thirds of their points. Ideally, before stepping out the door, they would hop on their laptops and enter estimations for things like the number of people, the quality of the ventilation, and the rate of mask-wearing for whatever activity they planned to do. The calculator would spit out a number of points based on these factors, and they would enter it in the sheet—they could update the numbers later, if reality turned out to be different.

The first thing Olsson noticed as she stepped into a new world of calculated freedom was that certain things mattered more than she had believed. By summer, going to the grocery store had started to feel normal again. But it involved spending time indoors with lots of people, some of whom inevitably wore ill-fitting masks, and it was eating up their points. The solution: communal shopping for the house with P100 masks, the kind you wear when using paint thinner, with the valve covered by a surgical mask. “It’s a strange fashion choice in Berkeley Bowl,” Olsson says, “but it’s good to know the PPE has got your back.”

For Olsson, the point estimation was more like tying her shoes or putting on a jacket before stepping outside. “I don’t find it overwhelming at all,” she says. She could meet up with a friend outdoors, and the combination of fresh air, masks, and distance would whittle the interaction down to just a few points—a blip in the budget. Dobro could take a Lyft to her office downtown and it wasn’t such a big deal, the calculations suggested—provided both she and the driver wore masks and kept the windows open.

Though the budget was set for a year, the housemates apportioned the points weekly, so that one person wouldn’t hoard their points and blow them on a 150-person unmasked indoor wedding. Sometimes the constraint produced challenges: A rare high-risk event, like a flight, might exceed an individual person’s budget. This was OK once in a while, provided the person wore a mask around the house and then got a Covid test.

Some activities were trickier to translate into points. First dates, in particular, would trigger a reversion to what Olsson calls a “one-off person-risk estimate.” The fact-finding missions these estimates required were a little strange and intrusive. The housemates wanted to know how often a new person shopped for groceries, who they lived with. Were they a gym rat? An ER doctor? Bachar found these interrogations uncomfortable. It felt as if she was implying that her friends were behaving badly. But others felt the questions were a reasonable concession to the pandemic. Dobro says that polyamory had prepared her for these awkward conversations around trade-offs. “We’re used to having conversations that are linked to risk,” she says. If you choose to be indoors with someone, the roommates agreed, make it count. Make it a deep conversation. Make it sex.

What if society had a budget for risk? In some ways, it does. This was the initial premise of shutdowns and social distancing and sheltering in place. Our common infection budget was tied to hospital capacity—the number of ICU beds and respirators and medical staff able to respond. For those who could work from home, the task was to contribute as little as possible to the overall sum. This left more points for those who couldn’t. Then, as the first infection curve began to flatten, the foundation of the societal budget seemed to shift. Yes, we still had to worry about public health, but that concern was being stretched by other considerations: business closures, job losses, some ideal of liberty, the desire to eat burritos.

As word spread through San Francisco’s group-house scene about Ibasho’s odd calculator, some people thought that, yes, budgeting seemed a little over the top, and the relentless data entry anxious-making. But people in that community tend to be quantifiers themselves. And for group houses, where each person depended on everyone else for their safety, it also looked like a solution to their own precarity.

Those group-house friends then started to send the tool to their own friends and family and coworkers. Those second-order contacts responded with gratitude. No official source had produced a tool that answered tangled questions with apparently clear answers. And so new questions were finding their way back to Ibasho: How about the gym? What about kids? Josh Oreman, who had done much of the epidemiological research with Olsson, was concerned about others adopting their model. What if Microcovid inadvertently pushed people toward taking more risk, by giving them a budget that was either too small—and thus ignored—or too great? “What if we say the wrong thing?” he recalls asking. There was only so much that an online risk calculator built for a group of healthy, childless, possibly polyamorous people in their twenties and thirties who lived in a San Francisco group house could do for strangers in unknown situations. There were people who had far less control over their lives than they did—people who worked in an ER or shared a bed with a store clerk.

Dobro felt that despite Microcovid’s limitations, people outside their pod would benefit from using it. As a doctor, she had noticed two extreme trends among her patients. Some had become agoraphobic. They were cleaning fastidiously and refusing to leave their houses even for necessities. Others were fed up with constraints and began taking too many risks. One of her patients, a man in his twenties, began coming in frequently for Covid tests. It turned out he had been bouncing between different groups of friends and jetting off on vacations, believing the tests a guarantee of safety. Dobro had started bringing up her own budgeting in these situations. She wanted to show him and anyone else how he could have some of the life he wanted while still being safe.

Olsson could see Dobro’s logic. Plenty of people still did not seem to understand how Covid-19 spread, and a tool like this might help them. She wasn’t thinking of the anti-maskers with beliefs diametrically opposed to her own—she couldn’t touch that situation. But she was thinking of the people still scrubbing their hands raw or living in fear of outdoor masked gatherings. Seeing how various activities compared, or how things like masks helped reduce the point totals, might help them find balance. Experts like Jimenez, who took a look at the tool, agreed. Communicating these beliefs was useful.

Still, Olsson hadn’t intended to step in as a public health authority. It was one thing to pass a spreadsheet around to like-minded friends. They knew how to approach the numbers, and they would happily do battle with her biases and assumptions. But risk communication wasn’t her expertise. There were abundant caveats. She believed in the numbers—or, at least, she believed in the things they had decided were important for their equations. But with more strangers using the tool, she couldn’t help but wonder if perhaps they had oversimplified. What if they had dismissed important routes of transmission, or included a spurious data point that threw everything out of whack?

One of those strangers was Bob Wachter, the chair of internal medicine at UC San Francisco and a frequent public commentator on all matters Covid-19. As a doctor, Wachter was used to probabilistic thinking—working through a range of uncertain pros and cons and arriving at the least worst decision. But in September, he found himself stymied. He was considering a trip to Florida to visit his father, who was 90 and ailing, possibly for the last time. A colleague who happened to know about Ibasho’s budgeting heard about his predicament. Soon Wachter was on Microcovid, inputting the details of his flight.

According to Microcovid, the risk of this flight was 200 microcovids, equivalent to a one in 5,000 chance of infection. The estimate was rough, he knew; for that matter, it was hard to judge what he thought of that level of risk. But he found it grounding, if intangible. “Everyone has to make about 50 risk decisions a day, and they really do need more practical guidance,” he says. “The CDC isn’t offering that.” He wouldn’t act on Microcovid’s points alone, but it was an input into another cost-benefit question: “If this goes badly and I get infected, and if I infect my parents, will I look back and say that I felt like this was a bad decision?” Wachter says. He decided he wouldn’t regret this visit. The decision would be sound, even if the outcome turned out badly. So he flew.

In October, I visited Ibasho. We gathered in the backyard. The space was cramped, but the housemates had recently undertaken projects to make it more homey: sprucing up the plants, redoing the masonry. We sat under a lemon tree drooping with fruit, masked and distanced, and I was the only stranger. They weren’t using many points. Much of the time, freedom was simple and involved just stepping outdoors or putting on a better mask. Living with Covid-19 was just living. The calculator had achieved what Olsson had set out to do. They had returned to harmony, in which everyone could exercise their own independence around a common set of beliefs.

Beyond this Eden, cases of Covid-19 were spiking again. Society was blowing its budget. The housemates had continued modifying Microcovid. The calculator, gussied up with a website and an interface, now also offered a color-coded risk category based on how much an activity contributed, relatively, to a 10,000-microcovid budget. There was an option to calculate risk using a tenth of Ibasho’s budget, for people who felt they were at higher risk, and preset scenarios for things people were asking about, like flights and one-night stands. They had started sorting out what to do about kids. There was always more to do.

A few weeks later, as the winter surge reached San Francisco, I checked back in. Lives at Ibasho were getting more complicated. The housemates felt they knew how to live, and yet the rising infection rate meant their summer habits were putting them over budget in winter. There were new questions too, like whether a new variant believed to spread faster would mess with their calculations, and how to figure in the risks of people around them who had received a vaccine.

Bachar had initially found the points frustrating. When the rate of infection had been lower, it had been both overwhelming to keep track of everything—who else was in the line at Walgreens that day, what kinds of masks they wore—and also a little futile, because she never hit her limit anyway. The tabulations didn’t come naturally to her as they did to Olsson. But now she found the process oddly comforting. She could see how being around friends who did not budget their risk was getting more expensive. It was odd to think of her friends as “expensive.” She did not like the sound of it, and she would have to see them less. Still, the points gave her a structure. It was a way of coping, however imperfectly, with this strange new life she was living. It was, she had calculated, the best she could do.

• 📩 Want the latest on tech, science, and more? Sign up for our newsletters!
• Vaccines are here. We have to talk about side effects
• Gear to safely make it through a pandemic winter
• Test positivity is a bad way to measure Covid’s spread
• The “healthy building” surge will outlast the pandemic
• I tested positive. What does that really mean?
• Read all of our coronavirus coverage here